Episode Transcript
Transcripts are displayed as originally observed. Some content, including advertisements may have changed.
Use Ctrl + F to search
0:00
Oh,
0:17
hey. Stop. It's WAN Show. Gonna
0:19
get some bitches. Get some pussy.
0:22
Sub silent loop. In
0:25
all seriousness though, we've got a lot of great topics
0:28
for you guys and it's gonna be a bit of an unusual
0:30
show. YouTube's updated
0:33
swearing in violence policies have caused
0:35
masked monetization of gaming YouTubers
0:37
videos. Wizards
0:42
of the coast has backpedaled on
0:44
OGL changes after community backlash.
0:46
Pezo moves ahead without them.
0:50
There's been some interesting developments around
0:52
the seventy nine hundred XTX and the problems
0:54
it's been facing courtesy of our friend
0:56
Roman Dairbauer And
0:59
Luke had oral surgery, so
1:01
he can't really talk. Hence,
1:03
the get ups. Let's go
1:05
ahead and roll that intro. Oh,
1:12
no. I'm the one who does that. Alright.
1:35
So why don't we jump
1:37
right into our main topic
1:39
for the day? Prominent gaming
1:41
creators had stoked community backlash
1:43
against YouTube over recent changes
1:46
to its ad friendly content
1:48
guidelines, which has caused many
1:50
of said creators videos to be
1:52
age restricted and or
1:54
demonetized. Hence, the title
1:57
of the video and I
1:59
mean, I guess, live benchmark
2:02
of YouTube's content filtering because
2:04
I went out of my way to attempt
2:07
to break their new filters or
2:09
rather break the rules within
2:11
the first eight seconds of the video. The
2:14
livestream went in went live almost
2:16
instantly on YouTube, which I'm not used to.
2:18
Normally, there's about a five to fifteen
2:20
second delay. I might have missed the window,
2:23
but I I took my best
2:25
shot at it. Now starting in October
2:27
of twenty twenty, the guidelines stated
2:29
that frequent use of strong
2:31
profanity could result in limited or
2:34
no ads. But in April twenty twenty
2:36
one, the guidelines changed to allow
2:38
creators to use moderate profanity like
2:41
shit and bitch within the first
2:43
thirty seconds of a video. In
2:45
November twenty twenty two, YouTube
2:48
issued a major update though to these
2:50
guidelines. Not only could profanity be
2:52
used within the first seven seconds?
2:55
Oh, not only could profanity used within the first
2:57
seven seconds lead to limited ads, but
2:59
all profanity was now treated equally
3:02
except for hell and dam,
3:04
which apparently are no
3:06
longer being treated as profanity anymore.
3:09
Hell, like damn. Damn
3:13
is not even a swear word anymore. We're
3:17
all going to hell. Profinity
3:19
used after the first eight seconds
3:21
though was apparently fine as long as it
3:23
was not excessive. The update
3:25
also specifies that standard
3:28
gameplay where gory injuries are
3:30
present within the first eight seconds
3:32
would lead to limited adds. On
3:35
December twenty fifth, Daniel Quorum
3:38
of RT Game asked YouTube for help
3:40
after one of his videos was set to limited
3:43
ads. And in addition
3:45
to denying his appeal, YouTube subsequently
3:47
flagged over a dozen more of
3:49
his past videos. Condrian believes
3:52
that they were flagged because he escalated
3:54
the issue. Oh,
3:57
you know what? Looks like Riley and I are
4:00
Kind of of the same mind here. Riley says, yeah,
4:02
it does look like that. But as all
4:04
powerful as the algorithm is, it is
4:06
actually not scanning every YouTube video
4:08
all the time. So when attention is brought
4:10
to offending content, it humans
4:14
are more likely bought. Probably
4:16
checks related content. Pro
4:20
z d has apparently tested the policy
4:22
already waiting eighteen seconds before
4:24
swearing and the video still got
4:26
demonetized. Hey, Dan.
4:30
Can we just throw up our sponsors down here?
4:32
To just just thank them for being
4:34
part of making sure that we can run this demonetization
4:37
experiment. Do wanna just throw
4:39
throw them up on the on the bottom for me just one
4:41
after the other. Hey, look at that. We got Notion
4:43
today. We got FreshBooks. We
4:46
got Squarespace. Those are
4:48
some damned good sponsors. Hell
4:51
yeah. This
4:55
morning, YouTube gave a statement
4:57
to the Virgin and we've actually gotten a statement
4:59
up here which I have not had the pleasure of
5:01
reading yet because I was busy getting dressed
5:04
up for the show. In
5:06
recent weeks, we've heard from many creators
5:08
regarding this update, YouTube spokesperson
5:11
Michael Asaman. Oh.
5:14
So he's an Aspenman. I
5:16
thought maybe more of a more of a boobman, but
5:20
an Aspen. Okay. Good. I told the village.
5:22
That feedback is important to us, and we are in
5:24
the process of making some adjustments to this policy
5:27
to address their concerns. We will follow-up shortly
5:30
with our creator community as soon
5:32
as we have more to share.
5:34
Well, that's not really a statement. Is it?
5:38
No, it's not. He agrees. Thank
5:40
you for that. We've
5:44
got a few discussion questions here.
5:46
First of all, what responsibility does YouTube
5:48
have to inform creators of changes to its
5:50
guidelines, both current or upcoming? think
5:53
Honestly speaking, that's the
5:56
biggest problem here from my point
5:58
of view. It's not that the rules
6:00
exist. Right? Like, It's
6:02
been pretty common sense for us for a
6:04
long time that you gotta
6:07
understand. Add safe guidelines
6:09
don't come from YouTube. They
6:12
come from advertiser tolerance
6:15
towards a particular thing. So
6:17
we all have our own colonial,
6:19
you know, moral and taboo
6:22
compasses. Right? Like, for some
6:24
people, full frontal nudity
6:26
is, like, whatever. Like, my understanding as
6:28
a North American where that's like,
6:32
is that in Europe? You can, like,
6:34
have a billboard on the side of the
6:38
fucking road. K? Look
6:40
for benchmarking the thing. Okay? You can
6:42
have a billboard on the road with, like,
6:45
full frontal nudity on it. It's like, yeah, what?
6:47
You know, like Holland or whatever. Over
6:49
here, I mean, you show
6:51
a little bit of crack And that's,
6:54
like, like, not the kind you smoke. Like, the kind
6:56
out, the back of your butt or the front of your chest.
6:58
Right? Like, you're So
7:01
edgy. Right? I mean, that was more true
7:03
twenty years ago. But even
7:06
even today compared to compared to other
7:08
regions of the world, it's it's
7:10
different. Right? So everyone has kind
7:12
of like their own their own ideas, whether
7:15
it's nudity or violence or profanity
7:17
for for what can be considered
7:19
tolerable. By the way, for those of you who
7:21
watching who do not like a profanity riddled
7:24
win show, we'll be back to normal next
7:26
week. Feel free to
7:28
let your kids tune in. We will not
7:30
have the video marked. We are trying to get
7:32
demonetized. That's how you'll be able to know the difference.
7:36
Riley apparently is adding
7:39
notes in real time here. Note
7:41
excessive or excessive
7:43
profanity can Okay.
7:46
This is does he take it? He's not a good
7:48
typing board keyboard guy. Excessive
7:52
profanity can lead to So
7:54
don't overdo it, I think, is what he's is what he's
7:57
trying to say. Anyway, my issue is
7:59
not that YouTube has rules. Rules exist
8:01
for a reason and it's because brands simply
8:03
don't like their brand next
8:05
to content that they feel could reflect poorly
8:08
on them. Whether that is
8:10
Whether that's rational, like whether that has
8:13
any any any data
8:15
backing to it or not, sometimes
8:17
people make emotional purchases be
8:20
it a gorgeous tech
8:22
themed water bottle or be it
8:24
a marketing campaign. Okay? And
8:27
if they see something that they don't like, they
8:29
could experience buyer's remorse and
8:31
not work with that platform again. So I get it,
8:34
The issue from my point of view,
8:36
and I think Luke agrees, is
8:38
that this is just
8:41
happening. It's just it's just
8:43
happening in the night and people
8:45
aren't being informed. And clearly,
8:48
it wasn't common sense to
8:50
some of the people who use a lot of profanity
8:52
and expect advertisements to show up on their videos
8:55
just fine. And not only
8:57
was it you know, not common sense or whatever,
8:59
but the community has ultimately figured out
9:01
where the line was. People were
9:04
figuring out how to stay within it.
9:06
And then the line moved. That
9:10
is not okay. I
9:12
mean, Yeah. I think this is Luke's cursor
9:15
now because it typed a lot faster. You
9:18
know, Luke says it's kinda Luke figures,
9:20
you know, it's like when Elon randomly started
9:22
banning everyone that was doxxing
9:24
people on Twitter when what really happened
9:27
was he changed the definition of
9:29
doxxing. So if YouTube is banning
9:31
people or not banning, but demonetizing people
9:34
for using too much profanity.
9:36
When even just the words too much,
9:39
or excessive are
9:42
fucking meaningless. Right?
9:45
How how are how are people supposed to
9:47
deal with that? And you gotta You gotta understand.
9:49
Right? YouTube is not some rinky
9:51
dink little platform that, you
9:53
know, people upload to for fun anymore.
9:55
This is people's livelihoods that are at stake.
9:58
Unless they have really great sponsors like,
10:00
can you can you get those sponsors across the bottom
10:02
for me again? Just wanna just wanna shout out our
10:04
sponsors who hopefully, thanks,
10:06
Notion. Are comfortable
10:09
being next to this kind of, thanks, fresh
10:11
books. Unsavory
10:14
unsavory, you know, language. Right?
10:17
Squarespace. Squarespace too. Shout out Squarespace.
10:19
If if they haven't ditched me yet, they're
10:21
not gonna ditch me now. Now
10:25
our second discussion question here is
10:27
what response ability do creators have
10:29
to be aware of potential changes like this
10:32
and play it. Safe cough cough LMG.
10:35
And that's true. I
10:37
mean, you know, Luke
10:39
has not been in favor at
10:42
times of our family friendly
10:44
approach to I
10:46
mean, really a lot of things. Like,
10:48
we get edgy sometimes. I
10:50
think that the how to hide your pornography
10:53
video was among the edgiest things
10:55
that we've ever done. But
10:59
we mostly get edgy in, like, kind of
11:01
a silly toilet humor innuendo
11:04
kind of way. We don't go straight
11:06
for, you know, really, really
11:08
vulgar language or you know,
11:10
really graphic really
11:13
graphic language or or imagery.
11:15
And a big part of that has been
11:18
out of staying advertiser friendly. It was a
11:20
business decision. It was not because I
11:22
have any particular aversion to using
11:24
cuss words. As
11:27
for what responsibility creators have,
11:29
like, that's tough. Right? Because creators
11:32
are supposed to be diverse
11:35
Right? That's one of the benefits
11:38
of the platform. And there's
11:40
a certain I don't think that
11:43
for our type of content, And for
11:45
our message, there's any particular
11:47
need to swear. And personally,
11:50
I consider a bleeped
11:53
to be funnier than a not bleeped fuck.
11:55
I think that I just I
11:57
find it to be comedically superior.
12:02
Okay? So that's a a huge
12:04
part of the reason that we do things the way that
12:06
we do, but also because
12:09
we want to remain advertiser friendly.
12:12
However, I think that there
12:14
are creators whose personalities or
12:18
whose message could
12:20
be aided by speaking in a
12:22
way that is true to themselves, and
12:24
that might involve some cussing. And
12:26
so for me to say or for me
12:28
to pass some kind of judgment, right, on
12:31
the responsibility that creators have
12:33
to be aware of potential changes and
12:35
play it safe, I think it's
12:37
it's one of those things where my
12:41
opinion is irrelevant. And so is
12:43
his. Right? Like,
12:45
we're we're coming at this from a space, from
12:47
a personality that doesn't need to communicate
12:50
in that way, whereas Some
12:53
people do. And I I just I
12:55
want YouTube to be a place where
12:57
all kinds of creation can thrive. That
12:59
is as long as you're not spreading disinformation,
13:02
spreading hate, you know, engaging in
13:05
in illegal activities if you're recruiting
13:07
for ISIS or some shit, then as far as
13:09
I'm concerned, you can get the fuck off the
13:11
platform. Mhmm. Right? But man,
13:16
a couple a couple f bombs. I
13:19
don't know, it's tough. I think that I
13:21
think that if advertisers haven't already
13:24
gotten used to the fact that YouTube
13:26
is like kind of a
13:28
zoo. There's a little bit of everything.
13:31
You know? Sometimes it's the
13:33
kind of zoo where, you know, They had
13:35
more than two of that kind of animal. You know what I'm
13:37
saying? You know, they got they got a lot
13:39
going on. It's
13:41
cluster. Okay? That's what I'm trying
13:43
to say. Our
13:47
our next discussion question that Luke and I
13:49
will be discussing is what would
13:52
need to happen to enable a a YouTube
13:54
level platform that doesn't rely on potentially
13:56
British advertisers for revenue. Wait,
13:58
that's float plane. Except it isn't right,
14:00
because we know we know
14:02
it's not a sustainable model. Right? Like,
14:05
once you've got a certain amount of momentum,
14:07
like, LNG, which
14:10
I shouldn't call it that. Lyness Media Group
14:12
has over thirty thousand paying
14:14
subscribers on Flowplane that are anywhere between
14:17
we have, like, o g grandfathered in
14:19
members at a three dollar, three
14:21
US dollar a month tier. That is
14:23
no longer available, hasn't been for years. In
14:25
fact, I think the percentage of three
14:28
dollar tier users is really
14:30
dwindling these days. We have
14:32
a five dollar tier and then we have a ten dollar
14:34
tier that gets you four k which is higher bit
14:36
rate than YouTube's four k. And most
14:38
importantly, not necessarily image quality,
14:41
but our audio is uncompressed,
14:43
which is one of the big reasons that someone
14:45
like Dankpots is on float plane for
14:47
his garbage time streams. But
14:50
it's super cool because he
14:53
he drums. Right? And YouTube's
14:55
audio compression twitch for that matter
14:57
makes it so that, like, it
15:00
doesn't have the dynamism that
15:02
flow plane streaming can get him. Right?
15:04
And so it's great for
15:06
people like us who have that
15:08
kind of critical mass that
15:11
I mean, you guys do the math. Right? Thirty
15:13
thousand people paying anywhere
15:15
from realistically five to ten dollars
15:17
a month is not an insignificant amount
15:20
of revenue. It justifies a development
15:22
team. It justifies us
15:24
investing in a a group of
15:26
people that makes exclusive content just
15:28
for that platform. Right? Or for
15:30
someone like for someone like tank pods. Right?
15:33
Where he's got a specific use case
15:35
for it and a dedicated audience
15:37
that is really into that kind of content
15:39
that will pay for it monthly. But I've had
15:42
lot of people point out like, hey, liners.
15:44
I couldn't help noticing I have to pay for every
15:46
creator. That adds up too fast.
15:48
I can't afford it. I get
15:51
it. Hundred percent. You
15:53
are right. A
15:55
platform like vessel where you can pay
15:58
one time and get all you can't
16:00
access to every creator on the platform, that's
16:03
way better. But you
16:06
aren't YouTube. Right? So
16:08
let's say you have a hundred thousand users
16:10
or something like that. Okay? Even a pretty
16:12
good number of users. You got hundred thousand
16:14
users. Okay? They're paying five dollars
16:17
a month. That's five hundred thousand dollars
16:19
a month. That's a ton of money. For it
16:21
all, you can eat platform, okay, wait second.
16:23
Hold on. How many creators do I have on here?
16:25
If I've even got fifty
16:28
creators. Right? All of a sudden, now
16:30
we're talking We've got five hundred thousand.
16:33
Oh, boy. Divide
16:36
by five. Okay. That's twenty grand a month.
16:38
That's that's a ton of money. Man,
16:41
yeah, how can we get to how can we get a hundred thousand
16:43
subscribers on FlowPoint? Man, not okay. Anyway,
16:47
the point is, you would have more than fifty creators.
16:49
If if we if we built a platform that
16:52
was like an all you can eat, that was like a a
16:54
smorgasbord model, what would happen
16:56
is the share for those individual creators?
16:59
If we had fifty that were the size of LTT,
17:01
for example, the share of that revenue
17:04
for each of those creators that's left by the time
17:06
you deal with payment processing fees, which are significant.
17:10
Man, we haven't talked about the exact numbers in a
17:12
long time, so you're gonna have to just kind of
17:14
let me know. If I recall correctly,
17:16
it was fifteen percent flat per transaction
17:18
plus a percentage, give
17:20
or take? Is it twenty five flat? don't know. The
17:22
point is it is a significant
17:25
amount of money when you're looking at a five dollar
17:27
transaction. The lower the transaction, the
17:30
bigger the proportion it is because you have to deal
17:32
with that flat fee no matter what. So
17:34
essentially, you're just in rich in credit card
17:36
companies, which the world definitely needs more
17:38
of, obviously. So
17:42
as soon as you start to divide it, between
17:44
more and more and more and more and more creators, they
17:47
lose their incentive to actually
17:49
even bother to upload on the platform
17:51
because You know what? I'm gonna get
17:53
my like sixty I'm MKBHD. Right?
17:56
What? I'm gonna get my sixty dollar check-in
17:58
the mail or even my six hundred dollar
18:00
check-in the mail. You know, I don't wanna you
18:03
know, obviously, I don't know exactly
18:05
what his numbers are, but I work
18:07
in the same vertical. I have some idea.
18:09
I can tell you that a six hundred dollar
18:11
check to Marquette is pretty much a rounding
18:13
error. Right? And with his
18:15
costs, you you guys gotta understand.
18:18
I'm not saying that because he's like so out of touch
18:20
and just has so much money. I'm just saying
18:22
that he's running a business over there. He's got a
18:24
dozen plus employees nice
18:26
space, good equipment. He's trying
18:28
to do a great job of what he's doing. Okay? It costs
18:30
money to do that. And that
18:33
amount of money Ain't
18:35
gonna make a huge difference. So
18:40
oh, okay. Hey. A. J. Is posted in a
18:42
Flippling chat. I was apparently way off it
18:45
is actually thirty cents per
18:47
transaction, plus a percentage. So
18:49
on those three dollar transactions in the
18:51
early days, We were giving up
18:53
ten percent right off the bat
18:55
to a flat transaction fee. Now, I
18:58
haven't been looking at the chat lately, but I'm sure
19:00
some people are asking about Nebula. Nebula had
19:02
a very very different model
19:04
where essentially they have a partnership with
19:07
Curiosity Stream that makes it
19:09
so that you are basically
19:12
getting Nebula for free if you
19:14
sign up for Curiosity Stream who
19:16
sponsors a lot of Nebula creators
19:19
effectively paying
19:21
creators to promote getting
19:26
their nebula subscription for
19:28
free. So it's like, is
19:31
anybody actually paying for Nebula? Probably.
19:34
But are more
19:36
people likely paying for Curiosity Stream and
19:38
getting Nebula for free? Probably.
19:42
So it's a really interesting model over there.
19:44
And then the creators on Nebula
19:47
are stakeholders in Nebula. And
19:49
my understanding is that that is dependent
19:51
on the size of the creator and overall
19:53
contribution to the momentum of the platform. And
19:56
so what that means is that even
19:58
if your check from nebulized four dollars
20:01
a month or whatever because whatever
20:04
depends how many people are watching, I guess. They
20:06
don't have any external viewership stats or
20:08
anything like that, which usually means that it's
20:10
not heavily used. So
20:12
it doesn't matter though if you get a sixty dollar
20:14
check. Because essentially you're getting
20:17
paid to promote it by one
20:19
of the the associated media
20:21
companies. And then you are
20:23
maybe in the event of
20:26
of an exit, in the event that Nebula
20:28
gets sold, you are getting some
20:30
some cut of that of that sales check.
20:32
So it's a very very different model.
20:35
It's maybe someday something
20:37
will happen, and also there's this very
20:40
convenient synergy here with
20:42
Curiosity Stream that I'm sure
20:44
makes sense for Curiosity Stream on some level.
20:48
Very, very different dynamic from what Flowplane
20:50
is, which is the money from
20:52
the subscribers you can drive today,
20:55
right now, and is sustainable. We
20:58
don't need we don't need any external
21:00
backing nothing like that. We are
21:03
we are profitable now. It's amazing
21:05
how many people I still see talking about the
21:07
epic failure that was flow plain. Guys,
21:11
the numbers are right on the site.
21:14
They're right on the site. Flowplane
21:16
is doing through LTT alone.
21:20
At least. What is it?
21:22
Calculator. Okay? So
21:24
five times thirty
21:26
thousand. K? Here
21:29
we go. Five
21:33
times thirty thousand times twelve. Flowplane
21:36
is doing at least one point eight million
21:38
dollars US a year in revenue
21:40
and that's just on LTT.
21:43
We don't have a ton of creators on the platform.
21:45
I admit The model doesn't make a ton
21:47
of sense for everyone, but what it is is
21:50
it is sustainable. And
21:52
Luke and I, I think, have been proven
21:55
Maybe not right, but not
21:57
wrong multiple times since
22:00
Flowplane's inception. I mean, look at
22:02
all the controversy there was around Patreon
22:04
and Vimeo. When Patreon was
22:06
effectively having people just
22:08
put up unlisted videos on YouTube for
22:10
the longest time. And then there was the integration
22:13
with Vimeo where Patreon
22:15
creators were able to use video to host their videos.
22:17
And then all of a sudden, video
22:19
kind of went, Bandwidth
22:22
and storage are expensive. If you guys
22:24
don't start paying for ProTear Vimeo
22:26
hosting, all your videos are
22:28
gonna go away. See you later. And then now
22:30
I think Patreon is self hosting video,
22:32
but I haven't actually looked at the quality
22:35
of the service since then, so I can't really
22:37
speak to if it's if it's any good or anything
22:39
like that. Yeah.
22:43
Level one tax has another eight hundred followers.
22:45
I I know I know garbage time streams,
22:48
like, hundreds. I know that forgotten
22:51
weapons has hundreds, if not.
23:00
Oh. Forgotten weapons does not
23:02
disclose how many subscribers are on Flowplane.
23:05
Forgotten weapons, good guy. And
23:08
Good channel. Doing good. Why don't I
23:10
just say that? The good news is I actually didn't
23:12
know if that number was accurate, so
23:16
you can assume it was wrong. Nate.
23:22
Let's move on to our next topic. Wizards
23:25
of the coast has backpedaled on
23:28
OGL changes after community backlash,
23:31
sort of. We've got a big
23:33
update for you guys, but first for those of
23:36
you who didn't tune into the show last week.
23:38
Let's give you a little bit let's get
23:40
you up to speed a little bit. Wizrs on the coast
23:42
owned by Hasbro. Runs, Dungeons and
23:44
Dragons, basically,
23:47
had a license agreement in
23:49
place with whether it was third
23:51
party content creators or
23:54
whether it was, like,
23:56
community run companies or even
23:58
actually companies with significant
24:00
revenues based on their Dungeons and
24:02
Dragons IP. They had a license
24:04
in place that pretty
24:06
much allowed you to create derivative
24:09
works without paying any
24:11
kind of royalties and without worrying
24:13
about any sort of legal action. The license
24:16
was perpetual, and it is a
24:18
huge part of why the
24:20
DND community has grown so vibrantly
24:22
over the last couple of decades. I
24:24
mean, we're talking everything from third
24:26
party add ons to I
24:29
mean, you know what? A perfect example
24:32
of someone that could be affected by this
24:34
is that local creator, Filthy Lot.
24:36
Where they're doing those d and d
24:38
live reenactments, where they
24:41
where they play the game and they do these, like, super
24:43
high production value reenact of their
24:45
tabletop games. I could
24:47
see someone like that. Even
24:49
if they ultimately aren't infected and
24:52
they don't need to pay twenty five
24:54
percent of their revenues to whizzards
24:56
to the coast, I could see them being worried
24:58
about the legal ramifications of the new
25:01
general license. And
25:03
that's one of the things that changed all of a sudden,
25:06
almost overnight with almost
25:08
no warning. Any companies that
25:10
were profiting off of Dungeons and Dragons
25:14
were going to have their license agreement
25:16
changed. So this included
25:18
a twenty five percent revenue garnish
25:22
that the fact I think you had to submit basically
25:24
anything to do with Dungeon and The Dragons to
25:26
them for if not approval, at
25:29
least for them to to look at
25:31
and be aware of. There were
25:33
a handful of other particularly awful
25:35
things and it was basically gonna happen like in a
25:37
couple of weeks. Now the thing leaked and
25:39
the community got understandably outraged
25:42
because This is wizards of the coast basically
25:44
coming in and going, you
25:47
know what? The D
25:49
and D community, the players, are
25:52
woefully under monetized. And
25:56
that's not even me just
25:58
like trying to sound like an evil executive.
26:01
The CEO of wizards of the coast actually
26:04
said that, which is terrible.
26:07
Describes the player base as under
26:09
monetized. And it's
26:11
like, even
26:15
if you're right. Okay?
26:18
Even if your NBA ass
26:20
knows what you're talking about. Come
26:25
on, man. Come
26:28
on. You can't.
26:31
That's like it. It's
26:33
Oh, man. How do I how do I it's
26:35
just it's just disrespectful. Like,
26:37
it's like if I it's like if I refer to
26:39
you guys as as walking wallets
26:42
instead of our community.
26:45
Right? Even if I saw you
26:47
that way, even if I fucking
26:49
did. Right? You walk
26:51
in wall at pieces of shit Even
26:54
if I felt that way, you
26:56
don't say the quiet part out loud, it's
26:59
not tactful. It's disrespectful. So
27:07
anyway, let's get you guys up to speed.
27:09
D and D Beyond staff, oh,
27:12
Oh, okay. Have released a statement about
27:14
the recent OGL changes. They
27:16
have three goals in mind.
27:18
Oh, they had three goals in mind.
27:20
Okay. To prevent the use of
27:22
DND content from being included in hateful
27:25
and discriminatory products. Okay. To
27:29
address those attempting to use DND
27:32
in web three slash blockchain slash
27:34
NFT material, I'm already calling
27:37
bullshit on this. Because the
27:39
document Okay. So the original the original
27:41
license agreement was like, how
27:43
many pages? I don't remember, but the
27:45
new one's like nine hundred pages and
27:47
has extensive documentation about,
27:50
like, web three voicemail blockchain, NFT
27:53
crap. And as far as I can
27:55
tell, This is not about preventing others
27:57
from doing it. This is about making sure that
27:59
they can do it and others can't. Anywho.
28:02
Hold on. And reason number
28:04
three, to ensure that OGL is
28:07
for the content creators slash home brewer
28:09
slash aspiring designers, etcetera,
28:12
as opposed, I guess, to anyone who's
28:14
looking to commercially profit off of it. But that's
28:16
the thing. You gotta understand. When
28:19
you create an open ecosystem, where
28:21
people can profit, you create a financial
28:23
incentive for creativity. That's
28:27
the whole point. That's
28:29
what you benefited from. And
28:32
so now you're basically going, this
28:35
is going great. Let's get
28:37
more benefit. Alright.
28:44
There were apparently two principles driving
28:46
these goals. Be good stewards of
28:48
the game and that the OGL exists for the
28:50
benefit of the fans. Nothing,
28:52
this is a quote. Nothing about these principles
28:54
has wavered for a second. All
28:57
of the royalty language was apparently meant
28:59
to apply to large corporations attempting
29:02
to use OGL content and wasn't meant to
29:04
impact the vast majority of the community.
29:07
Man, that's problem when the lawyers get involved.
29:09
Even if that's true, even if that was the
29:11
intent. The problem is that
29:14
the OGL Seems
29:17
to give them a lot of flexibility to crack
29:19
down on pretty much anything that they didn't
29:21
agree with. The next OGL
29:23
will contain provisions Let's going
29:25
back to back to this. To protect and cultivate
29:28
an inclusive environment, education charity
29:30
livestreams cause play, etcetera, will remain unaffected.
29:33
Okay? But, man, there's
29:35
some really there's some really fine lines
29:38
here. Right? Like, okay.
29:41
What is cause play? If
29:44
I dress up as some
29:46
DND IP character
29:49
or class of character, What? I
29:51
mean, can can you can you
29:53
copyright a fucking elf at
29:55
this point? Right? I
29:58
don't think so. So okay. So I
30:00
so I so I I'm cosplaying some
30:02
d and d something. And it's it's clearly let's
30:04
say it's clearly infringing. Right? What
30:10
if I cause play and I'm in
30:12
the video? What
30:14
if that video is monetized? What
30:16
if I didn't profit, but the maker
30:18
of the video profited? What if
30:20
I what if I'm not in a video? What if
30:23
I'm just at a con? I'm at
30:25
I'm at like some I'm at tax or I'm
30:27
at comic con or I'm I'm at something.
30:29
I'm at some kind of I'm at some kind of gaming
30:31
convention. What if
30:34
What if, you know, I mean, imagine me.
30:36
Okay? I wanna be like I wanna
30:38
be a sexy elf. Okay? You know,
30:40
I've got a bikini. As part of my as
30:42
part of my cause play. Right? Exactly.
30:45
Right? So someone slips a
30:47
someone slips a one to my g strength.
30:50
Okay. Have I profited now? Is this commercial?
31:00
There's clearly some fine lines here is what I'm
31:02
trying to say. Now they
31:04
also say Content released under
31:06
one point o a will remain unaffected.
31:09
No royalty structure included. No
31:11
life back provision under the new OGL,
31:14
you will own the content you create. Okay?
31:17
They've also said that the OGL update will
31:19
not release today, but it is coming. What we
31:21
saw was a draft sent out to content creators.
31:24
Nobody drafts a
31:26
document like Okay. Yes.
31:29
You do draft a document like that. But
31:32
a draft of a document like that
31:34
is unlikely to be so
31:36
far off the mark. If
31:38
it was all about cultivating
31:42
and inclusivity as
31:44
opposed to where's my money?
31:47
Where's my money bitch? Alright.
31:57
There was a very cheeky and salty line
32:00
that was applied to
32:02
this announcement that is being made fun of online.
32:05
Second, you're gonna hear people say that they
32:07
won and we lost because making your voices
32:09
heard forced us to change our plans. Those
32:12
people will only be half right. They
32:14
won, and so did we? The
32:19
fuck is wrong with you. Who
32:25
talks like that? Who
32:27
sends a letter like that to
32:30
their customers? That's
32:32
like that's like them releasing a
32:34
fucking t shirt? That's
32:36
like Trust
32:46
me, bro. The new OGL
32:48
will be epic. Okay.
32:53
But seriously seriously
32:57
though, in the case of the warranty
32:59
drama, our track record
33:02
and our intent was
33:04
clear. We have always
33:06
had a basic policy for
33:09
customer care. Make it right.
33:12
We got a little slow. Okay? We have
33:14
quadrupled the size of our customer care team
33:18
We are actually I was told that
33:20
by today, we would be down to twenty four hours,
33:22
and we haven't actually told
33:25
anyone yet. But some of the temps, I
33:27
believe, will actually be retained because
33:29
we're gonna be building new systems whenever there's
33:31
downtime when it comes to actually responding to
33:33
tickets. One of things we're gonna start doing
33:35
is combing through reviews on the site for feedback.
33:38
We're gonna be compiling that for product development
33:40
team. So, essentially, like,
33:43
customer customer, you
33:45
know, ticket answer department
33:47
is going to turn into more of, like, customer
33:50
experience department and is going to
33:52
be designed to continue to make
33:54
everything better from response
33:56
times to to
33:59
future products to even even, like,
34:01
midstream reviews or midstream reviews.
34:03
Midstream improvements on existing products, you can
34:05
expect to see all of that from us in the future.
34:08
Oh, apparently, it's called customer satisfaction.
34:10
Is that what the department's called? Okay.
34:13
Whatever. Whatever. Whatever we end
34:15
up calling it, it's it's it's not just
34:17
about answering tickets. And The
34:22
t shirt wasn't miscalculation. I'm
34:26
not gonna lie. I still think it's funny.
34:29
But I live in my head. Where
34:31
I knew my intent all along. And the thing
34:33
that I missed in all of that, the thing I totally
34:35
missed in all of it was that a
34:37
lot of you don't. Right?
34:40
You don't. You don't know what my intention
34:42
was. And I've told you guys, you know,
34:44
don't don't trust or at
34:47
very most, trust but verify. Right?
34:49
And so, for me, it
34:51
was very obvious. I'm looking at it going well, I've
34:53
got this super long track record of taking
34:55
care of our community. Also, even
34:57
if I didn't, I would be insane.
35:00
I'd be insane to screw anyone
35:03
over. If we actually did,
35:05
right, Like, if we
35:07
actually did, look at how much it
35:10
blew up when we didn't
35:12
actually screw anyone over. If
35:14
we actually just started denying
35:17
warranty claims on products on
35:19
our store, it would explode.
35:22
It would explode the entire techosphere. I'm
35:25
not a sucker for that kind of punishment. So
35:28
from my point of view, It was, hey, I've got
35:30
this track record and b isn't this obvious.
35:34
But from your point of view, we feel you were
35:37
a first time viewer of that show. Maybe
35:39
you're a first time viewer of this show. It's usually
35:41
not like this. Maybe you were first
35:43
time viewer of that show. Right? Maybe you'd
35:45
never ordered from the store before. Maybe you actually
35:47
had no context whatsoever for what
35:49
our policies were. And maybe all
35:52
you saw was quite
35:56
wealthy, Influencer.
35:59
That's a reference to last week's show as well.
36:03
Basically, you know, saying, you know, if
36:05
you want a warranty, well, you know,
36:08
go fuck yourself basically. Right? And
36:10
that was never my intention, but
36:13
I think we've made good on it. And I
36:15
think that in the same way, even
36:18
though they issued this spectacularly
36:20
stupid statement, they
36:23
still have an opportunity to not
36:25
screw this up. They
36:28
still can. They haven't actually
36:31
released the new OGL. Until
36:34
they do, they can still
36:36
reverse course. Well,
36:41
yeah. Okay. So Luke's Luke's
36:44
got some thoughts. They have already
36:46
permanently alienated large parts of the
36:48
community. This is and all in the name of what
36:50
inclusivity. Good
36:52
fucking job. And
36:57
they have also, if the idea
36:59
here, was to not allow potential
37:02
competitors to benefit from
37:04
their content. Man. This
37:06
was the biggest shot in the arm that
37:08
they could have possibly given to their
37:10
competitors. Just this
37:12
morning, Pieso began solidifying
37:15
their own OGL. It will apparently be
37:17
system agnostic, perpetual, and
37:19
irrevocable an open
37:22
RPG creative license, ORC,
37:24
they are looking for a nonprofit organization
37:26
with a history of open source values to own
37:28
the license, this would be similar to the Linux
37:31
Foundation and are
37:33
essentially coming in and saying,
37:36
even if these guys reverse course,
37:39
What they've shown is that
37:41
you're a walking wallet and you can't
37:43
trust them. I
37:46
mean, yes. Technically, the
37:49
original OGL. Technically,
37:52
the original OGL while it was perpetual
37:55
was not you're revocable. They're
37:58
not probably outside of
38:01
what is legal. But they are outside
38:04
of what is in my opinion morally
38:06
acceptable. When I see a perpetual
38:09
license, That doesn't mean to
38:11
me, hey, this
38:14
is going really great. I'm creating
38:16
all of this content. In
38:18
this amazing ecosystem that
38:21
is going to be be mine as part
38:23
of this ecosystem forever. And I can
38:25
I can monetize and I can use to hire great
38:27
creatives to make more content. This is gonna
38:29
be amazing, and we're gonna do this forever.
38:32
And oh, by the way, one day,
38:34
you can say, actually, no. And
38:36
then all of that will change, and now you're gleaning
38:38
twenty five percent of my revenue. No.
38:42
When I hear perpetual, what
38:44
I understand is not perpetual, but
38:48
like then, and then, like, less perpetual
38:50
later. I think that I think, you
38:52
know, from a non lawyer standpoint, I
38:54
think that means perpetual. Right?
38:57
Like, we've got a document called the
38:59
perpetual land show document. Okay?
39:02
That doesn't mean that every that this land
39:04
show document is perpetual until
39:06
we decide to make AAAA
39:09
femoral wan show document.
39:13
Okay? It's perpetual.
39:15
This is the only land show document. We
39:17
used to have a separate document every week, but
39:19
we now have a perpetual land show document
39:21
because that way we don't have to make sure shared to all
39:23
the stupid accounts on all the different computers and everything
39:26
every week. Yeah.
39:28
TeamViewer. Jam fuck nine
39:30
nine ninety one nineteen in full yeah,
39:33
exactly. When I buy a perpetual license
39:35
to your fucking software, that means
39:37
it's mine forever, not it's perpetually mine
39:39
until you decide it isn't mine anymore. No.
39:42
No. No. Forget it. Alright.
39:45
So our discussion questions here. Who should
39:47
take charge of ORC? I mean, honestly, I don't
39:49
have enough experience in it to say. How
39:52
truthful? Oh, this is great. Some of our other discussion
39:54
questions are really great though. Luke,
39:58
if you think it was truthful. The wizards
40:00
of the coast's intent was to stop bigotry
40:03
and n f t's. Say nothing.
40:06
Or wait. No. Don't say say something.
40:09
Crap. I set that whole thing up and then I
40:11
got it completely wrong. Fuck. Fitch.
40:21
Okay. And not
40:23
capitalizing at or as in Orcs
40:25
versus humans not ORC. I have no
40:27
idea what you're talking now. Anyway, I'm talking about
40:29
now. Luke says,
40:31
I think you can even say things. Like
40:33
your fan slash user base is woefully
40:36
under monetized. But yes.
40:38
Yes. This is what was
40:40
trying to explain the situation to Yvonne over
40:43
dinner last night, where she was like
40:47
sorry. Like, what's what's going on? Like,
40:49
she never ever played dungeons and dragons
40:51
like I mean, she she has
40:54
kind of a knack for legal documents and,
40:56
like, license agreements and stuff these days because
40:59
in the absence of a company lawyer.
41:01
She did she ran a lot of legal documents
41:03
for us and had to learn to
41:05
understand them for the most part. Because, like,
41:07
even when we had a lawyer in
41:10
many of the early days, we couldn't just
41:12
afford to have that lawyer like sit and look at
41:14
everything So we would look at it, figure
41:16
out what our questions are, what sections they needed
41:18
to look at, and just utilize them as little
41:20
as possible. Lawyers be expensive. Alright?
41:24
So she hadn't even seen the document, but I'm, like, kinda
41:26
trying to explain what's going on. And I'm,
41:28
like, yeah, they're basically what
41:30
they're doing is they're taking this existing
41:33
vibrant ecosystem and they're going, how much
41:35
can I squeeze it? How much
41:37
value can I extract instead of
41:39
asking themselves? How much value
41:41
can I provide? And I don't
41:43
think anyone would have complained if
41:46
they had come up with some amazing
41:49
new offering for their customers. And
41:51
said, hey, but this
41:54
costs money. They
41:57
well, oh, they might have complained. Sure.
41:59
But it wouldn't have been like this.
42:01
It didn't have to be like this. That
42:06
will increase monetization Yeah.
42:08
So the example that Luke was thinking
42:10
was, you know, how do you how do you
42:12
increase your monetization by offering something
42:14
new? So instead of offering just
42:17
T shirts on LTT store. We
42:19
developed a backpack. We developed
42:21
a screwdriver. The point is
42:23
not to just increase the price of
42:25
t shirts, which if you guys know, we have
42:27
never actually done. Our t shirts
42:29
are still nineteen ninety nine US,
42:32
which is twenty dollars. I know.
42:34
There's still twenty US dollars just
42:36
like they always have been regardless of whether
42:38
they're blank or printed. By the way, I have a
42:40
bit of an update for you guys. We are apparently
42:42
working with three or four printing
42:44
shops locally, trying to get some samples
42:46
in. The thing that drives me most crazy
42:48
about our previous printer is that the quality
42:51
was great. We actually loved
42:53
working with them. Oh,
43:00
I keep saying ORC when
43:02
referencing the OpenRTB creative
43:04
license, but it's apparently pork. That
43:07
makes sense. Hilarious. I'm
43:10
also not technically wrong. It's all caps.
43:14
Technically right. The best kind of right.
43:17
Actually, the quote is technically correct. But
43:21
let's get let's get pedantic today, shall
43:23
we? Why
43:26
don't we why don't we get to a couple of
43:28
merch messages actually? If you guys wanna
43:30
send a message into the show, the way to do
43:32
it today is through a new feature.
43:35
That we developed to add additional
43:37
value. We saw a feature that
43:39
was broken and sucked merch messages,
43:42
which don't show up in our dashboard properly,
43:44
and we went, hey, that's
43:46
stupid. Look at
43:48
this. Here's here's two merch. I'm sorry,
43:50
did I say merch messages? Here's two
43:52
super chats. K? Here's two superchats.
43:55
Oh, good. One of them is here.
43:57
What happened to the other ones? I don't
43:59
fucking know. Nobody fucking
44:01
knows. It's gone. So we
44:04
saw something was broken. Also, by the way, why
44:06
are you just giving Google money to build features
44:08
that don't work properly? That
44:10
has been broken for two years. Unbelievable,
44:15
unacceptable, merch messages,
44:18
men. How will it take?
44:20
I'd like I'm I'm pretty sure the first iteration
44:22
of merch messages was ready in less than a week.
44:25
It was amazing. Okay? And
44:27
what's best is that instead of getting
44:29
Google, a big cut of the money that you
44:31
send to us, you actually get to give
44:33
it to people like, well, used to
44:35
be our t shirt printer. But anyway, You get to give
44:37
it to people like A Mega Pro
44:39
and pH molds and the companies that we
44:42
work with on this crude driver. You get to give
44:44
it to our our creator warehouse
44:46
team who gets paid out of that money. You
44:48
get to actually get something in the
44:50
mail. Right? You don't just give that
44:52
money away. You give it to people who are
44:54
actually making real things that you might
44:56
actually wanna have like a really nice insulated
44:58
water bottle. It's better. So
45:01
so merch messages were developed so that you guys can
45:03
interact with the show in a way that is better for
45:06
everyone. Does it cost money? Yeah.
45:09
But we have received like zero
45:11
pushback on merch messages. Some people
45:13
complained we were doing too many of them, but then
45:15
for every person who complained we were doing too many of
45:18
them, there was at least three people saying
45:20
merch messages are my favorite because it's basically
45:22
q and a time. So our response to that
45:24
was we do a couple of them early on in the
45:26
show, so I can talk about how to spend how to
45:28
spend a merch message. Friday and
45:30
slip. Love it. How to send a merch message,
45:32
you go on LTT store dot com, you pick up something,
45:34
doesn't that? If you're not, You know, if we don't have anything
45:37
fresh and new that interests you, you can always just pick
45:39
up a gift card. Okay? You can spend it later.
45:41
So you go to all to d store dot com and the checkout, you'll
45:43
see a place to leave a merch message. It goes to
45:45
our producer, Dan. Stop producer,
45:47
Dan. He will
45:49
either respond to you. He'll just show your message
45:51
if you'd like, hey mom or whatever happy birthday
45:53
to a friend or whatever, you can just have it show up
45:55
down there, or sometimes he'll curate them
45:57
so that Luke and I can check them out later.
46:00
And if we don't
46:02
get to it, well, hey, at least to get your
46:04
order in the mail. Heck yeah. Any
46:07
who? Let's do a couple of them now and then Well,
46:09
right. So I was gonna say our compromise was we do
46:11
a couple of them early in the show. We talk about how to send
46:14
them, and then we address the rest
46:16
of them sort of more towards the end of the show when it
46:18
evolves into absolute chaos. Dan
46:20
hit me with couple merch messages. Let's go. Okay.
46:22
got one here from Vincent.
46:25
A sinus out of Navidea, Intel
46:27
and AMD, who's overall the biggest liar,
46:29
and out of those, which lie was the most
46:32
personally offensive. Why
46:35
did you ask this on this funny land show?
46:37
Corporations is a liar sometimes.
46:39
Right? Wow.
46:42
You specifically called out Intel
46:45
AMD and Nvidia. Now,
46:48
here's the thing. Companies
46:51
are made up of departments, which
46:53
are made up well, okay. No. Hold on.
46:56
Companies are made up of
46:58
business units. Which are made up of
47:00
departments, which are made up of
47:02
teams, which are made up of people.
47:05
So if I was to
47:07
if I was to zero in on
47:10
an egregious lie, I
47:13
mean, I could pick I could pick any one of them.
47:15
Man, I just
47:18
about pulled an all nighter. K?
47:20
When AMD's bulldozer processors
47:22
came out, because our
47:25
AMD rep at the time told
47:27
me they were freaking awesome.
47:29
Sorry, excuse me. Fucking awesome.
47:31
He was like, man, these CPUs
47:34
slap bitches harder than Dana White. Oh.
47:43
Sorry. I'm trying to get into character. Okay?
47:46
You're gonna kill the pair of us. Anyway,
47:53
he he tells me, Get
48:02
it together, Luke. Anyway,
48:05
he tells me that the CPU is
48:07
amazing. At the time, right,
48:10
I am I am ghost writing for hardware
48:12
Connex, which was owned by NCI X.
48:15
And I'm working on like our launch
48:17
day, you know, coverage. I'm trying to overclock
48:19
it or whatever. And I'm sitting here.
48:21
I'm benchmarking late into
48:23
the night going. This thing's a piece of shit. But
48:26
I was told it's great, but it's the middle of
48:28
the night right the night before the embargo
48:30
because I only got it like that day. And
48:34
I'm sitting, you're going, everyone else is
48:36
gonna make me look like an idiot. I can't publish
48:39
these numbers. This thing is dog slow.
48:42
And then it comes out, and it's like,
48:44
it sucks. And I'm sitting here going, well, that
48:46
would have been great to know. But,
48:51
I mean, Okay. was about
48:53
to say that's not AMD's fault.
48:55
AMD does have a culture of
48:58
overhyping crap products that
49:00
that suck. It's
49:02
usually closer to, you
49:04
know, white lies, I guess, or
49:07
don't know if they're white lies. It's usually closer
49:09
to bending the truth. I
49:11
get is it AMD
49:14
lies? Okay. Intel lies, lots.
49:18
NVIDIA. NVIDIA is just
49:20
a bunch of insidious fucks. Like,
49:23
let's face it. They they have a culture of being
49:25
insidious fucks. And
49:28
don't kid yourself, you know, just because
49:30
just because Intel dresses nicer and is
49:32
more respectful to you, doesn't mean I do leisure
49:34
businesses. Right? They exist to
49:36
separate you from your money. However,
49:40
there are people there. I wanna
49:42
I wanna go back to the other side of this coin.
49:45
Whether it's Nvidia or whether it's Intel
49:47
or whether it's AMD. I have
49:49
met some of the most genuine, passionate
49:52
enthusiastic people that I've ever encountered
49:54
in the industry. Because all
49:56
of these companies, insidious fucks
49:59
that they are, are profitable, which
50:01
allows them to spend lots of money
50:04
to attract super passionate, super
50:06
intelligent people to work on their products.
50:09
So what I'm trying to say is
50:12
that corporations is a liar sometimes.
50:14
Absolutely. But corporations
50:16
is actually also full of enthusiastic, amazing,
50:19
highly intelligent, honest people.
50:23
Right? Like, I remember running into one
50:25
of the people that advocated for
50:27
us for skull trail at Intel.
50:30
And he was just like, yeah, that was like my passion
50:32
project. I was like, you're awesome.
50:35
That was so cool. It was basically Intel
50:38
server platform in gamers clothing
50:40
and overclockable at the time. Right? It
50:42
was awesome. Bleeding edge
50:44
made no commercial sense, but they did
50:47
it. Right? You know, and it's it's
50:49
the same thing for companies like Nvidia, man.
50:51
Man does Nvidia ever have great engineers?
50:54
Like, these are people that are just so smart
50:56
that you sit down and talk to them for, like, fifty
50:58
Like, Holy shit. I
51:00
could have gone to school for a year and I
51:02
wouldn't have learned half as much as I did
51:04
just now. Right? And so when
51:07
I say these companies are just like full
51:09
of liars and and and deceivers.
51:11
It's true. But these companies are
51:13
also full of honest, amazing,
51:16
passionate people who would do it for
51:19
zero dollars. So what they're making
51:21
money? That's great. They should. But they would
51:23
have done it anyway, which is amazing. You gotta
51:25
love that passion. Right? You
51:29
know what? Let's have a spicy win show.
51:32
It's not one of those companies, but the
51:35
closest that I would say that
51:37
I've been to being lied
51:40
to outright by a company
51:44
I would say this is fair to say because it
51:46
was directly from the CEO and
51:49
it was a direct deception
51:52
something that they would have had intimate knowledge
51:54
of. This was around a key product
51:57
launch. It wasn't from, you know, a
51:59
low level sales rep or marketing rep
52:01
or even a high level sales rep or marketing
52:03
rep. This was directly from the CEO, and
52:05
I was told from a razor, okay,
52:09
that their gaming switches back
52:11
when they first released razor switches
52:14
were not re bodged
52:17
Kaewa cherry knock off switches.
52:19
I was told that they were razor
52:21
switches inside
52:23
and out engineered by razor.
52:26
And that was at
52:28
best, at
52:31
best, a misrepresentation, at
52:34
the very best.
52:38
Because they
52:41
were kaiwa, cherry clones. Moving
52:45
the actuation point point
52:48
two of a millimeter is not engineering
52:50
a switch. K? The
52:52
Romer g is engineering
52:55
a fucking switch. And you can say what
52:57
you want about the Romer g switch. K?
53:00
But that switch was engineered by
53:03
Almirall with input from
53:05
Logitech. The razor
53:07
gaming switch was not engineered
53:10
by razor. It involved
53:12
engineering. You'd be amazed
53:14
how hard it is to build a screwdriver. And if
53:16
it's hard to build a screwdriver, I promise
53:18
you. It's hard to build a keyboard switch.
53:21
It's a complex mechanism It's
53:23
small. The cost needs to
53:25
needs to be low. It must be low.
53:28
It can't be dollar or a switch. Like, you
53:30
you you can't I mean, you can sell three hundred dollar
53:32
keyboards. But most people will not buy
53:34
a three hundred dollar keyboard. There's a limit.
53:36
Okay? And it's gotta be reliable
53:38
within that limit. The millions of times,
53:40
you have to be able to press it. I get it. There was engineering
53:43
involved, but you did not
53:45
engineer a switch. A
53:47
razor gaming switch at least back
53:49
then. Right? Okay. So this is like seven
53:52
eight years ago. Right? At least back then,
53:54
it was a cherry clone. That's
53:57
it. I love that this is the yeah.
53:59
This is great. D BOSS twenty two eighty nine float
54:01
plane says, linus is gonna wants to be
54:03
spicy today and Luke doesn't have any power to stop
54:05
him this time. Yeah.
54:09
He yeah. He could. Okay. He could pull them, mate.
54:11
Let me get together fuck out of here. So
54:15
there, I think that's the I think
54:17
that's the closest I've ever been to
54:19
just like or I think that's the
54:21
most egregiously. I have ever been
54:24
misled by a company. And
54:27
to be clear, I'm not counting I'm
54:29
not counting anything that was unintentional. You
54:32
know, like, even if you if you
54:34
go back and look like, the principal technology
54:36
scandal at Intel, I had a theory
54:38
back back during that time. That
54:40
no one would ever go on the record and talk to
54:42
me about. But I'll
54:44
say that my theory is basically there
54:47
there were ignorant executives and
54:49
their egos involved. There
54:51
was skilled people who
54:54
didn't have the time or chance to
54:56
review things. Involved, basically,
55:00
bottom line, don't attribute to malice, what
55:02
you can attribute to incompetence or whatever
55:04
whatever the quote was. I suspect that
55:07
that was a clown show comedy of errors,
55:09
not AAA willful attempt
55:12
to deceive the enthusiasts community. We're the only
55:14
people watching those stupid presentations anyway,
55:16
and who we're obviously gonna catch that.
55:21
You know, that's that
55:25
I don't I don't classify the
55:27
same way. I just I just don't Whereas
55:30
when you tell me, yeah, we built this thing from
55:32
scratch and you, like, actually didn't. That's
55:35
that's different. Maybe
55:40
it was another one. K.
55:43
Yeah. Sorry about this. Lay there. I've got one here
55:45
from Well,
55:48
we'll do this one from oh,
55:51
jeez. I don't know. How about from
55:53
Reed? No. I wanna save LTX
55:55
for later. Cody. Love the
55:57
products, guys. I was wondering if you have any IPP
56:00
IP V4 address space
56:02
and use BGP for anything. And if
56:04
you guys would ever consider making a video about
56:06
how all that works, keep up the good work. Love
56:08
you guys. And that's a that's a no.
56:12
I mean, yeah,
56:14
we we definitely have a small block of
56:16
IPV4 addresses, but
56:18
like we're not we're we're not like
56:21
hoarding them or anything. We,
56:23
you know yeah. We
56:25
just we we just have them They're
56:27
useful. Oh, we have a
56:29
a fair number? Oh, okay. Cool.
56:32
Make sure they seldom. Alright.
56:34
Cool. Sure. Luke wants to hold
56:37
on to them. One more? Yeah.
56:39
Sure. Give me one more. I got one here from
56:41
Eric. Jay and Bob, I've been feeling
56:43
a ton of burnout at work lately, and
56:45
I'm considering changing career paths. My question
56:48
is, do you have any advice regarding
56:50
when to go to the higher ups and complain about
56:52
issues and when to know they aren't listening
56:54
and you just must move on. Oh,
56:56
man, that's a good question. Well,
56:59
one thing that I will say as
57:02
an employer is I
57:05
really man. And this is gonna
57:07
This kind of ties back into a conversation we
57:09
were having recently about how I
57:11
really respect people who who wanna
57:14
better themselves, who want feedback. And
57:16
I find the legal framework around employment
57:18
here in BC very frustrating because it
57:20
prevents me without opening myself up to
57:22
liability. From providing
57:24
feedback to people who are being
57:27
let go. You know, hey, here's how you could probably
57:29
make it go better next time. I I just can't do
57:31
it. And
57:33
so, you know, as
57:36
much as it's, like, sucks that
57:39
it's, like, a double standard, I guess, is
57:42
what I would say as an employer is
57:44
please talk to me. Right?
57:47
Give me a chance. If there's something that's
57:49
making you unhappy, The absolute
57:51
last thing in the world that
57:54
I want is for you to just
57:56
quietly sit there
57:58
and faster on your unhappiness and
58:01
be miserable and resentful when,
58:04
like, Honestly, maybe it's
58:06
something that we could have just fixed.
58:08
And maybe it might maybe it wouldn't happen
58:10
overnight. Like, for example, we
58:13
didn't have an employee retirement savings
58:16
plan until very recently. That's
58:18
something that we added. I think we've got pretty
58:20
kick ass plan now. It's like
58:23
very competitive, but
58:26
we didn't have one before. Why?
58:28
Well, because there's a lot of administrative overhead,
58:31
it's a significant cost. And
58:33
I mean, well, yeah, that's why. Basically, there's
58:36
and as we've grown and as we've become
58:39
more profitable and as we've added more
58:41
people and as we're looking for ways to
58:43
continue to grow and become a
58:46
better place to work, like my whole thing. Right? Has
58:48
been since day one. And I've always I've always done the
58:50
voice, so I'll do the voice again. I want to be a real company.
58:52
Right? Is how do we
58:55
how do we keep improving? What
58:57
if what if you were upset
58:59
and resentful and the thing that
59:01
you were so upset about was
59:04
something that, like, We
59:06
were two months from announcing and you
59:08
up and quit and that would have been
59:10
it and you would have been happy. Or
59:12
what if it's something that, you know, I had been
59:14
sitting there going like, man.
59:17
I really wanna do this, but like, I don't even
59:19
know how much appetite there is for it, but all
59:22
of a sudden if I if I just start talking about
59:24
it, then people might get the wrong idea.
59:26
Maybe only, you know, maybe maybe
59:29
ninety percent of people would just rather reinvest
59:31
that into hiring more people so that they're not working
59:33
as hard. Like, these are conversations that especially
59:35
in the early days we had a lot. Like, I'd
59:37
sit down with, you know, Luke
59:40
and Brandon and Taren and OG's
59:42
like that and I'd say, hey, look. Coming
59:44
into this year, I
59:47
could pay you more. However,
59:51
You'd have to keep working at this pace. Or
59:55
I could hire more people. I'm
59:58
kinda leaning towards door number two
1:00:00
here. What do you guys think? And,
1:00:02
you know, I I think got it right for
1:00:05
the most part. And, you
1:00:07
know, obviously, that doesn't mean that we were just
1:00:09
freezing pay at at
1:00:11
any stage and all of that. Like, we were
1:00:13
we were we were constantly trying to
1:00:15
to to do better as far as that one as well. It's
1:00:17
a balancing act. Right? But we were
1:00:19
also trying to so we're trying to make the place
1:00:21
a better place to work while also adding
1:00:23
more people, building more infrastructure. The
1:00:25
point is that I wanna know. I
1:00:27
need to know. Because if I don't
1:00:29
know what people want, I'm just guessing.
1:00:33
What's up? Oh,
1:00:42
oh, yeah. Yeah. Okay. You
1:00:44
know, I think Luke has some strong feelings
1:00:46
about this. If you really
1:00:48
need and want to fix, don't
1:00:51
just bring the problem. Find
1:00:53
a reasonable solution and suggest it.
1:00:57
Maybe even make them think it was their
1:00:59
idea. By
1:01:02
bringing them to that conclusion, I'm sure he's done
1:01:04
it to me. Yeah.
1:01:08
I mean, it's it's communication
1:01:11
101. Whether you're whether
1:01:13
you're in the supervisory or
1:01:15
the reporting to how
1:01:18
do I describe these
1:01:20
rules? Whether whether you're higher on the
1:01:22
on the pyramid chart, the
1:01:24
the company What what what what
1:01:26
is it called? Pyramid. Pyramid. Is that what
1:01:28
it's called? What's the company org chart? Whether
1:01:31
you're higher on the org chart or lower on the org
1:01:33
chart, it is such an important
1:01:35
communication skill to
1:01:37
bring people ideas in a way that is non
1:01:39
confrontational. If I can bring
1:01:41
you an idea and make
1:01:44
you feel like you can relate
1:01:46
with my struggle and that
1:01:48
my solution sounds super reasonable
1:01:51
to you, then the chances of implementing
1:01:53
it are much much higher. With
1:01:55
all of that said, the second part of your
1:01:57
question is, hey, when do I go
1:02:00
this is a lost cause and just
1:02:02
bail. There's a solid chance
1:02:05
that that's gonna be it. One of
1:02:07
one of our one of our rock ours. Okay?
1:02:09
Kyle from Creative Warehouse Engineering. Love
1:02:12
Kyle. He's great. You
1:02:14
know, a big problem for him at
1:02:17
a previous position. I'm not gonna
1:02:19
say it was the one right before this. I'm not gonna say
1:02:21
if it was two before this. We're not
1:02:23
gonna name any names here. But at
1:02:25
a previous position, he had
1:02:27
expressed some concern about
1:02:32
basically some costs
1:02:34
that were associated with his remaining
1:02:37
employed there. It hit him remaining employed
1:02:39
there. Okay? The situation
1:02:41
was going to become untenable for
1:02:43
him. He wasn't even asking
1:02:45
for a raise. He wasn't just
1:02:47
like, give me more money. He
1:02:50
just he just had there were some
1:02:52
costs that were associated with him
1:02:54
working there, that he needed
1:02:56
some solution to. And
1:02:58
in effect, they basically said, deal
1:03:00
with it. And he did. And
1:03:03
he came here. Right?
1:03:05
Well, eventually. So
1:03:09
That could happen. Some people just
1:03:11
They don't care about you. They
1:03:14
they see you as, III
1:03:16
hate the word. I hate the term HR.
1:03:19
They see you as human resources. Effectively,
1:03:23
warm bodies that can be mined for
1:03:26
productivity. And
1:03:28
and that's just the way that's just the way it's gonna be.
1:03:30
And you just have to, you know, karma is a bitch. Right?
1:03:32
Like, you just have to hope that at some point, that's
1:03:35
gonna bite them in the ass. No one's gonna wanna work
1:03:37
for them. And honestly, I think we're seeing a lot
1:03:39
of that today. Labor shortage? There's
1:03:41
no fucking labor shortage. How
1:03:43
many resumes you get? Thousand resumes or
1:03:45
whatever for some positions? Yeah. lot a
1:03:47
lot. No labor shortage.
1:03:50
You gotta pay and benefit shortage. That's
1:03:53
your problem. Figure it out.
1:03:55
And like the thing that drives me most crazy
1:03:57
is like McDonald's food doesn't fucking
1:03:59
cost more in places where they pay living wage?
1:04:02
It's still cheap. Why? Because
1:04:04
it is. You're just taking
1:04:07
more of it. So just
1:04:09
pay more of it. Anyhow.
1:04:15
Alright. Let's move on to We should get some
1:04:17
sponsors. We should get some to go the way. We gotta
1:04:19
pay for gotta pay for all this somehow.
1:04:23
The show is brought to you by our wonderful
1:04:25
sponsors During whose reads
1:04:28
here, I am not going to use any profanity
1:04:30
because that's just not in my nature.
1:04:33
Starting with, help me out, Dan. Thanks
1:04:36
to MIT Mobile for sponsoring today's show. It's
1:04:38
twenty twenty three, which means it's time to stop paying
1:04:40
insane amounts money every month for your
1:04:42
phone bill. Switching to Metmobile will help you do exactly
1:04:45
that. As the first company to sell premium wireless
1:04:47
service online only, Metmobile lets you order
1:04:49
from home and save a ton. Phone plan
1:04:51
started at just fifteen bucks a month. This
1:04:53
means you can finally buy that thing you've always wanted with
1:04:55
how much money you'll be saving on your phone bill.
1:04:58
All phone plans come with unlimited talk and text plus high
1:05:00
speed data delivered on the nation's largest five
1:05:02
g network, and you can even use your own
1:05:04
phone with any mint mobile plan and switch easily
1:05:06
in minutes using eSIM. To get
1:05:08
your new wireless plan for just fifteen bucks a month
1:05:11
and get the plan shipped to your door for free,
1:05:13
go to mint bubble dot com slash wancho.
1:05:15
That's mint mobile dot com
1:05:17
slash wan show. Cut your phone
1:05:19
bill to just fifteen bucks a month at mint
1:05:22
mobile dot com slash wan show.
1:05:25
Next topic. Man,
1:05:27
I'm kind of feeling the Windows eight
1:05:29
point one end of support. What do you think?
1:05:32
Yeah. Let's do it. Windows eight point
1:05:34
one reached end of support on Tuesday,
1:05:36
January tenth. This sucks. I
1:05:38
actually still have an active Windows eight
1:05:41
point one. Right now. It's
1:05:43
not like normal Windows eight point
1:05:45
one. It's Windows nine. That
1:05:47
that sort of stripped down eight point one
1:05:49
embedded or whatever it was called
1:05:53
that we made a video about. Man, I looked
1:05:55
at that thing. I was, man, it's
1:05:57
fucking snappy. It was amazing.
1:06:00
And so I immediately set it up
1:06:02
on a VM that to this day I still
1:06:04
use. So I'm gonna have
1:06:07
man, I'm gonna have to like update
1:06:09
that thing. That sucks.
1:06:11
Because like that's my sketchy VM that I've
1:06:13
just used for like, oh, this is like a weird
1:06:15
program. Okay. Let's let's
1:06:19
disconnect it from rest of the network and
1:06:21
see what it does. Like
1:06:24
and it's so because it's so bare bones, it's
1:06:26
really easy to tell if anything is there that's not
1:06:29
supposed to be there and stuff like, oh man, I just
1:06:31
I love that VM. It's
1:06:33
it's been through a lot with me anyway.
1:06:36
Technical support, software updates,
1:06:38
and security fixes will no longer be
1:06:40
provided. You're not gonna say anything. And
1:06:42
Microsoft is recommending that customers move
1:06:44
to a more current version of Windows. They
1:06:47
are not offering a free upgrade path
1:06:49
from Windows eight point one to Windows ten or eleven
1:06:52
And unlike what happened with Windows seven, Microsoft
1:06:54
will not be offering an extended security
1:06:56
update program for Windows eight point one, probably
1:06:58
because nobody would even want it. The
1:07:01
ESU program for number of older products actually
1:07:03
also came to an end on Tuesday, including Windows
1:07:05
seven professional and enterprise versions and Windows
1:07:08
server two thousand 8R2. Fun
1:07:11
fact. I was actually using Windows Vista
1:07:13
earlier this week. And
1:07:15
Windows eight point one is not gonna be
1:07:18
as bad as that. But I can tell
1:07:20
you using an outdated version of Windows,
1:07:23
it gets pretty rough after a while.
1:07:25
So it was good up until I
1:07:27
think it was about two years ago. We got
1:07:29
our hands on the Dell XPS M twenty
1:07:32
ten. I think it's called, it's
1:07:34
a wild laptop. I can't
1:07:36
imagine they made more than like hundreds
1:07:39
of these if even that. Like,
1:07:41
I can't even imagine who would buy this thing,
1:07:43
but we got our hands on it and that thing
1:07:45
shipped with Windows Vista. And
1:07:48
apparently, it was pretty good up until just a couple
1:07:50
of years ago when support
1:07:53
for when Chromium stopped
1:07:55
supporting Vista. So effectively, you
1:07:58
couldn't do anything in browser. I
1:08:00
think valve pulled support for Vista
1:08:02
and Steam around that time as
1:08:04
well. Don't quote me on that. III could have the timing
1:08:06
a a little bit off. But, man,
1:08:09
you try to do anything on
1:08:11
Vista today. And it's bad.
1:08:14
Even ignoring the security issues, let
1:08:16
me tell you. I
1:08:18
tried I tried to load up I tried
1:08:20
to load up rotten tomatoes dot com.
1:08:22
K? So that I could check reviews for
1:08:24
for a blue ray to determine if I was willing
1:08:27
to risk losing it in the
1:08:29
in the pop up motorized
1:08:32
optical drive on the top of this laptop, guys,
1:08:34
do not miss that video. I
1:08:36
was like, do I care about this blu ray?
1:08:39
And then I could manually
1:08:42
calculate the percentage of thumbs up and thumbs
1:08:44
down reviews because I could
1:08:46
I could see how many there were but the
1:08:48
actual images weren't weren't loading.
1:08:52
It was pretty rough. Our
1:08:55
discussion question here is
1:08:57
where would you rank Windows eight slash
1:08:59
eight point one on a best to worst
1:09:01
list of Windows versions? This is
1:09:03
one of those things that's kinda like ninety
1:09:06
eight, ninety 8SE and
1:09:08
Emmy where the lines
1:09:11
are a little blurry. You can't
1:09:13
ask me where would you rank Windows eight slash
1:09:16
eight point one? Because Windows eight and
1:09:18
eight point one, Let's
1:09:20
just say Microsoft learned a lot
1:09:23
from Windows eight. And when Windows
1:09:25
eight point one was far less of
1:09:27
a piece of shit. I
1:09:30
would rank Windows eight. Oh, man.
1:09:32
See, I liked Vista. So
1:09:34
let's go Matt, you know what? That'd be kind of
1:09:36
a fun video. Like, ranking
1:09:39
ranking windows. And, like, just
1:09:41
like Windows tier list. Yeah.
1:09:44
Oh, that could be a fun float plane exclusive. Doing
1:09:47
like a hey hey, Dan. Can
1:09:49
you can you ping the social team and
1:09:51
actually CC James and let them fight it out
1:09:53
over whether that's an video or whether it's
1:09:55
a float plane exclusive just like kind of casual
1:09:57
social
1:09:58
video. But we should do we should do like
1:10:00
a Windows tier list. Nice
1:10:02
to Windows tier list. I wasn't listening at all.
1:10:04
Yeah. Windows tier list. So we take all the
1:10:06
additions of Windows, and we have people we
1:10:08
have people rank them best to worst. Even
1:10:12
good pre n t windows,
1:10:15
like DOS based windows ninety
1:10:17
eight SE, was not very good.
1:10:20
It crashed a lot. It had a lot of compatibility
1:10:22
issues in spite of the fact that it was running on top
1:10:24
of DOS. DOS programs just would
1:10:26
not work a lot of the time. I
1:10:29
think it's hard to call Windows eight legitimately
1:10:32
worse than anything before
1:10:34
the NT colonel, but
1:10:36
I would say that Windows eight was
1:10:39
probably the worst of the post
1:10:41
NT colonel operating systems for me.
1:10:45
Like, eleven has its issues,
1:10:47
but they're not stability issues. They're
1:10:50
not just like utterly making it
1:10:52
utterly unusable issues
1:10:54
with the keyboard and mouse. Right? You
1:10:57
know, and and Windows
1:10:59
eight, as I talked about before, with
1:11:01
the the Windows nine experience, could
1:11:06
get wildly better. Like,
1:11:08
I would I mean, I've told you
1:11:10
guys, eight point if if I if
1:11:12
I could have had the latest versions of
1:11:15
DirectX supported. There
1:11:17
is no reason why I couldn't
1:11:19
have just continued running eight point one
1:11:21
embedded with modifications. We've kept running
1:11:23
Windows nine until today. Mhmm. There is
1:11:25
no reason for me to not do that. It is so
1:11:28
usable. Start menu search works fine.
1:11:30
It's much easier to get at your network configuration
1:11:33
panel. Oh, you don't have that
1:11:35
bullshit, terrible control panel
1:11:38
that's just like sits on top of
1:11:40
the regular control panel and makes everything
1:11:42
more difficult. Actually, I think you
1:11:44
did have a little bit of that. But it wasn't as bad.
1:11:47
The whole regular control panel is still
1:11:49
there, damn it. Yeah.
1:11:51
No. I I liked eight point one with
1:11:53
the right tweaks. But then I liked
1:11:55
Vista. So what the fuck do I know?
1:11:58
I'm fucking idiots. I
1:12:02
was running Vista on modern hardware. Mhmm.
1:12:05
I didn't try to plug in my old
1:12:07
ninety eight sierra printer. That's
1:12:09
a big difference. Right? Windows
1:12:13
Vista worked great on a
1:12:15
brand new computer playing current gen
1:12:17
games. Because it was what everyone
1:12:19
was validating with. Yeah.
1:12:22
Okay. Luke,
1:12:24
on the other hand, worked at Geek Squad in
1:12:26
the Windows Vista days, and it
1:12:29
was a bloodbath of low spec laptops.
1:12:32
Yeah. That's I think that's very fair
1:12:34
to say. If you had the specs
1:12:36
for it, it rang great. If you were but
1:12:39
if you were running like thirty two bit VISTA
1:12:41
on some piece of crap, laptop,
1:12:44
it was it was a pretty bad
1:12:46
time. Pretty bad time.
1:12:54
Alright. Why don't we do let's do two more
1:12:56
merch messages and then we'll move into another topic.
1:12:58
Hit me Dan. Okay. I gotta find them. Okay.
1:13:02
But that's your whole job over there. He's finding merch
1:13:04
messages. Yeah. But I mean, he also monitors
1:13:07
the audio. A scroll down to me. He helps you
1:13:09
guys listen to the echo. Go. Go.
1:13:12
Go. Go. Go. Go. Go. Alright.
1:13:15
Let's see. I
1:13:19
think we got that one done too. You
1:13:22
got this. Do you need to do a quick topic while
1:13:24
you look through it? I got Matthijs. Loving
1:13:27
the super edgy episode, laughing at
1:13:29
the whole show. Question for linus, do you take any
1:13:31
actions to protect the long term health of
1:13:33
your voice with all the hosting of episodes
1:13:35
and live shows. Excited for LTX and
1:13:38
LDD store not com. No.
1:13:41
I can't say that I do, but one of
1:13:43
the things that I do do --
1:13:48
Nice. -- one of the things I do
1:13:50
is I just use
1:13:52
my natural talking voice. I'm
1:13:55
not I'm not putting on a voice for you guys.
1:13:57
I I'm louder and that's
1:13:59
something I've tried to work on. But
1:14:02
every time yeah. Every time
1:14:04
I'm quieter. Okay.
1:14:07
Why don't I try? I will simply
1:14:09
speak, I say I can't do it. I immediately
1:14:12
start ramping up. Part
1:14:14
of it is that when I'm on camera,
1:14:17
I'm usually talking about something that I can get
1:14:19
kind of fired up about. Otherwise,
1:14:21
why are we making a video about it? Like, I was,
1:14:23
oh, man. I was in script review
1:14:25
with Oh,
1:14:28
crap. What's that? Oh, yeah. Right.
1:14:31
I was in script review with Tanner today. And,
1:14:34
is anyone outside of my office
1:14:37
probably thought I was laying into
1:14:39
him? Because I was sitting there. I was
1:14:41
sitting and going, hell the fuck
1:14:43
Is this even still a problem in twenty
1:14:45
twenty three? How's this
1:14:47
even fucking possible? Right.
1:14:50
You know, but I was
1:14:53
I was actually, you know,
1:14:55
Tanner's more of a subdued kind
1:14:57
of guy. He wouldn't say it like that. But
1:14:59
it's not like he didn't agree. What I was talking
1:15:01
about was the challenge of
1:15:04
sharing files from one device
1:15:06
to another. You know, Apple's kind of got it
1:15:08
solved, but it doesn't count as solving it when you
1:15:10
only solve it for people that are
1:15:12
made of money and can afford an entire
1:15:14
ecosystem of your products. That's fuck you.
1:15:16
Right? Like, that's not a solution. That's
1:15:18
not a real solution. What
1:15:21
I want is a real solution. If I
1:15:23
need to beam you a file from my phone
1:15:25
to your laptop, That shit
1:15:27
should be simple. It
1:15:29
is twenty fucking three.
1:15:33
Right? So anyway, The point is that even
1:15:35
when I'm not on camera, I can I can
1:15:37
get kind of I can get kind
1:15:39
of passionate about things, but That's
1:15:42
just my that's just like my voice.
1:15:44
And so when I'm on camera, that happens a lot
1:15:47
and I mean, drink
1:15:49
water. You
1:15:51
know? LTT store dot com,
1:15:54
bitches. I
1:15:56
should have waited to eat. Right? Almost
1:16:00
got him earlier with the slap. Do
1:16:06
do you want another one? Or if we Yeah. Give me
1:16:08
one more. Okay.
1:16:11
This is from Daniel, not me.
1:16:14
Getting some birthday merch Line
1:16:16
us, what is a product that you will probably
1:16:19
never be able to make, but would like to
1:16:21
make? Oh, man. I mean, there's there's
1:16:23
a lot of things I'd I'd love to be able to
1:16:25
make Oh.
1:16:30
Oh, yeah. Yeah. Sox. I
1:16:32
mean, we'll get it. We're
1:16:34
gonna we're gonna nail the socks. Don't you
1:16:36
worry? Sandals.
1:16:39
Sandals may never happen. Just the
1:16:41
the mold costs for all the different sizes that
1:16:43
you gotta do. I'd love to do better sandal.
1:16:45
Like, man, the See?
1:16:48
Get all passionate again. Okay? The
1:16:50
failure point on every pair of sandals
1:16:52
I've ever owned is the same. So
1:16:54
why don't you just reinforce it? You
1:16:56
fucks. That's your
1:16:59
whole job is to make sandals. So
1:17:01
what? You don't know how they come apart? Like,
1:17:04
how is that even possible? No one man.
1:17:08
Oh, man. Okay. Oh,
1:17:10
man. Okay.
1:17:13
I was trying I'm trying to navigate to
1:17:16
like a local rec center on my
1:17:18
way home. Okay? And I'm using
1:17:20
my voice because I'm a responsible person
1:17:22
I'm operating a motor vehicle. So
1:17:24
I press my button on my steering wheel. Hold my button
1:17:26
on my steering wheel. I go, okay, navigate
1:17:29
to whatever the name of it is. Now
1:17:31
I happen to have an entry in my address
1:17:33
book. That is some
1:17:35
other place of business that starts
1:17:37
with that same region. Okay?
1:17:40
So I'm not gonna say exactly what it is. I don't
1:17:42
need to disclose any of this. But, you know,
1:17:44
let's say say for example, I was trying
1:17:46
to go to Alder
1:17:49
Grove Recreation Center, and
1:17:51
I also happened to have an entry for
1:17:54
Alder Grove Dental Clinic.
1:17:57
Three fucking times in
1:17:59
a row. Mhmm. It utterly ignored
1:18:02
recreation center and
1:18:05
navigated me to
1:18:07
the dental clinic. Three
1:18:09
times three times in a
1:18:11
row. With all the
1:18:13
AI shit that Google
1:18:15
does. In twenty twenty
1:18:18
three, how is it fucking
1:18:20
possible that when I say
1:18:22
call Yvonne Ho.
1:18:26
It says, I'm sorry, I don't
1:18:28
have number for Yvonne Hoam. Who?
1:18:33
Who? How
1:18:38
is it possible that when I
1:18:40
say call Hoffman Wong,
1:18:43
it says, I'm sorry, I
1:18:45
don't have an entry for Hoffman. Sometimes
1:18:49
it spells with two fucking ends.
1:18:53
You know, I fixed it. I took the second
1:18:55
end off the end of his name. How's
1:18:57
that even possible? If
1:19:00
I say, call Jake Ivy. Okay?
1:19:04
You check. How many fucking
1:19:06
jakes are in my address book? Jake
1:19:10
Tivity, close enough. If
1:19:13
I say call Jake Tivy, it works. K?
1:19:15
It's the same with James. Right?
1:19:17
Call James Stripe. Calling James
1:19:19
Street. No. No. No. No. No. I said it this
1:19:21
way, you repeat after me. How
1:19:23
is it even possible? Okay?
1:19:26
Because I know they have it. I know
1:19:28
they have the technology to take
1:19:30
whatever the phonetic version
1:19:33
of that would be, cash
1:19:35
it and then give you and
1:19:37
then return a probability match.
1:19:40
Probability match. Okay?
1:19:42
Not a perfect match. Fine. Give me the
1:19:44
eighty five percent one that's in my local
1:19:46
storage. Sorry.
1:19:54
What were you talking about? I
1:19:57
don't remember don't remember what the question was.
1:19:59
What what kind of product do you want
1:20:01
to make? Yeah. I don't know. How about
1:20:04
How about a voice assistant that isn't complete
1:20:06
dog shit? How about one of those? Could
1:20:09
use one of those? Alright.
1:20:15
Let's see another topic. Oh. Intel
1:20:18
releases the first six gigahertz CPU.
1:20:21
Good for them. This
1:20:25
is funny. Mercedes is gonna
1:20:27
be offering level free self driving in America.
1:20:30
No. No. I'm kidding. I'm I'm kidding. Poor Anthony
1:20:32
put together this topic for us. We're down at we're gonna
1:20:34
read it. After CEO, Pat
1:20:36
Gelsinger teased it during the innovation twenty twenty
1:20:38
two keynote, Intel on Thursday formally released the
1:20:40
I nine thirteen nine hundred k s. It
1:20:42
has been apparently through a unique selection
1:20:45
process, so bidding.
1:20:48
It has the same core in cash layout, but
1:20:50
with higher base and boost clocks. It is apparently
1:20:53
a world's first six
1:20:55
gigahertz CPU. This is versus five
1:20:57
point eight for the non s
1:20:59
version of it. It's
1:21:02
E Corus boost the same though. It's only
1:21:04
the p course that will boost higher, which is fine because
1:21:06
from a gaming standpoint, that's all you really
1:21:08
need. Hardware and box got their
1:21:10
hands on the new chip and released a review at
1:21:13
launch unsurprisingly, they're fast.
1:21:16
Yep. They're fast. They're at the top of
1:21:18
the graph. There's a slight
1:21:20
problem, though, aside from the whopping
1:21:22
two hundred and eighty watts of power
1:21:25
that it drew at its five point
1:21:27
five gigahertz all core frequency. Wow.
1:21:30
K? It's
1:21:33
really expensive. It's a hundred dollars
1:21:35
more than the CPU it's based on. That's seventeen
1:21:37
percent more for about three percent
1:21:39
better performance and this
1:21:41
does not factor in AMD whose
1:21:44
four hundred dollars seven thousand seven hundred x isn't
1:21:46
that far off in gaming. To
1:21:49
get the best gains out of the thirteen hundred k
1:21:51
s, it goes without saying that you will need to also
1:21:53
spend a bunch of you'll need
1:21:55
fast DDR five seventy two hundred memory
1:21:57
that gained an additional three percent FPS
1:22:00
though that might apply to the thirteen hundred
1:22:02
k as well. So the question becomes If
1:22:04
I really need three percent more performance, are
1:22:08
there other ways to do it, maybe? And then
1:22:10
do I need another three percent? Let's walk you, go.
1:22:14
Let's get to chaos. Yeah.
1:22:20
I mean, Luke's got a good point. You
1:22:23
know, he thought it would actually be more than
1:22:25
a hundred dollars more. Very few of
1:22:27
these things will probably exist and going
1:22:29
back to, you know, like, third party
1:22:31
bidding services, like Silicon Lottery back when
1:22:33
they existed, these highly bidding
1:22:35
ships would often cost a lot
1:22:37
more. So From
1:22:39
that point of view, you're probably right. It's
1:22:41
downright reasonable because you have
1:22:43
to understand what you're buying. You
1:22:45
are buying essentially handpicked silicon.
1:22:49
Maybe not by an actual hand, but certainly
1:22:51
a robot hand. Certain
1:22:55
things that you'd really prefer human hand
1:22:57
do. But I think robot hand is good enough for this. You know
1:22:59
what I'm saying? You know what I'm saying? Silent
1:23:01
Luke. Yeah.
1:23:04
Should've waited for him to drink. I
1:23:07
wanna get that computer wet. And
1:23:09
computers love me. I turn them on. I'm
1:23:14
sorry. I really shouldn't be making him laugh. For those
1:23:16
of you who are wondering what the fuck is going on
1:23:18
with the WAN show, we
1:23:20
are we are doing a little day in silent
1:23:23
Bob cause play because Luke
1:23:25
had some fairly significant oral
1:23:27
surgery earlier this week. He
1:23:31
cannot really speak comfortably,
1:23:34
but we've got a bit of a street going. We
1:23:36
gotta figure out exactly which was the
1:23:38
first show that we
1:23:41
that we did or
1:23:43
not the first show we did, but we've gotta figure out
1:23:45
how far back we have to go before it
1:23:47
was not me and
1:23:49
Luke on the land show. I I'm pretty
1:23:51
sure we're over two years at this point where neither
1:23:53
of us is mister land show, regardless
1:23:56
of work trips, family
1:23:58
vacations, statutory
1:24:00
holidays, surgery, And
1:24:05
in the interest of keeping the streak alive,
1:24:08
Luke actually scheduled his procedure
1:24:11
as far away as possible from
1:24:13
WAN Show and
1:24:15
had intended to be on the show today
1:24:17
talking, but has had some complications
1:24:20
today that prevented him from fully
1:24:22
participating So I don't remember
1:24:24
whose idea it was, but we came up with the
1:24:26
idea of, okay, well, if you're
1:24:28
gonna what's this mean? Okay. Sometimes
1:24:31
having stupid fucking ideas. So
1:24:33
we came up with the idea of cosplaying
1:24:36
as j and silent Bob. So there would be a
1:24:38
reason for him to be
1:24:40
silent. And then this topic with
1:24:42
YouTube demonetizing channels for swearing
1:24:44
excessively, kind of was
1:24:47
perfect because I needed an excuse
1:24:49
to talk about fucking pussy and Right.
1:24:51
Like Fuck. If I'm gonna
1:24:54
if I'm gonna be the character what? I'm gonna I'm gonna
1:24:56
like, bleep things. Can't
1:24:59
do that shit. Anyway,
1:25:03
Intel has dreamed of speeds of six gigahertz
1:25:05
for twenty years, with their last serious
1:25:07
attempt being the Pentium four net burst
1:25:10
architecture derived Tejas. Tajas.
1:25:12
I forget how to pronounce Tajas. But
1:25:15
the Pentium Pentium five.
1:25:17
These chips were intended to push past gigahertz
1:25:19
to an ultimate goal of ten gigahertz plus,
1:25:22
but they just couldn't do it. And Tejas
1:25:24
was canned by two thousand four in favor
1:25:26
of the lower clocked Pentium M based
1:25:29
core CPUs. Gigahertz doesn't
1:25:31
matter, came true. As
1:25:34
we reach the limits of simply adding more
1:25:36
cores, and with both teams pushing clocks
1:25:38
again instead, might gigahertz matter once
1:25:40
more. The answer is they
1:25:42
always will, they always have, and they always
1:25:45
will. All else
1:25:47
being equal. More gigahertz is
1:25:49
more faster. But the thing
1:25:51
is that a lot of efficiencies were
1:25:53
gained by building a more a
1:25:57
more power efficient architecture, especially
1:25:59
as we moved into the multi core era,
1:26:02
such that gigahertz
1:26:04
was not the only answer to
1:26:06
the problem. So I
1:26:08
think there's always gonna be a little bit of of
1:26:11
ebb and flow. Are we chasing gigahertz? Or are
1:26:13
we chasing architectural efficiencies? Realistically,
1:26:17
for a long time, it's just kind been a little
1:26:19
bit of both. Discussion
1:26:22
question is what direction do you see the industry
1:26:24
going and following this milestone? More clocks be
1:26:26
more cash, more cores, more power, something else.
1:26:29
I don't think we can really push power much higher,
1:26:31
at least not for monolithic dies. We
1:26:33
saw with AMD's release of the 9X7
1:26:36
thousand series CPUs that at
1:26:39
even pretty high power draw, a chiplet
1:26:42
design, spreading out that
1:26:45
that power dissipation helps a lot
1:26:47
with thermal management. Like, yeah, it's still a ton
1:26:49
of heat. You still need a big fat heat sink,
1:26:51
but you're not gonna have these hotspots that are
1:26:53
absolutely gonna kill any attempt to
1:26:55
cool these things. Like, that's the thing, right,
1:26:57
is you could have a one hundred watt
1:26:59
chip. Right? That is impossible
1:27:02
to cool because the dye is
1:27:04
so small that the physics
1:27:06
of moving the heat away from it fast enough are
1:27:08
impossible. Right? Or
1:27:10
you could have a four hundred watt chip in
1:27:13
a server that's this fucking big
1:27:15
has chiplets all over the damn thing And
1:27:17
yeah, you need a big heatsink on that bitch, but
1:27:19
like, but
1:27:24
it's not a problem to get the heat
1:27:27
out of the chip into that heatsink.
1:27:30
So I don't see pushing power much
1:27:32
higher for desktop chips where realistically
1:27:35
wafers are going up and price not down
1:27:37
and we're not going to see significantly larger
1:27:40
dies So we're we've just reached point
1:27:42
where we can't really move heat. I mean, Intel's already
1:27:44
thinning out, not just the IHS. They've
1:27:47
been thinning the die. For multiple
1:27:49
generations now to try to get heat out of it more
1:27:51
efficiently. Like like we are at the we
1:27:53
are at the razors edge of what's possible
1:27:55
now. I
1:27:58
do see them I do see them continuing
1:28:00
to try to push clock speed as much as they can, but it's
1:28:02
hard to do without more power. So man,
1:28:05
what are you what are you gonna see?
1:28:07
More cash is gonna be tough. Man,
1:28:10
I'm sound I'm sounding like
1:28:12
a bad news bearer here. More
1:28:14
cash is gonna be tough. We've seen news coming out
1:28:16
of TSMC that the latest node
1:28:18
trinkets are not
1:28:21
really making cash any smaller. Which
1:28:23
is pretty
1:28:26
tough for AMD's strategy of throwing more
1:28:29
cash at their chips to dramatically boost gaming
1:28:31
performance. I think the x three d's of this
1:28:33
generation are gonna be pretty freaking expensive.
1:28:35
AMD's gonna be able to point at their non x's
1:28:37
and say, hey, we're still a great value. Right?
1:28:40
AM five is great platform, great value,
1:28:42
but you wanna go fucking fast. You're gonna have to
1:28:45
pay some fucking money. You know what I'm saying? I
1:28:50
mean, I think I think Luke
1:28:53
Luke's right, chiplets are a hundred percent the future.
1:28:55
But, I mean, what? You want do you
1:28:58
really want more course? I mean,
1:29:00
yeah. Yeah. Your team works in development. You're gonna
1:29:02
parallelize everything. Good fucking luck. Ultimately,
1:29:13
Single core. Yeah. Single core performance is
1:29:15
always gonna matter. Yeah.
1:29:18
It's always gonna matter. And it's
1:29:20
still it's still a bottleneck today. Like,
1:29:22
no matter what you're doing, it still matters.
1:29:27
Alright. Let's move on to our
1:29:29
Oh, this is oh, this is a big topic.
1:29:31
Praise Saint Cook, hero of the people.
1:29:34
Tim Cook voluntarily takes a forty percent
1:29:36
pay cut. On Thursday, Apple
1:29:38
announced that CEO Tim Cook would be taking
1:29:40
a forty percent pay cut going from
1:29:42
eighty four million dollars last year
1:29:45
to just forty nine million dollars
1:29:47
this year. Okay.
1:29:50
I don't really know if I like your attitude. This is
1:29:52
some pretty sad shit right here. How's
1:29:57
Tim Cook's family gonna have generational wealth
1:29:59
now? This
1:30:03
decision was made. After the company's
1:30:05
board committee. I mean, it doesn't
1:30:08
sound like an exciting committee, I guess, to get to
1:30:10
talk about how much more money Tim Cook makes than
1:30:12
them. The
1:30:15
decision was made after the company's
1:30:17
board committee on executive compensation, balanced
1:30:21
shareholder feedback, Apple's
1:30:23
exceptional performance, and
1:30:25
a recommendation for mister Cook. Okay,
1:30:27
For real though. To adjust his compensation
1:30:30
and light of feedback received, Tim's
1:30:32
annual basic salary will remain
1:30:34
three million dollars with bonus of up to six
1:30:36
million dollars. What's changing
1:30:38
is his stock award target, which will be cut
1:30:40
to forty million as opposed to the seventy five
1:30:42
million that he received last year. Last
1:30:45
year Apple apparently wanted to pay Cook a ninety
1:30:47
nine million PayFac package,
1:30:49
but shareholder advisory firm ISS
1:30:52
said it was too much. It's
1:30:57
very interesting. Like, once you get into these
1:30:59
numbers that are just so far beyond, like,
1:31:02
any reasonable amount of okay.
1:31:04
Tell me something, Luke. Okay. What what what
1:31:06
what does this work out to? Even now, forty nine million
1:31:09
dollars a year. So that works out to a seventh
1:31:11
of a million dollars a day. So
1:31:13
about about a hundred and thirty million
1:31:15
dollars a day. Alright? Alright.
1:31:17
It's a thousand thousand. Sorry. Excuse me. hundred
1:31:19
and thirty thousand dollars a day. Yeah.
1:31:23
Can we become a shareholder firm that says too
1:31:25
much? Yeah. That's advisory
1:31:27
firm. Okay. So a hundred thirty thousand dollars
1:31:30
a day, Luke. Could you spend a hundred
1:31:32
what was that workout to per per waking
1:31:34
hour? So so you're awake for what?
1:31:36
Let's say sixteen hours. Okay? So
1:31:39
you make about
1:31:42
eight thousand dollars a waking hour.
1:31:45
Oh, man. That was pretty close. No.
1:31:47
Nice. Okay. You make eight thousand dollars
1:31:49
an hour. Could you spend eighty
1:31:53
dollars in a minute? Could
1:31:58
you spend eighty dollars every minute? Like, what
1:32:00
kind of food what kind of fucking food would you
1:32:02
eat? Like, how long is it?
1:32:05
At the rate that you eat, my God,
1:32:07
you could never He
1:32:11
could eat forever on that amount of money.
1:32:13
He could I mean, this fucking guy could eat
1:32:16
forever on, like, a thousandth
1:32:18
of that. AA1 hundred thousandth
1:32:20
of that got you eat so slow. So you think
1:32:22
he wouldn't be able to get as big as he is.
1:32:29
Oh, oh, wow. Yeah.
1:32:31
You're gonna be pretty slow eating now.
1:32:34
Unless you're eating that pussy. Look,
1:32:37
I'm I'm supposed to be Jay. Right? That's
1:32:39
what he would say. Gets
1:32:42
right into it. I'm
1:32:52
actually hurting him. I
1:32:55
I'm sorry. Anywho,
1:33:00
I wanna I wanna I wanna make it clear.
1:33:03
We actually did
1:33:05
praise the late CEO
1:33:07
of Nintendo for taking a fifty
1:33:09
percent pay cut after the Lukewarm
1:33:12
launch of the 3DS and the failure of the wii
1:33:14
u. Right?
1:33:17
Like that, it it is actually a pretty cool thing.
1:33:20
For a CEO to take a pay cut when the company
1:33:22
doesn't perform well because why
1:33:25
should why should other people's jobs get cut
1:33:27
like that stupid, sad CEO
1:33:29
talking about, like, crying about how hard
1:33:32
it was to lay people off. Crocodile
1:33:36
tears. Right? Like, how how much are you gonna cry when
1:33:38
it's your family that's worried about where their next meal
1:33:40
is gonna come from. And to be clear,
1:33:42
I'm not saying that I'm some kind of
1:33:44
like, you know, softy. I would never, you
1:33:47
know, fire anyone. Like, that that happens.
1:33:49
That's business. But, you know,
1:33:53
when you're doing it because the
1:33:55
company overall didn't perform,
1:33:58
Well, shit rolls up the hill. Like,
1:34:00
the executive team has to
1:34:03
has to take accountability for that.
1:34:05
Right? The difference
1:34:07
is that Awada's basic annual
1:34:09
salary was seven hundred and seventy thousand
1:34:11
dollars, with up to two point one one million
1:34:14
of performance based bonuses. It
1:34:17
wasn't three
1:34:19
million dollars with six million dollars of bonuses
1:34:21
and another seventy five million
1:34:23
dollars of performance based bonuses. Or
1:34:26
a stock to stock stock award target. Jacob
1:34:33
Scholes says, apparently, internally at Apple,
1:34:35
everyone spent the entire day mocking him. I can't
1:34:37
validate that. You know,
1:34:39
I have no idea, but it wouldn't surprise
1:34:42
me. Like Anyway,
1:34:47
so, yeah, good luck with that, Tim. Hopefully
1:34:51
your hopefully your private jet payments
1:34:53
don't fall behind. Alright.
1:34:56
What else we got? This
1:34:59
is an ultra rapid fire topic.
1:35:02
The LTT subreddit brought to
1:35:04
my attention that Anchor
1:35:06
was still using me as a spokesperson
1:35:09
on their site. Hey, thanks, subreddits.
1:35:13
You know, post on the forum next time.
1:35:15
But, no, the subreddit is cool too.
1:35:18
Anyway, it's gone. We've got
1:35:20
that I'll dealt with. So
1:35:22
here's their site. If you guys
1:35:24
weren't keeping up with the drama, we dropped Anchor
1:35:26
as a sponsor after their Yuffie
1:35:29
sub brand security cameras. Were
1:35:31
marketed on big fat lies. We
1:35:34
were we were not super happy about the way
1:35:36
that whole thing went down. Oh, that's great. Okay. So Lou
1:35:38
just gets more It
1:35:41
just gets a wider spot, I guess.
1:35:44
Alright. Cool. Good
1:35:48
stuff. Alright. We got a couple
1:35:50
more topics here. Mercedes will be offering
1:35:52
level three self driving in the US.
1:35:55
And they will be the first in
1:35:57
front of Tesla. They got
1:35:59
approval from the state of Nevada and
1:36:01
they will be the first to offer a production vehicle
1:36:04
with level three driving. The
1:36:06
jump from level two systems, which include Tesla's
1:36:08
autopilot and Cadillac's super cruise
1:36:11
systems, is a substantial leap and
1:36:13
includes apparently environmental detection
1:36:15
capabilities that allow the system to make informed
1:36:18
decisions such as accelerating to
1:36:20
pass a slow moving vehicle. Level
1:36:23
three systems, to be very clear,
1:36:25
still require a human driver to remain ready
1:36:27
to take over control and the systems quest,
1:36:29
so you can't just, you know, have
1:36:32
have someone suck in your dick while you're behind the wheel.
1:36:38
I mean, I guess you could as long as you're ready
1:36:40
to take over control at the system's request.
1:36:42
Just don't let it just make sure it's not a make sure
1:36:44
it's not a good blow job, not too distracting. So
1:36:49
that's good. And, anyway, Audi's
1:36:52
twenty nineteen a eight sedan was supposed
1:36:54
to be the first production vehicle with Level features in
1:36:56
Europe, but ultimately they decided the market and
1:36:58
infrastructure wasn't ready back in twenty twenty.
1:37:01
This is this is pretty exciting. I still man,
1:37:03
I My
1:37:06
new car has some, like, assist features, like,
1:37:09
even like lane keep assist. Yeah. Man, I
1:37:11
drive without it on. I just I like
1:37:13
to drive my car. I feel like I
1:37:15
thought I was gonna be all automation all the
1:37:17
time when it came. I don't know.
1:37:19
I don't know if I'm not into it. Maybe, like,
1:37:22
sometimes, like, if I gotta if I'm work if I have
1:37:24
to work. Or something like sit in the back
1:37:26
seat and work while the car drives itself, that'd be
1:37:28
kind of cool. But I I
1:37:31
like to drive. I
1:37:33
also have a pretty nice car. I'm
1:37:36
I'm actually really liking the new car. It's
1:37:38
III never really cared. I've
1:37:40
never been a car guy but I
1:37:43
I definitely am I definitely am liking
1:37:46
the new car. Yeah.
1:37:54
Still yeah. Yep. Yep. Yep.
1:37:57
Luke likes it too. What
1:38:00
what is it not good enough for blow jobs? Is
1:38:02
that wait. Where did you say that before I
1:38:05
oh, my car's automated driving? Well, yeah. I don't know.
1:38:07
It's not even close. Let me tell
1:38:09
you. I
1:38:14
tried to give a blowjob in that car. It's
1:38:18
like awful. The bolsters on the seat, I can't even reach
1:38:20
the passenger seat. Alright.
1:38:28
Why don't we go ahead and do some some merch
1:38:30
messages here? Don't don't say that.
1:38:33
Okay. Merch. Don't. Okay. That's fine, though.
1:38:35
In that case now, we got some jobs. We got some jobs.
1:38:37
Let's go. Let's do jobs. Oh, wow. We have different
1:38:39
pages for different companies now. Look at this.
1:38:43
Okay? So there's there's
1:38:45
jobs, jobs, jobs, float plane jobs.
1:38:48
Junior back end developer, Luke, you gotta correct this
1:38:50
if any of this is wrong. Is this all right? Full time back
1:38:52
end web developer, full time
1:38:54
front end developer, full time
1:38:56
machine learning slash computer vision engineer,
1:39:00
Okay. Why work here? Equal opportunity statement,
1:39:02
job perks. Hey.
1:39:05
Hey. We don't have our don't have our bloody
1:39:08
thing. The good thing that we have, the the
1:39:10
GRSP plan. Okay. We gotta get that
1:39:12
shit in here. Hey, Dan. Do you mind just sending
1:39:14
message over to HR? Yep.
1:39:17
And whoever manages this website,
1:39:19
that should be really quick to put up there
1:39:21
because of Squarespace. Okay.
1:39:24
For creator warehouse, we need
1:39:26
a full time senior retention marketing
1:39:28
specialist. Okay. Whatever the
1:39:30
hell that is. Full time electronics engineer
1:39:33
slash product designer. A full time fit technician.
1:39:36
Gotta make more garments. Alright. Cool.
1:39:38
So we got that. And then, 0000,
1:39:43
Here we go. What's going on
1:39:45
here? Oh, are not in
1:39:47
hiring right now, but they will be this year
1:39:50
and will be using the resume. Oh, okay. So
1:39:52
bear that in mind. We will be
1:39:54
hiring those positions. And
1:39:56
honestly, like, if it's stellar, I'm sure
1:39:58
we would just make it work. But sometimes,
1:40:00
you know, timing's not always perfect. So
1:40:04
Okay. And then Lions Media Group is looking for
1:40:06
a full time procurement manager, full
1:40:08
time sales supervisor, full time
1:40:10
logistics coordinator, full
1:40:12
time accountant, full time bookkeeper,
1:40:15
full time video editor slash camera op,
1:40:17
full time social media coordinator, full
1:40:20
time writer slash video producer, full
1:40:22
time production assistant. And,
1:40:25
yeah, we're man, we're hiring a lot of positions.
1:40:27
You want you want me to look at your thing. Right? It's
1:40:32
alright. Also, the jobs
1:40:35
on the float plane page are not only for float plane, which is
1:40:37
confusing. Yeah. I think they're just jobs that you're gonna have to
1:40:39
hire for. That kind of makes sense,
1:40:41
sort of not really, but yes, but no,
1:40:44
but yes. Anyway,
1:40:46
thanks, Luke. Good chat. Alright.
1:40:51
Oh. No. We got one more topic. Let's do
1:40:53
our last big topic and then we'll get into some good
1:40:55
merch messages. You guys probably
1:40:57
noticed we didn't promote really anything on
1:40:59
the store today. That
1:41:01
is for us to have one more
1:41:04
big push this weekend. We actually do have a
1:41:06
couple products that we could launch. But
1:41:08
I talked to Nick about it. Adam,
1:41:10
our our new customer experience supervisor,
1:41:13
and we basically went, okay, look, Let's
1:41:16
not do a big burst of
1:41:18
sales this weekend. Let's
1:41:20
save it. Let's get everything completely caught
1:41:22
up. Our goal for this year is to measure
1:41:25
our customer support response times in
1:41:27
hours, not in days.
1:41:29
So we're gonna get everything completely cleaned
1:41:32
up next week. And then next
1:41:34
week, hopefully, we'll have like good promo for you
1:41:36
guys or a big launch or something like that. Alright.
1:41:39
Seventy nine hundred XTX problems. After
1:41:42
its release in December, The AMD seven thousand
1:41:44
nine hundred XTX has
1:41:46
had some issues. There have been number of users reporting
1:41:48
that it was hitting hot spot temps above hundred and
1:41:50
ten degrees and slowing the GPU down. Amy
1:41:53
started by saying this was normal, but soon
1:41:55
after they admitted that some reference cards suffered
1:41:57
a manufacturing defect in vapor chamber.
1:42:00
Aandy was then interviewed by Gordon at PC
1:42:02
World, and according to Durbauer, some
1:42:04
questionable statements were made.
1:42:06
In the interview, Scott Herkleman from Aandy
1:42:08
said, is there performance issue? What we
1:42:10
found is if you throttle at hundred and ten degrees
1:42:12
in certain workloads, you can see a small performance
1:42:15
delta. Okay.
1:42:21
This did not line up with Dervauer's testing
1:42:24
that showed that three out of four cards
1:42:26
he tested could only dissipate two
1:42:28
fifty to two eighty watts power consistently,
1:42:31
eighty watts lower than it should be, which can
1:42:33
drop performance by ten to twenty percent.
1:42:36
Whoops. You gotta
1:42:38
protect my voice. Whoops. The problem
1:42:41
appears to be stemming from not enough liquid in
1:42:43
the vapor chambers. What's strange is that
1:42:45
AMD says they traced the problem to a bad
1:42:47
batch of coolers and that customers can get
1:42:49
in touch with customer support to figure things out.
1:42:52
But that doesn't really make sense because if it only
1:42:54
affected a single batch, it should be relatively simple
1:42:56
to use serial numbers to figure out where those cards
1:42:58
are and preemptively recall them. I
1:43:02
mean, yeah.
1:43:04
Is it it's yeah. Like, why are they relying on
1:43:06
customers to test a thousand dollar GPU?
1:43:08
To figure out if it's defective. Like, what are they? What
1:43:10
are they? Your RMA department? This
1:43:14
either means that the defect isn't confined
1:43:16
to single batch or that AMD just
1:43:18
doesn't really care, and as long
1:43:20
as these GPUs don't die within
1:43:22
the warranty period, they're just ultimately not
1:43:24
gonna be their problem. The
1:43:27
larger problem is that AMD
1:43:29
is super low on seventy nine hundred XTX
1:43:31
stock, so only about one out of three
1:43:33
of customers affected will be able to get
1:43:35
ACE card in the first two weeks in Germany.
1:43:38
Dervé recommends just taking a refund from
1:43:40
AMD and buying an AIB seventy
1:43:42
nine hundred XTX instead of waiting for AMD
1:43:44
to replace your reference card. Or
1:43:47
if this is just kind of like fucking
1:43:51
stupid, you could just not
1:43:53
buy an AMD card, but then you're supporting Nvidia
1:43:56
who has their own problems. I
1:43:59
know. Right? There's
1:44:01
been some speculation that GPUs have died
1:44:03
as a result of this hotspot, so Jabauer
1:44:06
received a broken card from viewer to teardown
1:44:08
and you absolutely should watch
1:44:10
the moment that he turns it on for the first time,
1:44:12
not safe for nerds. With
1:44:14
with the disassembly done, looks like there was
1:44:17
faulty VRM. So the hot spot temp had nothing
1:44:19
to do with the GPU dying. With a better
1:44:21
cooler, it still would have led up the magic smoke though.
1:44:23
Some bonus not safe for nerds content, sixty
1:44:25
nine hundred s t's are also appearing to be dying
1:44:27
in large numbers with one
1:44:30
German shop receiving forty eight dead GPUs.
1:44:32
The picture of one of the GPUs is truly
1:44:35
horrific. Let's
1:44:37
see. Yikes.
1:44:41
Glad did that die just crack? What
1:44:43
makes a die crap? Oh, it came off on the cooler.
1:44:46
Why would it come off on the cooler? What
1:44:49
the hell? Wow.
1:44:53
Okay. With
1:44:56
that said, forty eight is not necessarily
1:44:59
an enormous number. It just depends
1:45:01
on how many this particular this
1:45:04
particular store sold. Yeah. Some
1:45:08
people are speculating that a driver issue caused
1:45:10
this, although much like the faulty VRM GPU
1:45:12
that Durbauer showed, it
1:45:14
is possible for driver kill a GPU,
1:45:17
like for real. Like, a driver could
1:45:19
potentially tell it to to draw
1:45:21
way too much power. Or
1:45:23
it could it could disable a
1:45:25
a thermal safety. You know, yeah, it's possible.
1:45:28
But I I think that's fairly unlikely Our
1:45:31
discussion question is if you were AMD, how would you try
1:45:33
and solve the bad vapor chamber problems? I mean,
1:45:35
we had some problems with screwdrivers. We contacted
1:45:37
everyone who bought one. We had them
1:45:39
keep their existing driver. We don't serialize
1:45:41
the screwdrivers because how,
1:45:44
like, realistically, there's
1:45:46
there's no there's no onboard programming. Like,
1:45:48
we wouldn't have a way of actually serializing them
1:45:51
that couldn't be that couldn't be forged.
1:45:53
Right? So what are we gonna do? Have a different
1:45:56
plastic mold for every stupid driver
1:45:58
that we're gonna put a sticker on it, like, whatever.
1:46:00
We don't serialize them. So what we had people do
1:46:03
was just mark the ones that we knew were
1:46:05
bad. To scratch an x, send
1:46:07
a picture to us of the scratched one so that we'd
1:46:09
know that no, that one does
1:46:11
not have any existing or does not have
1:46:13
any remaining warranty and then we sent them a brand new one
1:46:15
that has a brand new warranty. That's
1:46:18
how we would do it. A
1:46:20
and D has, you know, can serialize
1:46:23
cards though. So maybe they could,
1:46:25
you know, have people return the card
1:46:27
at least. And then they wouldn't be out
1:46:29
the entire cost of the card. I don't know. Would they would they
1:46:31
salvage memory chips? Probably not even. They're probably
1:46:33
just You know? I
1:46:36
would like to know the answer to that. What
1:46:39
does happen to
1:46:41
a a legit like, I'm not talking a customer
1:46:44
return. Because I know where that's gonna end up. That ends up
1:46:46
in the open box pile. Right? What
1:46:48
happens to a legit dead GPU?
1:46:51
Memory chip goes bad. Is
1:46:53
someone remanufacturing that? I
1:46:56
don't know. I I'd be really I'd be
1:46:58
I'd love to follow the journey of
1:47:02
IIIII legitimately dead
1:47:04
GPU. Does it just go into an e waste pile?
1:47:07
I don't know. Probably. But
1:47:09
then some of them are extremely valuable. Like,
1:47:11
you've got GPUs where the memory chips on
1:47:13
the malone are worth a couple hundred dollars. Right?
1:47:16
And the IT industry is notoriously
1:47:19
stingy. I have
1:47:21
I have no idea. Yeah. I mean, you could recycle
1:47:23
it, but that still cost you two hundred dollars. Right?
1:47:26
Unless someone's gonna use those chips, but then if you
1:47:28
use those chips. You can't sell it you can't sell it as
1:47:30
new unless you just do. I don't know.
1:47:32
I mean, it's under a cooler who's gonna know.
1:47:37
Alright. I don't think we have any LTX updates
1:47:39
this week, Dan. You wanna get me some merch messages?
1:47:42
You ready? I'm finally ready. Yes. I've got
1:47:44
one here from Joe. Hi, Joe.
1:47:46
I've been listening to this show in podcast
1:47:48
form for six to eight years now.
1:47:50
Just so you know, at least one of us does exist
1:47:52
You're a confused man, Joe. You're a
1:47:54
confused man. Six and eight years. Six and eight are
1:47:57
two whole completely different numbers. I
1:48:00
also play competitive Mario Kart
1:48:02
We. What's your favorite Mario Kart
1:48:04
game? Mario Kart competitively.
1:48:07
Goddamn it, Dan. Mario
1:48:11
Cart. It's a me.
1:48:13
Mario. Merrill
1:48:17
Still better than Chris Pratt. Mario
1:48:21
Cartes. Yes. Like a British
1:48:23
person, like I'm supposed to talk like British.
1:48:26
Mario. It's Mario. It's far more.
1:48:28
Mario. Mario. I don't know.
1:48:31
Oh. I can't do a British accent. My
1:48:34
favorite, man. One I probably played the
1:48:36
most was for the DS. I don't
1:48:38
even remember what that one was called. I
1:48:41
think it was the one legit game I
1:48:43
actually owned for that console. Everything else was on
1:48:45
my r four. No. No. I also bought
1:48:47
Phantom Hourglass. Got suckered into
1:48:49
that one. What a terrible game. Yeah.
1:48:52
That's probably the one that I played the most. And
1:48:54
it's the only one that I ever played online.
1:48:57
I don't know what the fuck that is. I
1:48:59
I for you, it'd say, yeah, I know.
1:49:02
I mean, they they didn't ask you. Okay.
1:49:04
Okay. They didn't even answer.
1:49:06
It's not like you didn't even on. The
1:49:09
one I probably played second most was sixty four.
1:49:12
So this is this just comes down to like what
1:49:14
consoles I owned and when I was at a time in
1:49:16
my life when I, like, had any time to
1:49:18
do anything. Yeah.
1:49:21
I'd say that I'd I'd say my my favorite
1:49:24
was probably Mario Kart DS. Was
1:49:26
that what it was called? I don't even remember. Yeah.
1:49:29
Alright. Let me. Okay. If you can say
1:49:31
Jif, I can say Mario. No.
1:49:33
Jif is correct. Mario is dumb and
1:49:35
bad. We
1:49:38
both have stuff to work on. This one's from Reed.
1:49:40
Yeah. I've got some stuff to work on. We work on this
1:49:42
guy. Yeah.
1:49:46
When should purchasers of tickets of the canceled
1:49:48
LDX look out for the email, allowing them
1:49:50
to rebuy the tickets they bought before
1:49:52
the official sale of the tickets. Also,
1:49:55
Luke, why don't casting seem to work
1:49:57
on Flowplane Android app or Chrome?
1:50:00
I take your silence as you're working on it.
1:50:03
If you message support, I'm sure they can
1:50:05
help you out with that. Yep. Message flow clean support,
1:50:07
not LTT store support, not someone
1:50:09
on the forum, don't post on the forum, message,
1:50:12
floor plan support, and they can they can get you an answer
1:50:14
for that. As for the purchasers of
1:50:16
tickets, I don't know.
1:50:22
I don't know if we have LTX support.
1:50:26
We should at some point. We should
1:50:28
figure that out. Hey,
1:50:30
Dan. Can you send a message to
1:50:32
Colton and Chase to figure
1:50:35
out what the hell are plan
1:50:37
for support for RTX's. Yeah. I've gotten some
1:50:39
questions about security and things like that. Okay. I
1:50:41
think a lot of it is just still in the works.
1:50:44
Okay. Yeah. But we should have somewhere
1:50:46
that people can start submitting questions
1:50:48
so that we can populate an FAQ
1:50:50
with, you know, FAQ.
1:50:54
Mhmm. Get it. FAQ, yeah, anyway.
1:50:56
The point is we can populate it with answers to those
1:50:58
questions. Alright.
1:51:02
I'll get you another one if you want. Yeah. K?
1:51:06
This is from g's floor. That's a very
1:51:08
strange first name. I'll be using this driver
1:51:10
to build my home lab. Is there
1:51:12
any interesting that
1:51:14
you is there anything interesting that
1:51:17
you're running on your home server besides
1:51:19
plex and home assistant? That you haven't
1:51:21
shared. PS has
1:51:23
creator warehouse ever thought of designing
1:51:25
a vest? We,
1:51:27
I think, I think we have a vest that
1:51:29
we worked on at some point. I don't know if
1:51:31
it's ever gonna come to light. I'm
1:51:34
not really a vest guy to be perfectly honest
1:51:37
with you. Like, I
1:51:39
mean, we could do a fedora, but
1:51:42
we won't probably unless we
1:51:44
do. I mean, Now
1:51:47
I kinda wanna do like a rock in fedora,
1:51:49
like a super awesome fedora. It's
1:51:52
like with the finest materials. Oh,
1:51:55
I'm gonna have a lot of velvet soon if you wanna
1:51:58
do a velvet shoe. No. I wanna
1:52:00
do it. I
1:52:02
don't know. We have, like, a Minecraft server, like,
1:52:04
just like a survival server that
1:52:06
we play a family Minecraft on once in a while.
1:52:09
I think, like, what else I got on there? Not not too
1:52:11
much honestly. I mean, I use it as a Nas obviously,
1:52:13
but that that's about it. I'm not a
1:52:15
super demanding user at home.
1:52:18
That's the thing about, like, working all
1:52:20
the time is you don't play much. Like,
1:52:22
I have the coolest toys and I hardly
1:52:24
touch them. Like, you
1:52:27
know, Ivan. Right? Formerly,
1:52:30
like, Ivan who used to work here was he
1:52:32
was begging me. He's like, hey. We should go, like,
1:52:34
to see the sky highway on the bike. So I'm like,
1:52:36
dude, I have literally not
1:52:38
once in my life gone for a recreational ride.
1:52:41
Like, nothing personal. I just
1:52:44
probably won't do that. But, like,
1:52:46
if you wanna grab lunch or whatever, that's a different
1:52:48
question. Like, I can probably I can probably make
1:52:50
that work. But like I got this cool bike and
1:52:52
I literally have only ever commuted on
1:52:54
it. Think okay. Okay.
1:52:59
This one's from Michael Linusluc. What's
1:53:02
your take on using blockchain technologies
1:53:04
to help resolve monetization issues
1:53:06
for creators? I think that it's unfortunate
1:53:09
that blockchain blockchain technology has
1:53:11
such a a negative vibe
1:53:14
around it right now, because there are legitimately
1:53:17
really interesting things that could be done with it.
1:53:19
The problem is just that it's
1:53:21
completely unregulated and has become
1:53:23
space where grifters and scammers can
1:53:25
exploit people. And so that that
1:53:27
harms the reputation that it has. And
1:53:30
I think harms its utility for
1:53:32
legitimately useful purposes. Like, I talked
1:53:34
a a little bit on a previous show about
1:53:36
this company that had this goal of creating
1:53:39
like a stock exchange
1:53:42
essentially for creators to publicly
1:53:44
list their companies using
1:53:46
blockchain technology as
1:53:49
the as the validation of the proof of ownership
1:53:51
of shares. So
1:53:54
instead of buying some fucking aper
1:53:57
picture or whatever, like just some bullshit,
1:53:59
you would actually be buying a share legally
1:54:03
in Linus Media Group Incorporated. That
1:54:06
could then go up in value with,
1:54:08
like, you know, the way that shares go
1:54:10
up in value with Linus Media Group Incorporated,
1:54:12
you know, outperformed its quarterly
1:54:15
estimates or whatever the fuck financial
1:54:17
shit happens. And We
1:54:22
would also make money from the initial offering
1:54:24
of shares, and then from any secondary
1:54:27
market movement of shares. And, like, the whole
1:54:29
thing just, like, kind of actually makes sense,
1:54:31
and blockchain would be a great way
1:54:33
to do that. But, yeah, this is a great
1:54:35
point. Roberto in the flip plane chat says
1:54:37
kinda like torrent. Yeah. Like, like
1:54:40
torrent technology. It's a super cool
1:54:42
tech that is getting a really, really
1:54:44
bad rep. A
1:54:46
relative of mine, a
1:54:48
member of my extended family, let's put it
1:54:50
that way, is working for a company
1:54:52
that is using blockchain technology
1:54:54
to to help
1:54:56
with aerial mapping, which is super
1:54:59
cool. So essentially, what you
1:55:01
do is you submit mapping
1:55:03
units like you you record
1:55:05
and submit map and
1:55:08
you get these tokens. And then
1:55:10
when people need aerial maps
1:55:12
of that place, they they
1:55:16
compensate your tokens.
1:55:19
Right? Are are they they have to buy
1:55:21
tokens that make your tokens go up in
1:55:23
value or whatever. Basically, there's like this
1:55:25
this mapping economy. So drone operators
1:55:27
in their spare time can just fly
1:55:29
around mapping shit and get compensated
1:55:32
anytime anyone utilizes the mapping.
1:55:34
That's so cool. Right? But
1:55:37
as soon as you say blockchain, people are like,
1:55:39
oh, bloves
1:55:41
in the -- Right. -- crypto. No.
1:55:45
The tokens make sense in this case, Jelliedi. Why
1:55:49
not? Because in this
1:55:51
case, there is a clear reason
1:55:53
for people to exchange real fiat currency
1:55:56
for the tokens. Right?
1:55:59
It's it's actually a good thing.
1:56:02
Super cool. Alright.
1:56:07
What else we got? Okay.
1:56:09
I've got another here from Yossak. Hey,
1:56:12
linus and Luke what are some challenges you
1:56:14
faced when redesigning the screwdriver to
1:56:17
the short version or the backpack
1:56:19
to the slimmer version? Hold on a second.
1:56:21
Uh-oh. Algarism in the twitch asked,
1:56:23
why do you need a marketplace for maps when they
1:56:25
can be shared freely on the Internet at barely any
1:56:27
cost? Because
1:56:32
the drone operators time is not fucking free.
1:56:35
Why should it be barely any cost? Why don't they
1:56:37
get compensated? Besides, we're not
1:56:39
talking satellite imagery that is freely
1:56:42
available on Google Maps. We're talking, like, three
1:56:44
d, like, topographical maps.
1:56:46
Right? Like, we're talking high resolution
1:56:49
maps. We're like moving
1:56:52
into the future. We're talking about, you
1:56:54
know, you could invest in a drone that is
1:56:56
capable of creating, like, super, super
1:56:58
accurate maps and anyone who needs
1:57:00
the resolution be it for, you
1:57:03
know, I don't know resource exploration
1:57:06
or whatever it is would have
1:57:08
a a very strong financial incentive to
1:57:10
pay for it. This isn't just for like random
1:57:13
to be like, oh, it'd be kinda cool to, like, see
1:57:15
a map of, like, this, like, forest bit here.
1:57:18
That's that's not what that's not what it's for.
1:57:20
And anything that has a commercial purpose.
1:57:24
Right? You should be paid for.
1:57:27
Like, we we have a
1:57:30
very clear policy here. If we are
1:57:32
going to make money on something,
1:57:34
we insist that we pay for your
1:57:37
work. Period. Like,
1:57:39
that's just that's just the way it is. There are
1:57:41
things that we don't make money on and we'll never
1:57:43
make money on, and we're very very grateful
1:57:45
to people for their for their
1:57:48
their their their very
1:57:50
very generous contributions. I mean, someone
1:57:52
like a colonel Mortise, for example, would really stand
1:57:54
out to me as someone who's just been
1:57:57
been an amazing member of our community over
1:57:59
the years. You
1:58:03
know what? I think there's another exception. We
1:58:05
take volunteers at LTX. I'll be honest
1:58:08
with you guys though. If if you had everyone
1:58:10
does it, and the model is
1:58:13
not feasible. It's
1:58:15
just literally not feasible if you don't take
1:58:17
volunteers to help you run a large scale event like
1:58:19
that. We like we looked
1:58:21
at the numbers, it's
1:58:24
wild. And they are still
1:58:26
compensated, so they get, I think, the way that
1:58:28
it works. If you guys wanna volunteer, I think the
1:58:30
sign ups are live on LTXX0 dot
1:58:32
com. The way it works is I think
1:58:34
you get two days of admission and you
1:58:36
volunteer one of the days. Is is
1:58:38
how it typically works. And then there's also, like,
1:58:40
some, like, some, like,
1:58:42
some swag pack type stuff and stuff like that.
1:58:46
And I got a message back from Chase about LTX.
1:58:48
yeah. They're trying to set something
1:58:51
up next week hopefully. Cool.
1:58:55
Okay. Okay. This
1:58:58
one's for Eric. From Eric. Sorry.
1:59:00
Oh, is this the one you read before? No. I think
1:59:02
this is a a different Did did you read 10I
1:59:05
might have read this one before. Yeah. The
1:59:07
LTT store dot com. Okay. The
1:59:09
things that will be at LTT store dot com
1:59:11
at LTX will mostly just be
1:59:13
like what you would find on LTC store dot
1:59:15
com. Yes, we plan to have screwdrivers and
1:59:18
or backpacks there. Glad
1:59:20
you picked up a water bottle. Yeah. We'll we'll definitely
1:59:22
have more stuff there. It's not gonna be like
1:59:24
the LTT store dot com concepts that
1:59:26
I shared before where we're gonna have like
1:59:29
stuff that was tested by the lab, you
1:59:31
know, like discounted electronics or whatever, I
1:59:33
still wanna do that, but that would definitely not
1:59:35
be an LTX. K. This
1:59:37
one's from Michael. How many t shirts do I
1:59:39
need to buy get line of mutis to say
1:59:41
snoochie butches? What?
1:59:46
think that's from one of the later silent
1:59:48
fog. And yeah,
1:59:51
Oh, okay. Yeah. See, I
1:59:53
I have not watched every I I talked about
1:59:55
this in the pre stream. I have I haven't watched,
1:59:57
like, every j and silent, Bob. Adjacent
2:00:01
piece of content. Sure.
2:00:04
Snooty botches, I guess. There you go.
2:00:06
What it means? What cost one t shirt? Is
2:00:09
it really bad? I have no idea.
2:00:12
Okay. I'm not gonna worry about it. This is
2:00:14
the you don't have to address it next week.
2:00:17
This is from China. I'm gonna blow
2:00:19
one of my few times that I'm gonna talk.
2:00:21
Is it gonna be out an intellectual monologue?
2:00:23
Unfortunately, this time, not It's snoozy
2:00:26
booty's, the person typos.
2:00:29
No. No.
2:00:31
Snoozy booty's? No. And
2:00:34
that's it. My jars. Cool.
2:00:37
I appreciate your sacrifice, Luke.
2:00:40
This was from Charles. Lionel's recently newly
2:00:43
wed in my early thirties, wife only
2:00:45
into mobile legends for gaming as of
2:00:47
now. Was Yvonne into gaming when you
2:00:49
first were together and any tips slash
2:00:51
games to introduce my wife? She
2:00:53
wasn't really, like, she played casual games.
2:00:55
One that she apparently played lot was called dangerous
2:00:58
Dave. Yeah.
2:01:00
I'm glad she doesn't play dangerous Dave anymore.
2:01:05
You know what I mean? Davey's
2:01:08
dangerous, a lot of STD's and shit.
2:01:17
I would say the best way to
2:01:19
to get her into gaming, Yvonne, really
2:01:22
enjoys playing co op games. Games like
2:01:24
trying overcooked if you don't like your
2:01:26
marriage. She
2:01:28
played a lot of like leopard ed and team fortress
2:01:30
too. With us back in the day, it takes
2:01:33
you'll definitely have to you'll
2:01:35
you'll wanna find things that are
2:01:38
pretty forgiving for newcomers.
2:01:40
And I would strongly recommend, you know, if you
2:01:42
can, involve involve
2:01:45
her in like the squad. You know? Like,
2:01:47
instead of having a four man group,
2:01:50
grab two squad members and, like,
2:01:52
carry her. And, like, make it make
2:01:54
it fun, be supportive, just
2:01:58
make it fun. That's the most important thing. I think
2:02:00
that's what a lot of people miss, because it doesn't
2:02:02
matter what game you play. It
2:02:05
matters that you make it fun. And
2:02:07
make it an experience that she'll
2:02:09
look back on and be like, yeah, I wanna try that again.
2:02:14
Okay? God. One here from Tim. This
2:02:17
one's a little more interesting. L
2:02:19
and L at all. Love
2:02:21
the show and channel. Curious to hear your opinions and
2:02:23
opinions on private companies
2:02:25
being treated as infrastructure, e
2:02:28
g Google, YouTube, Facebook, and
2:02:30
should they be government regulated?
2:02:32
Also, what would happen if one of them just shut
2:02:34
down overnight? Cheers. Yeah. We're
2:02:36
in a dangerous place. I mean, that's
2:02:42
This is a problem that's been developing for a
2:02:44
long time. I mean, I I remember,
2:02:46
man, back in the day, I forget I
2:02:48
forget who it was, but I I watched a
2:02:50
really cool video about the
2:02:53
the way that US telcos
2:02:55
essentially took government money,
2:02:58
took public money, to build out a
2:03:00
ton of infrastructure, just completely
2:03:02
fucking didn't do it. And
2:03:05
then just kept
2:03:07
jacking up rates anyway for
2:03:09
customers like it's you
2:03:12
know, at that point, you go, well, hold on a second.
2:03:15
You took public funding. Are you a
2:03:17
private company? What does
2:03:19
that even mean? And and and to be clear,
2:03:21
I'm I don't think by private companies
2:03:23
you meant companies that are not publicly
2:03:25
traded. I think you meant not government
2:03:28
run entities, so any private or public
2:03:30
company. And
2:03:33
so so this this problem, this
2:03:35
this train has been heading towards us
2:03:37
for a long time. Yeah. If if Google
2:03:39
suddenly disappeared overnight, it
2:03:42
would be disastrous.
2:03:45
I think I think it's fair I think it's fair to
2:03:47
use the d word. If
2:03:49
yeah, it'd be disastrous. What
2:03:52
would the recourse be? III have no
2:03:54
idea. I could, you know, Can
2:03:59
can you tell them you you you send
2:04:01
government agents to, you know,
2:04:04
point a gun at them and tell them to turn the server back
2:04:06
on? Like, Fuck
2:04:08
you do. Right? Yeah. I
2:04:10
don't know. Silent
2:04:12
loop doesn't know either. Okay.
2:04:17
I'm sorry. I just don't have the answer to that one.
2:04:20
Certainly a complicated one. Oh,
2:04:24
no. Am I allowed to swear on this stream?
2:04:26
No. Okay. I didn't think so. Why
2:04:29
don't you read this one out then? If
2:04:32
you two are j and silent Bob, who at
2:04:34
Lenis Media Group are Dante and Randall?
2:04:40
What do you think, Luke? Who
2:04:44
would be our randal? No.
2:04:54
I don't think so. Yeah.
2:04:56
I don't know. I don't know. I don't think either
2:04:58
I don't think we are a day and Bob
2:05:01
anyway. So I cool question.
2:05:03
Love it. Useful thought
2:05:06
experiment, but I I just don't have an answer
2:05:08
for you. I'm sorry. I'm I'm I'm over two.
2:05:10
I'm over two. Too.
2:05:12
There's one consort if you want. I don't I don't give
2:05:14
a shit. That's fine. Go
2:05:16
on here from Patrick. Hi there. You mentioned that
2:05:18
you had put in place a policy
2:05:21
preventing personal items, SD
2:05:23
cards, and such, to be used at work.
2:05:25
What problems caused this to
2:05:27
oh, it's gone. What problems
2:05:30
caused this to exist?
2:05:32
How many items were lost? Oh, it wasn't
2:05:34
that items were lost. It was that that's just
2:05:36
stupid. Why are you bringing
2:05:38
your own stupid fucking lens to work
2:05:40
when you can just put in a procurement
2:05:42
request and we'll buy a lens and you can keep
2:05:44
lens at home, where it belongs, and where it's not gonna
2:05:47
get damaged. Why are you putting
2:05:49
wear and tear on your personal devices? I
2:05:52
don't understand. There
2:05:54
were times when we did need people to help
2:05:57
out with their personal devices, admittedly. We
2:05:59
did not have a lot of money early on. And
2:06:02
there have been times when procurement
2:06:04
has been slow. We didn't always have a procurement department.
2:06:07
But at this point, if you can't get
2:06:09
something procured, then
2:06:12
you need to, like, figure it
2:06:14
out. Talk to someone in operations or procurement.
2:06:19
Let's get her done. Because
2:06:21
clearly, if you're having a problem, then someone else
2:06:23
probably also having a problem and that's something that
2:06:25
we need to solve. Yeah. I just I just didn't
2:06:27
wanna be responsible for someone's, like,
2:06:30
expensive personal item getting
2:06:32
lost or stolen. Right? Like, that
2:06:34
sucks. Hopefully,
2:06:38
my orders will be in soon. This
2:06:41
one is from Braden. Hey, guys. I've never
2:06:43
been able to catch the show live,
2:06:45
but I'm at least able to send a merch message today.
2:06:48
I'm curious what you guys think about the
2:06:50
leak of the steam deck two.
2:06:52
How did you want valve to change slash progress
2:06:54
with handhelds? Have a great week. I
2:06:57
do not see about this. I
2:07:07
don't know. Andy,
2:07:10
Little Phoenix, said, there seems to
2:07:12
be not a lot of backing
2:07:15
for this. So I'm not gonna I don't think I'm really
2:07:18
gonna weigh in on this one. But
2:07:20
how do I want valve to change slash progress?
2:07:22
I mean I mean, the steam deck is honestly really
2:07:24
great. More more
2:07:26
more of that and more more better. I'd love to a little
2:07:28
bit more battery life. I'd love to see it quieter
2:07:30
personally. Tire
2:07:32
resolution display. Honestly, I'd rather
2:07:35
have an OLED than a higher resolution
2:07:37
LCD at this point if it's a if it's a
2:07:39
cost question or if there's a trade
2:07:41
off there. I think that's about
2:07:43
it for me. Okay.
2:07:46
The last one I have curated is from anonymous.
2:07:49
When does Luke would want to retire?
2:07:51
Well, Luke would love to answer that question for you,
2:07:53
but his jaw's little sara, if
2:07:56
you know what I mean. Everyone's
2:08:04
in a while. I gotta play the character. Okay?
2:08:07
I don't think Luke wants to retire. I think
2:08:09
Luke plans to work until he drops
2:08:11
dead I don't think you'd have it
2:08:14
any other way. There's
2:08:19
a few more. Joshua C. Asks
2:08:22
Luke feel better. I've been revamping my resume
2:08:24
site to apply. Does over engineering carry
2:08:26
much weight in consideration? I would
2:08:29
say probably not too much.
2:08:32
It's gonna come more down to,
2:08:34
you know, the interview and any kind of skill
2:08:37
testing process that Luke has
2:08:39
in place with his team. I'm gonna
2:08:41
go ahead and archive that. Steven says
2:08:43
if you guys had a button that
2:08:45
you press wait. What if you guys had a button,
2:08:47
you had press at the same time on your steamedex
2:08:49
to start and stop the show, kinda like
2:08:52
those silly lunch sequences. Oh,
2:08:54
you mean StreamDEX. Confusing.
2:08:57
Like those silly launch sequences, that
2:09:00
would be fun and completely not
2:09:02
worth me paying anyone to code. I
2:09:04
forbid anyone from the development
2:09:07
team who is watching this to work on feature like
2:09:09
that. That is not valuable. Ellius
2:09:14
asks, what is your laptop and external
2:09:16
GPU enclosure for when you travel?
2:09:18
And still have enough performance to play demanding
2:09:20
titles. What is your laptop in
2:09:22
extra for when you oh, what what do you use?
2:09:25
Oh, I'm still using the Oh,
2:09:29
the FlowX thirteen. Yeah.
2:09:31
That thing that thing is sick with the external, like,
2:09:33
PCIE by eight, like,
2:09:36
EGPU thing that's pretty sick.
2:09:39
I don't have to use that anymore now
2:09:41
that framework properly
2:09:43
support Thunderbolt on their latest laptop,
2:09:45
and I have upgraded mine to the to the latest
2:09:47
platform. But I just like already have all
2:09:49
my games on that one and stuff. So I just
2:09:51
throw it in my bag. I don't travel much anymore though,
2:09:53
so it doesn't really come up too much. AJ0
2:09:57
says, I think YouTube often gives false impression.
2:10:00
Minus Media Group is, of course, a
2:10:02
business. I'm not asking for names, but how
2:10:04
many people at LMG would you consider a
2:10:06
friend, like hanging out with after work
2:10:09
or something? Man.
2:10:11
It's tough. I mean, there's too many people
2:10:14
here that I could realistically say, I'm
2:10:16
friends with everyone. And
2:10:18
I'd say that there's almost certainly going
2:10:21
to be a
2:10:23
skew towards people who have been here
2:10:25
longer just because I have a lot more familiarity
2:10:27
with and we know familiarity breeds contempt.
2:10:29
I mean, closeness.
2:10:35
I don't hang out with almost anyone after
2:10:37
work, like just at all.
2:10:42
So tough. Also, the lines can be a little bit blurry
2:10:44
sometimes. Like, I'll
2:10:46
see people that Yvonne hangs out
2:10:48
with after work sometimes, so that doesn't mean
2:10:51
that Like, I would have invited them over.
2:10:53
Like, in some cases, that would have been super
2:10:55
weird. Like, if I invited,
2:10:59
you know, Yeah.
2:11:02
No. I don't wanna give like too much I don't wanna give
Podchaser is the ultimate destination for podcast data, search, and discovery. Learn More