16 CORE RYZEN CPU!! WAN Show May 19, 2017
Linus Tech Tips
·Linus Tech Tips
·2018-05-06
·
12,950 words · ~64 min read
0:13
We've been live for a bit. We were just, you know, we were having like a a
0:18
contemplative moment. Did you know that whole time?
0:22
Yeah. You just let that happen? Yeah. Yeah. I was I was watching you over here on the monitor. I was just
0:26
like, he's just so majestic.
0:31
So, we've got a we've got a great show for you guys today. It's actually been a
0:37
heck of a week. Um, I don't know what you've been doing, but uh I've been I've
0:42
been traveling to the far away land of Toronto. Oh, wow.
0:45
Canada. That's majestic.
0:49
All manner of majestic. Who were you who were you visiting with? Tell the people.
0:53
Dad Mau Senko. So, Senko.
0:56
Yeah. Did you call him that? No, I did not.
1:02
I went with Joel. I figured that was pretty sick. Okay, cool. Yeah.
1:06
All right, so we've got a lot of great uh topics for you guys today. AMD
1:10
announces Threadripper.
1:16
Also, pretty sick name. The chip formerly known as Naples with a
1:20
far more controversial name. What else we got? Uh, well, Destiny 2 is going to
1:26
be exclusively on Battle.net, which is
1:30
actually kind of interesting because I believe before they were saying Steam or
1:34
just open to everything. And W to Cry
1:38
has only collected $92,000 in ransom,
1:42
which like considering the scope they're on is really not very much. We'll talk
1:46
about that later. Wow. Crazy. Wild. Fantastic.
1:51
I just almost turned off the screen. stopped the whole thing. I'm actually like my brain is pretty
1:56
tired. Like I did I went 3 hours one way
2:00
and 3 hours the other way in like which is like not enough to
2:03
56 hours kind of thing. Like it's and like generally speaking I handle the
2:08
time zone shift pretty well. But I also
2:12
uh was up till 3:00 a.m. Eastern time.
2:16
No, Western time still. So like
2:20
No, don't stop the stream. Oh, no. I'm not going to stop the stream. Savage
2:23
Jerky. Thank you, Savage Jerky. Maybe you'll wake line us up.
2:27
Fresh books and also the tech carnival.
2:32
Hooray. Do we know how many tickets we sold for that? Uh, people ask me all the time, and I'm
2:36
like, I don't know, but I could I could find out. Um,
2:40
yeah, I could definitely find out. I'm going to ask
2:46
um I love that guy. He's awesome. All
2:50
right. Well, why don't we now he's not doing the Wand do. Okay. Do you want to Yeah. Yeah. Every
2:55
all is forgiven. Yeah. Okay. So, should I talk about my trip?
2:59
Do you want to talk about what you've been up to this week or should we get into actual talk about your trip?
3:03
Cuz I I want to know about your trip. So, I signed an NDA
3:07
the minute I walked into the studio. Really? Yes.
3:12
What? So, I can tell you that much.
3:15
What? Okay. I will tell you I will tell you things that I'm allowed to tell you
3:19
because like here's the thing is like what what are you even working on? I
3:22
guess you can't say his his studio is like his house. So
3:27
like Oh, so he doesn't want you to talk about like
3:30
So I like met people and like
3:34
So it was an NDA but like the people that were there and the stuff that was
3:38
there and Oh yeah. Like I thought it was an NDA but what you're
3:41
doing and I was like I sort of think I know what you're doing and it's not that. Oh, it's it's an NDA of like
3:46
making sure that you properly respect someone's privacy who is a public figure
3:52
and totally makes sense. So, there's a separate release that we're going to get
3:57
signed for everything that we're publishing from the piece that we did
4:02
and then like he gets to sign off on it. But like basically what I can say is
4:07
that uh Joel uh Zimmerman aka Dead
4:11
Mouse, super chill guy. Yeah, we were supposed to be there from 1
4:16
until like the end of the workday. Um,
4:20
we didn't leave until the wee hours of the morning just like shooting the
4:27
and uh working on working on our video and just like
4:31
that's actually really cool. the guy is. So there's there's a huge difference
4:35
between people that are famous for like
4:40
um for for like being famous I guess
4:44
like socialites and people who are people who are famous
4:47
for just like really really caring deeply about their craft and the guy's a
4:53
gigantic geek like such a nerd. I've
4:57
thought about getting his master class, not because I want to learn how to
5:02
actually do that stuff cuz I would be horrible at it. We both know this. Um,
5:06
but I think it would just be interesting to watch, because I like watching people
5:09
that are that passionate about something just talk about it. Yep. And so, and it's it's what's cool
5:13
is that no matter how basic of a thing we were talking about, because for me,
5:18
electronic music is like explain like I'm five pretty much. I don't know much
5:23
about the scene. I don't know much about the artists. I I had I confessed it to
5:27
him. I was just like, "So, as part of my
5:32
research and due diligence for this trip, I put you into Google Play Music
5:37
and like listened to some of your music." I legitimately have listened to a lot of
5:40
his stuff, like probably everything. And the funny thing about it is he was just
5:44
crazy chill about even even like like for me that's pretty embarrassing but
5:49
it's less embarrassing for me to admit it up front right
5:52
than for him to be like you know something something something
5:58
do you think of this song and me being like oh is it one of yours
6:01
and it's like ghost and stuff or something
6:04
like okay good reference there you go because Google Play Music conveniently sorts by
6:09
popular so like so when when and stuff.
6:12
Nope. So, when um when we were looking at some stuff that
6:16
I can't talk about, but that is really cool and it's going to be in the video,
6:20
like I I recognized things that was
6:23
good. When we used to have lands, one song that we could play like on the speakers
6:28
because my parents and my friends and me all liked it was a DeadMouse song and it
6:33
was Strobe. Okay. I don't know that one. But like like it's it's just kind of funny that
6:37
like out of all the music between all the different genres that all the people
6:41
liked at these like really big lands with lots of people, it ended up being
6:44
some Dead Mouse songs, right, that were the easiest to play.
6:48
So So yeah. No, it was really fun. He's super cool. We uh spent the entire day
6:54
there. Um and basically it's going to be like a studio tour, but not like most
7:01
people would go in and do a studio tour with Dead Mouse. In fact, in some ways,
7:06
I think me not being a rabbid fanboy
7:09
probably helped us approach his studio a
7:12
little bit differently because I think, you know, what would a typical person
7:17
do? They'd walk in, they'd be like, "Can you wear the hat?"
7:21
Right? That's kind of lame. And so, and I would actually specifically want him
7:25
to not put the hat on. So, instead, you know, we're looking at
7:30
ah I like I don't know what I can say and what I can't say. Like, some G
7:34
series SS. I'm sure he'd be cool about it, but like we're looking at the tech behind
7:40
how he does things. Um, he even does like a short So,
7:44
because I was like, "Okay, explain like I'm completely stupid." Uh, how
7:48
electronic music works. And he actually does like like a like a kindergarten
7:53
level like demo, live demo. Super cool. And
7:59
just it'll be it'll be great for anyone like me. I am personally really stoked to watch.
8:04
electronic music is like that kind of
8:08
like it's it's that but what does that mean and and we get this very basic
8:12
explanation of it. It's really awesome. That's cool. Um so what were you up to?
8:15
I was grading projects. I was trying to
8:18
finalize contracts with people about things. Um I was modernizing something
8:24
that I'm not going to mention. Uh what
8:27
the heck else did I do? meeting with multiple accountants about weird really
8:33
annoying stuff. Um, one of which like
8:37
wants me to This is going to go a little bit too deep into personal stuff.
8:40
Shoot. We called the stupid stream afterparty. Ah, dang it. Carry on. Uh,
8:46
one of them, no, I told both of them that I'm a US and Canadian citizen, and
8:50
one of them like very aggressively
8:53
wanted me to move to the States, renounce my Canadian citizenship, and
8:57
get convince you guys that Floatplane
9:01
should move to the States and all this kind of stuff. What? What? And like had a giant picture of Trump on
9:05
his wall. It was a really weird meeting. Um, yeah, it was a lot of weird stuff
9:11
happened this week. That's probably how I'd frame that to be completely honest.
9:19
But yeah, there you go. Okay. Um, sure. Anyway,
9:24
tried out Okay, I actually tried out a really wicked keyboard. Okay.
9:27
Uh, it's a pre-release keyboard from Wuing and we're making a video on it, but it's
9:32
the optical key switches where it's like an analog gradient as you go down, so
9:37
you can do like steering wheel driving. We finally got one of those in. I think
9:40
we requested one like ages ago, forever ago, and they said they'd they
9:44
didn't want to send me an early one cuz they were like, "We want to send you one when it's done so you can see it when
9:48
it's ready." And it's still an early one because I asked forever ago. I think
9:53
around like Computex last year and Yeah.
9:57
So, whatever. It's Wuing like like Wing,
10:01
not Wuang. Did I say Wuang? No, you didn't. Yeah. Okay.
10:05
But Twitch chat, it's like Wuang. No.
10:09
No, it's not a Wu Tang keyboard.
10:13
Get out of here.
10:16
Oh, I was making the wings behind this. Yeah. No, he can he can he can he can
10:20
go. He's dismissed. All right. Yeah, let's So, this was
10:24
originally posted on the forum by Doc Swag.
10:28
I have more to talk about with Floatplane, but I was thinking of lining it up after
10:33
or not. Well, I also I want to line it up for when we do the little bit of a teaser
10:37
thing. Okay. Sure. Okay. All right, AMD.
10:41
Oh, what? Oh. Oh, did we move the we
10:45
moved the HDMI duads around? Hold on. What?
10:48
I'll be right back. Okay. Uh, should I talk about it? Yeah. Yeah. Yeah.
10:52
AMD will be introducing stuff called Threadripper. Ryzen Threadripper. The
10:56
they're AMD. Sorry. They will have 16
10:59
cores and 32 threads. That's why they're called the Threadripper, which is just
11:03
absolutely ridiculous. I believe there's a couple different Yeah, there will be
11:08
up to nine different Thread Rippers. These are rumors. The 1998X
11:15
and 1998 with 16 cores. The 1977X and
11:19
1977 with 14 cores. 1976X and 1956X
11:25
and 1956 all have 12 cores. And then the
11:29
1955X and 1955 with 10 cores. Why are
11:34
they making this so complicated? It's a little messy. Okay, it's really
11:38
messy. Um, but having it be AM4 and
11:42
being able to get 16 cores is pretty sick.
11:47
So, uh, this is all this is all Oh, so
11:51
these are claims from WCCF tech, just to be clear.
11:54
Um, wait, hold on. And they're also
11:58
saying that these Threadripper CPUs won't be compatible with the AM4 socket,
12:02
which sports 1331 pins, uh, but will demand a massive
12:07
4,094 pins. Does it have a picture of it? Was that
12:10
the Epic thing that was epic? Um, although it wouldn't
12:14
surprise me, it wouldn't surprise me if Threadripper
12:18
is um will look the same as Epic and uh
12:22
fit in the same socket. So much like what Intel does with their high-end
12:26
desktop, so their 2011 V3s and their
12:30
Xeon platform, which also fits in socket 2011 V3. I'm expecting AMD to have their
12:36
epic line of server chips and their Threadripper line of high-end desktop
12:42
and/or workstation chips exist on the same socket. So that'll allow them to
12:46
have more PCIe lanes, uh more pins for interfacing with more cores, uh up to I
12:52
believe it is six channel memory.
12:56
I'll try to find that. I read that wrong and thought it was on a different socket. Sorry guys.
12:59
Um six I can't remember. Is there eight
13:03
channel memory or something stupid like that? I can't remember. I think it's eight channel memory actually now that I
13:07
think about it. Um, and then the other
13:10
the other thing is that uh what's so
13:14
what's okay so what's interesting about this to me is that the way that Intel
13:17
differentiates their Xeon processors on LGA 20113 and their uh their Core i7
13:24
processors on LG 20113 is a couple of things and the main ones are dual socket
13:30
support. So I would expect that that
13:33
will continue to be a differentiator for AMD. But the other thing that Intel does
13:38
to keep them separate from each other is support for ECC registered memory on
13:43
Xeon. 16 channel DDR4 memory. 16 channel. I think that's 16 channels
13:49
to the dual uh wait 16 channel. Wow. Okay. Well
13:55
maybe uh maybe Twitch chat is epic.
13:58
So that's the epic. So for 30 if you want 32 dims.
14:04
So yeah. Um but if the Epic is on the same platform.
14:09
Yeah. I guess that that could be another way that they differentiate them,
14:13
right? So yeah. So rumor rumor there will be up
14:16
to nine Thread Rippers. I really I kind of hope they just call them Ryzen 9 to
14:21
make life a little bit easier. And I I I
14:24
really am not sure what I think of these convoluted kind of names for them.
14:29
The 1998X and 1998.
14:33
Um, tell me this. 16 cores
14:38
on a desktop. What's the value here
14:43
if you want to run like four
14:47
Lime Tech systems on one system? So,
14:51
I'll let you in on a little secret. I've chatted with the Lime Tech boys and u so
14:56
far virtualization on Ryzen is not a great
15:00
experience, but Okay, go ahead. Carry on. Okay. Um um
15:06
streaming to a lot of places at one time
15:10
at a lot of different like output resolutions. Sure.
15:14
And for some reason, you don't have like a separate box handling that.
15:17
Yeah. Okay. All right. What else you got?
15:20
Not much. Cinebench. Sure.
15:24
Amazing Cinebench. The best Cinebench. Um, so we actually
15:28
did a demo a little while ago. 3D rendering.
15:32
Yeah. Yeah. Okay. So, so like workstation.
15:35
Workstation. But he wasn't saying workstation. Yeah. Cuz I like I'm I'm talking like
15:40
competition with high-end desktop. Like to me, I'm kind of looking at Thread
15:44
Ripper going this is seems like a very niche product right now because Ryzen 7
15:51
already goes all the way up to eight quite high performance cores which for
15:56
desktop workloads like it's about like um like it's about like talking about
16:00
how your how high performance your SSD is when it requires Q depths of 16 in
16:07
order to even you know differentiate itself from a low-end SSD. Like that's
16:11
what we're talking about. We're talking about a product with all this theoretical power, but that a desktop
16:17
workload um pretty much wouldn't touch.
16:20
And to be clear, I am not complaining.
16:23
The fact that AMD is coming in and putting pressure on Intel about core
16:30
counts is great. But right now, I think the sweet spot is just six cores. So, a
16:36
lot of people in chat are naming a lot of things that you could do with it that aren't the scenarios that Linus is
16:41
talking about. A lot of what people are listing are workstation stuff. I don't remember all work.
16:45
One of them was like SETI at home and stuff. That's not a
16:48
which is like that's not necessarily workstation. No, but it's not really uh like it's a
16:54
it's it's it's more like a power virus. It's less like a an actual thing that
16:58
you're sitting at your computer doing. Yeah. To be clear, it has a value, but
17:03
most people remember we're talking about people saying like cryptography and
17:07
stuff. Who's actually doing that at home? And like you're probably doing with a
17:11
GPU and GPUs are a thing.
17:14
Yeah. Um so another guy said Bitcoin mining, but like again, yeah, I go with a graphics
17:19
card. Yeah. Or uh don't they have like a
17:23
video rendering? I'm I'm out of date on the whole Bitcoin thing.
17:26
Jake's trying to convince me to do like another Bitcoin. Still, you still kind
17:31
of just want to use your video card, though. Yeah. Oh, is it still just video cards?
17:34
Okay. Yeah, I was just like I think I like I could be
17:39
misinformed. So, basically what I what I
17:42
see as being positive about this is that it's putting pressure on Intel to
17:48
finally give us more than four cores. And it's putting uh it's not apparently.
17:53
Apparently, I'm wrong. Oh, GPUs are not a thing.
17:57
Well, no, they still are.
18:01
But
18:05
yeah, I'm not sure. Um, okay. Now, let's talk about the chip
18:09
formerly known as Naples. So,
18:14
here's what I don't get. Why is it that
18:17
companies that have great code names
18:20
like Zen Architecture and
18:25
Naples product code name, which is like
18:28
inoffensive, unoffensive? It's not offensive. What's offensive about Epic?
18:34
It's stupid. Yeah. Okay. It's like It's like you're you're
18:39
marketing this to like data center professionals.
18:43
Why are you calling it epic? Why don't you just call it elite while you're at
18:48
it?
18:51
Yeah. Yeah. Yeah. You said you said
18:54
inoffensive, though. Okay. No, epic isn't offensive.
18:58
It's not offensive. It's like um it's
19:02
just a little silly. I mean, to be clear, it's not like I
19:07
necessarily have a better answer. And
19:11
like it's possible they would have wanted to call it Zen server or
19:16
something like I don't know what they wanted to call it but there's there's
19:19
always complications around you trademark infringement and copyright and
19:23
all that kind of stuff. So you know maybe that was the only thing left. Um
19:28
but unless it like stood for something really cool then I I don't really I
19:35
don't really get this as like a um as an enterprisegrade product.
19:40
Now, with that said, we should we got to straw pull this because I was going back
19:44
and forth with another with another journalist this morning actually on in
19:48
Twitter DMs and uh we were kind of debating debating this whole this whole
19:53
thing. You know what? Okay, I'm not going to say what side
19:57
anybody was on, but it was uh Ian from Antech. We were chatting about it and
20:02
the other side of the argument that got brought up was, "Okay, well, hold on a
20:06
minute. What if your typical like data
20:11
center, you know, purchaser,
20:14
you know, is goes home and watches anime and like
20:19
plays video games and like thinks that
20:22
things are epic and you know, like is it
20:25
really that far off? I don't think it would be that far off.
20:29
Okay, so let's let's let's just straw pull this epic. Um,
20:35
you know what? At the same time, personally, if I was purchasing this, I
20:40
would worry about this being on an invoice that I would be handing to
20:43
someone. Okay. Okay. So, cuz I would be like,
20:47
maybe they just think I'm buying it cuz it's cool. It doesn't sound like
20:51
it sounds flashy and it sounds like something in Toys R Us. It doesn't sound
20:55
like serious infrastructure hardware. You know what I mean?
20:59
Yeah. But, you know, with that said, it's not like some companies haven't had
21:04
a ton of success with that. Oh, definitely. The red cameras,
21:08
the red dragon, the red epic, while we're at it, the red weapon.
21:12
Yeah. But then, like, at least they spelled it right.
21:16
They spelled it. Like, I feel I feel like But seriously, I feel like you'd be more positive about
21:20
this if they put an I there instead of a Y. Like, would you like it more if it
21:23
was spelled right? Cuz a big part of your defense this
21:27
whole time is that it's been silly. I'm going to be honest. No, I wouldn't like
21:31
it better. Interesting. Now that they've gone down the path of
21:35
Ryzen being R Y.
21:39
Okay. So, every the second letter just always has Nope. That's not even the
21:42
second one. The eyes. K. R Y. And like
21:46
so lame. It it Well, yes, but
21:50
I hate that. But if you're good, I hear what you're saying. Be consistent. I I got you. I got you. I
21:54
just hate that. Okay. So, so far most professional Yeah. are saying
21:59
I'm on and that makes sense because I'm on the unprofessional to neutral side of
22:02
things. I don't really care. But I would a part of me would be like gh like I
22:06
might have to defend this to someone cuz they're going to be like why did you buy something called an epic?
22:10
Why are you buying epic processors like like we just need some infrastructure
22:14
stuff. What are you doing? And I'd have to be like gh okay like uh
22:17
epic is really good cuz it has a lot of threads and it's cheaper and it has a
22:21
lot of memory bandwidth and then by then their eyes glaze over and hopefully they
22:25
just sign off on PO. Yeah. Yeah. It's like, okay, everything's fine. But like, that was an
22:29
unnecessary amount of communication that had to happen. Yeah. And you know, like the stupid
22:34
thing is like we're probably going to we're probably
22:38
we're probably going to all like agree that it doesn't matter anymore in a
22:42
number of years. But some someone named Lionus Rockin Mandals
22:48
in the chat. Um I don't know what's going on there, but said no because
22:53
Intel owns the I letter for processors and like obviously like they don't own
22:58
it. But that's an interesting point. They might be trying to stay away from the I in anything.
23:03
There are lots of words that don't have I's in them. Yeah. Like blind.
23:10
What? I'm genuinely confused.
23:17
No eyes. Oh. Oh.
23:21
Oh, you can have eyes and be blind.
23:25
Twitch is booing so hard. It's just like boo.
23:28
But that doesn't even make sense cuz you can be blind with eyes still. You don't
23:32
have to like have your eyes have been removed to be blind.
23:36
Oh, I'm terrible. Ah, I should Yeah, I should just end it all now. Um
23:40
Oh my god. Okay, so at any rate,
23:43
that was brutal. Let's talk Let's talk about the chip formerly known as Naples. Okay, I was
23:48
right. 8 channel DDR4 16 channel DDR4.
23:52
What are these articles? So, single for EP,
23:56
now known as Epic. Single socket processor scales up to 32
24:01
cores. And as far as I know, there will be dual socket variants as well.
24:05
Oh, that's fancy. are we getting are we getting some I have I am in touch they want to know
24:11
so so basically what um what uh oh I
24:14
don't know how much of this I can say um
24:17
okay what I can say I can talk generally about server hardware generally speaking
24:22
if someone's going to seed you server hardware they're going to be like what
24:26
are you going to do with this because if my answer is we're going to run in a
24:31
bench they're going to be like cool go buy one
24:36
Um, so if we come up with, for me, I want to come up with a really cool
24:40
project. So I want a couple of 32 core
24:43
epics for a total of 64 cores and I want
24:47
to figure out something completely bananas that we can do with it. So
24:51
assuming that this What about speed transcoding YP9 8K
24:55
footage? Sure, dog. Like you So there's that.
25:00
That would be really cool. I had another idea. That would be fun. Um is is that like
25:05
something that you so does does that would be wonderful scale
25:08
with more CPU cores basically infinitely? Uh there's we're I still talking to
25:13
boiler about it. Boiler wants to do this thing which completely makes sense and
25:17
I'm totally on board but we can't do it right now because we don't have the infrastructure right now but it's like
25:21
distributed transcoding where you like slice up the video and then transcode it
25:24
in a bunch of different places on a bunch of different servers with a bunch of different cores. But if we had a 64 core machine would
25:29
that be unnecessary? Maybe do it in one place, right?
25:32
I don't know. It's still I was talking to Boiler. Apparently, it's like 30
25:35
times harder. Yeah, because key frames. But then if you just have like a
25:39
monstrous computer then
25:42
maybe low. Yeah. Okay. Okay. So, I'd be totally down for
25:46
that. And then one of the ideas that I wanted to look at is if they've got if
25:50
they've got virtualization sorted out a little bit more, and to be specific, I'm
25:54
talking about uh KVM. So, Red Hat KVM because that's what Unrade uses. I think
25:59
it would be amazing to get my hands on
26:02
like a a multi-GPU barebone server much
26:05
like what we did for uh 10 gamers one CPU
26:10
is just like yes exclamation mark and what get this what I'm thinking is
26:15
the entire editing den running off of
26:19
one tower six editing six 10 core
26:23
editing workstations off of a single tower Nice.
26:29
So, that's kind of something that I have in mind for what I'd like to do with it.
26:32
So, I'm going to I'm going to try and come up with some cool projects, but um
26:37
yeah, basically it's this this thing is
26:41
unbelievable. So, 8 core memory up to 16 dims per CPU. Um
26:47
they've got uh Epic provides up to 128
26:51
PCI Express 3.0 no lanes allowing you to connect more GPUs directly to the CPU
26:57
and that's really important for machine learning type applications. We've got
27:01
our unboxing of SFUS the fastest
27:04
supercomput in Canada. So I I went down there and they've got they've got a ton
27:09
of GPU nodes and they would love to have
27:13
the GPUs connected directly to the PCIe lanes on the CPU. In fact, we were
27:17
talking about that for it's in my notes uh but I don't remember exactly what
27:22
workload specifically it was where intergpU communication is incredibly
27:28
important and the more hubs that you have to pass through the worse. And then
27:31
I I even talked to it's so cool talking to like these highle architect guys. I'm
27:35
like, "Okay, well, what about Envy Link?" And he's like, "Yeah, the cost
27:38
just wasn't right for what we're doing." And it's it's all these kinds of
27:42
tradeoffs where normally as a consumer, you hear about this crap and you've got
27:46
NVIDIA on a stage with freaking wood screws and whatever it is that they're
27:50
talking about going, "This is the future of blah blah blah." But then, you know,
27:55
to us that this is all totally unattainable stuff. Whereas you're
27:59
talking about people who are working on data centers that are in the value of 10
28:04
to 20 million where they are actually
28:08
taking like a $2 million solution and a
28:11
$3 million solution and evaluating it the same way that we would like two
28:15
different laptops. Yeah. But that they're just working on a
28:19
different scale. It's all the same conversations. It's like do I want to
28:23
pay this much more for this much more performance increase? and they just for
28:27
their design it didn't end up making sense and uh and it was going to add
28:30
some cost and I don't know the exact numbers so there's no like ND8
28:34
information that I could hope to give you I just know that they didn't use it and they opted for these uh they opted
28:40
for certain GPU nodes some of the GPU nodes for certain workloads are
28:45
connected directly to the CPUs and some of them are not
28:49
depending on what they're needed for and these ones are higher cost I was just like oh cool
28:53
that's pretty fancy you guys are so baller I love
28:56
Um, so Epic is expected to launch before
28:59
the end of June. You can tell AMD is in full-on crisis mode about their stock uh
29:05
dropping what was it two weeks ago. Oh my goodness, we have so much stuff
29:09
coming. We have stuff. Please just wait. We have
29:13
so much stuff. Um, even Radeon Technologies Group was
29:17
showing off stuff. So, this was originally posted by PC Gamer 324 on the
29:21
forum. The original article here is from
29:24
uh oh three. Did I say three? Okay. No,
29:28
digital trends. Here we go. AMD's Radeon Vega Frontier Edition is
29:33
the most powerful graphics card yet according to AMD.
29:40
All right. So, the numbers put forth by AMD are as
29:46
follows. It'll have 64 next generation
29:50
compute units. So that's 13 teraflops of peak single precision 32-bit compute
29:56
performance, which should put it in the neighborhood of NVIDIA's top cards. I
30:00
think their top card is 12. Uh, and that's a Quadro
30:05
uh P6000. Don't quote me on that, but I'm trying
30:10
to look this up as quickly as I can. Uh,
30:13
Quadro P6000. Yeah, 12 teraflops for
30:17
their top-of-the-line Quadro, which is about a $5,000 graphics card. What else
30:23
do we know about it? It I wasn't sure about all the stuff you
30:27
said cuz I was checking in with something. Uh 16 gigs of HBM2 memory. So, that's
30:31
pretty impressive on a super wide 2048 bit bus.
30:34
Have you Have you shown them the Shroud? Uh the Shroud looks freaking sick.
30:39
Yeah, AMD finally gets it on the Shroud.
30:43
Although, if it's like plastic and kind of cheesy looking, then I really don't know if that's going to be that
30:48
impressive. To be fair, the last one that they came out that was plastic and cheesy looking,
30:51
once it was in a case, it was actually a little hard to tell,
30:56
right? Okay. Fair. Still, but like not amazing, but it wasn't that bad. Um, so they're saying
31:01
that Vega's memory architecture can access terabytes of memory. And we've
31:07
actually seen them kind of move beyond
31:10
the like the memory that's like HBM2 or GDDR
31:16
memory that is soldered to the board. And uh what was that uh what was that
31:20
professional card that was using SSDs? Like it was using NAND in order to have
31:25
I I forget how much memory it was on it. I'm sorry you guys. I wish I had like a
31:30
more exact uh set of talking points for this, but they were showing it off at
31:35
CES and uh the idea was that they were using NAND flash in order to
31:39
dramatically increase the amount of of the basically the size of the project
31:44
that you could work on because it's still faster since you've got a nice
31:48
lightning fast interface than trying to go out to like a scratch disc or
31:52
something. So, apparently AMD is claiming that they've done several
31:55
benchmarks with it. And the Vega Frontier Edition is apparently 70%
32:01
faster in Solid Works.
32:04
Okay, which is like pretty insane. Another program that I haven't personally heard
32:09
of called Katya, I'm hoping I'm saying that correctly, is it's 27% faster in
32:16
Okay, which is pretty wild. I mean, AMD has
32:20
been known from time to time.
32:25
Sorry, I didn't mean to get an ice cube in my mouth and now,
32:29
okay, it's gone now. So, AMD has been known to cherrypick benchmarks from time
32:33
to time. Um,
32:36
but still, it's promising and like that happened last time. You
32:41
just reorganized everything. Control Z that. um that happened last time with
32:44
their CPUs to a certain degree, but then
32:47
a big part of what they're claiming did end up being true. That's right. And what we do know based
32:52
on that it's not getting completely stomped on is that we're going to have
32:57
competition. Yeah. At least until NVIDIA trickles down Volta
33:01
and like Yeah, fair. But uh the like
33:04
with their CPU, what they were claiming did completely end up being true. Yeah.
33:08
So if you're into Solid Works, there's one really cool thing that just
33:11
happened. Yeah. But again, it depends on how you're benchmarking that specifically,
33:16
what kind of projects you're working on, what you have in Solid Works at that time.
33:19
This is basically the AMD show at this point. Laptops are coming. Ryzen Mobile,
33:24
Ryzen Pro, the original article is Trusted Reviews. Let's go ahead and pull
33:28
that up. Don't worry, we'll get through this eventually. Ryzen finally comes to
33:31
laptops. This I'm actually excited about. I would love to have like a six
33:36
or an 8 core laptop. That would be awesome. I don't know if they're going
33:40
to be able to get the power low enough and I haven't actually looked that
33:43
closely at this yet, but like six core laptop would be freaking awesome. AMD is
33:48
claiming 50% more CPU performance than its previous 7th generation APUs. Lordy,
33:54
I would hope so. Yeah, and a lot of these claims are a little
33:57
whimsical, I think, cuz they're from kind of, you know, older stuff. So, 40%
34:02
better graphics performance and a 50% cut in power, but you're jumping quite a
34:06
few categories there. quite a few generations um of improvement that
34:10
Intel's been making. So, they might not be leaprogging Intel, but I guess the
34:14
promise here is that they're going to be competitive. And then they also
34:18
announced Ryzen Pro, a version designed for business PCs.
34:24
Wow. So much actually kind of weird. Like I I get it, Pro Professional, but
34:28
for a long time Pro has been like a gaming thing. Yeah. Like higher just like higher
34:32
performance. Yeah. So that's actually a little bit weird.
34:36
Um, but oh well. Like I feel like Ryzen's
34:40
smart or like even just Ryzen business
34:44
would have been cool. So come back up a little bit.
34:48
Did it help them to kind of announce all this stuff?
34:52
Kind of looks like it. Um, I mean they haven't quite recovered from
34:57
uh this was when they did their uh their quarterly earnings report, right? Yeah.
35:01
Okay. Well, yeah. We'll see how that goes. I mean, I don't know what here's the
35:05
thing. I don't know what people were expecting
35:09
from that quarterly earnings report. Like, if I That's fairly accurate to what they
35:13
guessed. If I invested in tech companies, I I
35:17
would actually love to know what our viewers think. Like, should I be allowed to invest in tech companies? Cuz I I'd
35:22
love to I'm allowed to, but I have to disclose
35:25
it. Okay. Yeah. But like, it's it's a matter of
35:28
whether they will allow me to. I I would have I would have probably
35:33
bought some AMD right after that drop because it was the drop was based on
35:38
people having completely unrealistic an unrealistic reaction to what AMD has
35:44
been doing over the last quarter. It's like Ryzen 7 and even Ryzen 5 are the
35:50
tip of the iceberg, you guys. That's like the enthusiast gamer chip and then
35:55
the like slightly less enthusiast gamer chip. Where they're going to move volume
35:59
is Ryzen 3 and where they're going to make margin and start to be profitable
36:03
is going to be products like Threadripper and Epic. So I yeah I don't know
36:09
what people were asking for there. Um,
36:12
not only that, but you know, any new product is going to take some time to
36:16
ramp up in terms of yields, in terms of having stock on the shelves, in terms of
36:21
uh the platform maturing a little bit. I mean, I don't know if you guys noticed,
36:24
but you didn't hear from us, run out and
36:28
buy Ryzen in the first 6 weeks because
36:32
it was it it was promising. And we were
36:35
confident that AMD and their partners like ASUS and Gigabyte and MSI, we were
36:39
confident they were going to get all this stuff addressed,
36:43
but it wasn't yet. And when you buy a
36:46
computer, it needs to work on the day you buy it. So, I mean, a lot of people
36:50
criticized us for doing a Core i5 build guide uh shortly after the launch of
36:55
Ryzen, but the reality of it was when we were filming that, we weren't ready to
36:59
say go buy a Ryzen PC yet. Now to be
37:03
clear, it has improved a lot like a lot.
37:07
So I am expecting over the next quarter or two these kinds of results to change
37:12
in a big way. Sorry. Oh, no problem. Uh oh, they want
37:19
me to do a straw poll. Okay, you know what? I I'm going to I'm going to do a
37:23
straw poll whether I should be allowed to invest in tech companies.
37:26
Oh,
37:29
see like Yeah, I don't know. I I don't know.
37:34
So, to be clear, I actually Oh, this is something I've been meaning to talk
37:38
about on W show for quite some time. Um,
37:41
we are working on like a private deal
37:44
with a much much smaller tech company. So, maybe I'll talk about it after our
37:49
after our ad break. But, um,
37:52
what is actually bad about that? Well, it creates a potential conflict of
37:57
interest. Like, if I own Oh, yeah. Then then yeah,
38:00
if I own a bunch of NVIDIA stock as a viewer I cuz I thought I thought we
38:04
were looking at the like insider trading side of things. No, I don't think that's an issue. Um
38:09
because I can just not buy when I know anything. And that's
38:15
pretty easy to do because I think you know if you were to just go, okay, the
38:20
embargo's up 3 minutes later now that everyone knows all the same things I
38:24
know go by. Go by. I mean, you can you could kind of
38:27
beat the rush on that kind of thing if you wanted to. Is it an advantage? Yes.
38:31
But is it against the law? As far as my understanding goes, no, it wouldn't be,
38:36
right? So, as a viewer, I would say no. You would say no. Go ahead and cast your
38:41
vote. Um cuz so for me, where I where
38:46
I've kind of drawn the line already because like I said, we are working on
38:49
an investment deal with a really really small company. Um, so where I kind of
38:54
draw the line is that if it's stuff that we cover um
39:00
competiti like uh the competitors of like if it's stuff where it's kind of
39:04
one option or the other and where I'm in a position to inflate the appearance of
39:10
let's say Intel versus So would you be declaring in every single video that you
39:16
cover this one mono company on? I'm saying mono because you're not covering
39:21
other ones in its space that you have ownership.
39:24
No, I think because to a certain degree that makes every single video an ad.
39:28
Yeah, I think the disclosure uh regulations are such that you talk about
39:33
it only in the context of talking about
39:36
that company. So let's say I owned a bunch of shares of Intel. If we are
39:41
going to talk about an Intel news story, I need to have a thing up that says um I
39:48
disclaimer I am long Intel or something like that. Like you have to clarify your
39:53
your position. You say I am long Intel. Yeah, long Intel. So So I am I I have a
39:58
long-term investment in Intel. Whereas you would say I'm short Intel. If you
40:03
think Intel's going to have a big quarter and you want to kind of put in
40:07
some money, grab it, back out. Yeah. So, long or short? Yeah.
40:11
Um, and I I think medium, but I'm not sure. Like, I'm not an expert on this
40:15
stuff. I've only actually bought stock like twice.
40:18
Um, so, so there you have it. Uh, we should
40:22
probably go through our sponsors for the show. Speaking of, uh, you know,
40:26
disclosures, I guess that's not really a disclosure. Fresh Books.
40:30
Hooray. So, if you guys are running a small business, let's say you are, come up
40:35
with a small business we've never talked about before. Come on. trying to
40:38
distribute videos on the internet. No, not like that. That's not that's not
40:42
trying to uh uh
40:47
terrible detail cars. Yes. Okay. So, if you have a small like
40:52
shop out of your garage where you work or even a mobile one cuz you could
40:55
accept payments through your phone or let's say you're a locksmith. Okay.
40:59
So, you got a and then we'll go with you know the usual one that's more likely in
41:03
our audience. Let's say you choose do smalltime tech repair. If you run a
41:07
small business, FreshBooks is the way for you to take your accounting solution
41:11
with you on the go. It's got all the same functionality in the mobile app
41:14
that it does in the desktop version and keep track of everything from expenses
41:19
to hours worked on a given job to your invoicing. You can send professionallook
41:24
invoices in a matter of minutes. And you
41:28
can accept payment through the FreshBooks platform on your own terms.
41:31
So you can take deposits and then take the rest later or you can take
41:35
everything after the fact and you can even see when your client has seen the
41:38
invoice so you know that uh well so you
41:42
know that they saw it. So if this sounds pretty good actually oh and I forgot
41:46
right they've got uh they've got free tech support when you call FreshBooks
41:50
you talk directly to a human being and they walk you through what you're having
41:54
trouble with. No robot portals which are insanely
41:59
frustrating. I had to deal with one, not going to mention who, but it was really
42:02
brutal. It took a long time and it was really annoying. I had to call my credit card and I ended
42:06
up like five or six tears in. We're
42:10
sorry. We're now closed. I'm like, so
42:13
basically, you just wasted like 9 minutes of my time. Oh, that's brutal.
42:17
And I can't even talk to anybody about why this transaction didn't go through.
42:20
Perfect. Thank you. Uh, anyway, not with
42:24
FreshBooks. So head over to freshbooks.com/win
42:27
and you can try it out for free. Tool around with the software. It's super
42:31
easy to use and see if it's right for you. When you are done deciding it's
42:36
right for you, go to freshbooks.com/teips. Wait.
42:41
Okay. Well, I guess it doesn't really matter. And enter the when show in the how do you hear about us section.
42:47
All right. Ah, yes.
42:51
It's nice. Savage jerky time. Yay! I get Reaper
42:56
again. Well, you don't have to eat Reaper. You could have the uh
43:00
You could have the ghost pepper buffalo sauce. I'm going to do ghost pepper
43:03
buffalo sauce today. Yeah, I'm doing it. Which one's hotter? I don't remember.
43:06
Is that six or five? Take that one. No, take that one. No, not taking the reaper. I don't I
43:11
don't want to do the reaper. This one's your favorite one, right? Uh no, no. All of my uh I the maple the
43:17
maple bacon one. Maple bacon. Yeah, the maple bacon one. I just ate it
43:20
all after the show. I really like the Mojo habanero. That's
43:25
my favorite one. Yeah. So, good amount of spice. You can tell it's like the second or
43:31
third Savage jerky spot when it's like the stuff that we absolutely cannot
43:36
resist is just all gone.
43:39
Okay. Well, this is my favorite one, but he hid it from me,
43:44
so I couldn't have any more of it.
43:52
Are you good so far? You had the ghost pepper.
43:56
So, you didn't have quite the hottest one. This one's actually still sealed.
44:03
The ghost pepper and buffalo sauce
44:06
is hot enough for me. But basically,
44:09
you want some ghost pepper? Jake does really good with spice.
44:13
I would probably throw it. He's he's he's very spice tolerant. All
44:18
right. So, Savage Jerky.
44:21
See that one? Where? Where'd it go? Mojo habanero gives you a bit of a kick.
44:26
Mhm. But it's not like like it's not doing that to me. Like I feel good. It woke me
44:30
up a little bit. It's definitely spicy. It definitely hits you. My mouth is
44:33
burning, but it's like Yeah. an okay amount of burning. I'm not I'm
44:38
not having uncontrollable hiccups. I get hiccups when stuff's spicy.
44:43
So, you can buy a mix. You can get like cracked pepper and sea salt. Yeah.
44:47
And a few other ones. And like find your find your sweet spot like I did with
44:51
Mojo. Now, to be clear, your friend is good. It's just what you
44:56
can handle. Yeah. So, Savage Jerky is made with the best
44:59
ingredients without nitrates or or preservatives.
45:03
And the goal was to create a snack that's full of flavor and spice that
45:07
isn't bad for you. The inspiration came from the flavor of
45:10
garlic, lime, and cilantro as it brings out the natural flavor in premium beef.
45:16
And uh this is in my notes. People seem
45:19
to love the Sriracha bacon, maple buffalo bacon, and the traditional
45:22
flavors. They also make barbecue sauce. Wow, I had no idea.
45:27
That's cool. So, use offer code LTT to save 10% over at savagejerky.com.
45:34
The cracked pepper and sea salt is like a really good but more traditional
45:38
styled beef jerky. Yeah. The problem with that one is that you can rip through a whole pack of it
45:42
in like 2 minutes. Yeah. So yeah, if you want to like troll your
45:47
friends and get a ghost pepper or reaper or something like that, it works.
45:52
Or you might just like it. If you first bite, it's actually easier.
45:57
It gets a little bit easier. Yeah, it does get easier.
46:01
All right. It's It's It's also when you stop,
46:05
though. Mhm. Cuz it gets easier after the first bite
46:08
and then for a while, but then when you stop eating it. Oh, yeah. No. Then then it hits you
46:12
again. So LTX LTX is the uh tech carnival that we're
46:18
doing in the summer. We're going to have booths, games, prizes, and us, the whole
46:23
LMG team will be there. Um, a few examples of the kinds of booths that
46:27
we're going to have. Like, we wanted it to be just kind of goofy. So, it's going
46:31
to be stuff like guess how many CPUs are in the jar. Um,
46:35
video editing with Taran. Yeah, like the jelly bean thing, but like
46:39
CPUs. I think FreeGeek's probably going to be in front of So, they'll be there with like, you
46:44
know, just some Oh, free geek's going to be there. Maybe. Maybe. Okay. TBD, but like we're we're in talks with
46:49
actually a lot of pretty cool folks. Um, uh, one other example would be the case
46:53
toss. So, we're going to get a case manufacturer to give us uh, some cases.
46:57
Is that nailed down now? Um, not exactly who, but someone's going
47:00
to do it. Okay. Okay. Cuz I'm pretty excited about that, and I know there's a few people
47:05
that are pretty excited about that. We're going to do like a VR with Luke demo. So, basically, you'll be like kind
47:09
of showing people the Oh, that's exciting. That's cool. I've been wondering this
47:13
whole time what I was going to do. That's really cool. And then Taran's probably going to be
47:17
doing like he'll have an LTT video and
47:21
he'll just be working on it the whole time. So, you could actually watch Taran
47:24
edit. You can ask him questions. That's been a great idea. We're not expecting him to actually get
47:28
anything done because people will be like blah blah blah. But, like, if
47:32
people want to see how it's done in person, I thought that would be really cool. We're just going to like pick up
47:36
his workstation and take it there. So, we've got a lot of really cool ideas.
47:39
There's at least a few hundred people coming so far.
47:44
So, it's not going to necessarily be like over a thousand people and crazy,
47:48
but that may actually be a blessing in disguise because at least that way we're
47:52
going to be covering with uh like a
47:55
significant title sponsor. We should be able to cover most of the cost of the
47:59
venue already. And um and it'll actually
48:03
it could end up being a pretty intimate event. Although, now that I say that, if everyone's like, "Oh, yeah. I want to go
48:06
to an infinite. So, no guarantees.
48:10
How many people are going on your limo ride? Um I think that sold out.
48:13
Whoa. 10. Yeah. Cool.
48:16
So, that was that happened faster than I expected. Um
48:20
I thought at when I first read that that by the limo they meant the Civic.
48:24
No. No. Then I was like, "Oh, that's too many people." Yeah.
48:27
Yeah. Yeah. 10 people in my Civic would be like pretty intimate
48:34
as that goes. All right, I want to talk about this.
48:40
Um, W to Cry collects basically no
48:44
ransom despite infecting 200,000
48:51
machines. So, the deadline is uh is
48:54
beginning to pass here. The original article is from Bloomberg.
48:58
I have a theory. Um, and apparently the attackers have
49:01
claimed only about $92,000 in payments.
49:05
Huh? How about that? So, my theory is if this happened like a
49:12
number of years ago, like when I was still in high school, I think it would
49:15
have actually been a lot more impactful because now I think if people like have
49:20
a baby now or get engaged now, all their
49:23
photos go up on social media. So, it's not their one backup that's on
49:27
their computer. And I think what they ran into here was if people had super
49:32
important stuff, the 300 bucks was not
49:36
good, but probably not the end of the world. And then other people are like,
49:41
"Oh, well, this is a fairly major inconvenience, but I can go get my
49:45
computer reformatted and everything's fine because everything I do is in a
49:49
browser." Notice how they're releasing computers that are like almost just browsers these days.
49:53
So like for a lot of people it's like, "Oh, what the heck is this?" they show it to some
49:57
techie friend or they bring it to Geek Squad. They're just like, "Oh yeah, we can do it cheaper than that ransom price
50:02
anyways, but we have to do some stuff." And then people that are like, "Oh my
50:05
goodness, this is mission critical information. 300 bucks for like a big
50:09
business sucks, but like
50:12
not the end of the world." So if even 1% of people had paid the
50:16
ransom, it would have worked out to over six times the amount they actually
50:21
collected. So I feel like almost no Joe Schmoes
50:24
actually paid for it. And it gets worse because of the attention on this
50:29
particular case. The authorities think it's unlikely the monies the attackers
50:34
have received will be transferred from Bitcoin to more traditional currencies
50:38
due to the risk of it being detected. With that said, there's a lot of stuff
50:41
you can buy with Bitcoin. So, I don't think that's as big a problem as some
50:44
people might. Yeah, but they could they could still track that to a certain degree,
50:49
right? I guess that's true. Like that money is sketch depending on but but okay. But they
50:54
might do for sure. They might buy very illegitimate things,
50:58
scatter it around, and then Yeah. What up?
51:01
Very easy to like mix the Bitcoin.
51:05
Yeah. No, it should be fine.
51:08
Um Oh, really? Apparently, people sell like mixing it up and obscuring it as a
51:12
service. That's cool. I don't know enough about Bitcoin. Why do you know this? Go away.
51:18
Jake is secretly running Silk Road 3 right now
51:21
out of his house. He told me this offline.
51:25
Great. Yeah. Yeah.
51:29
Jake, can I have a favor? Yeah. Can I have more water, please?
51:33
Oh my god. You have to pay him in Bitcoin through
51:36
his secret onion website. Thank you. I'm dying.
51:40
Thank you. And ice, too. Thank you.
51:46
But it's okay. Is that how that works?
51:49
Yeah. I like have like mut built up in
51:53
my throat. I can hear the Oh, your throat's even red here.
51:56
Well, that's probably from I always have a redneck. Okay.
52:00
Yeah, I was born in the sticks. Um Oh,
52:04
okay. Saying you need milk. I don't think we just have milk. Yeah, we don't, unfortunately. And yes,
52:09
I know milk works a lot better than water. Um, so this was originally posted
52:13
on the forum by WM Groom AK and the original article is from the tech
52:17
report. Optane dims and companion CPUs will
52:22
arrive in 2018. So if you haven't been
52:26
paying attention to this, Optane DIMs
52:29
are going to be super high capacity modules, too.
52:33
That will fit in Kidding. I'm kidding. It's fine.
52:40
Ah, much better. Super high-capacity
52:43
modules that will fit in otherwise normal RAM slots as far as I know. Then
52:48
there will be a like a a special BIOS
52:51
configuration that allows you to use it as a slower kind of uh so if you got
52:56
your level one cache and your level two cache, your level three cache, I guess
53:01
in some cases there's level four, but don't worry about it. Then your next tier is DRAM. And then traditionally
53:06
your next tier was NAND but it's like many orders of magnitude slower. So
53:11
Optane sits in between. And uh what this
53:16
will allow you to do is have massive effective
53:20
memory capacity for your system without needing so many dim slots and without
53:27
spending so much. So, when we were checking out SFU's supercomputer, one of
53:32
the things that they told us was that
53:36
Oh, okay.
53:40
Their uh their systems that needed 1 and
53:43
a half terabytes uh excuse me, yeah, wait. Yeah, one and a half terabytes of
53:48
memory had to be 2U just so they could fit all those RAM slots. Then
53:54
the systems that were going to be used for workloads that needed even more
53:58
memory than that, the 3 TB memory systems of which I think they only have
54:01
four had to be four U systems with quad
54:05
sockets. But get this, the quad socket
54:08
CPUs they were using were eight cores. So their quad socket systems actually
54:14
had the same number of cores as their
54:17
dual 16 core socket systems
54:20
or their dual socket 16 core systems. So 30 uh yeah 32 total and so because the
54:26
only reason they needed it was so they could have more RAM. So this could solve
54:30
that problem as long as the workloads that they're using can are okay with the
54:36
data being a little bit slower but not as slow as trying to put it on NAND
54:41
flash. Someone just asked me what I'm eating. We literally just did a sponsor spot.
54:45
It's Savage Jerky. I'm just still eating their stuff. Um, another cool thing is
54:49
that they're saying that it's supposed to be cheaper, lower prices than DRAM.
54:54
Yes. Much cheaper than DRAM, which like usually happens with new
54:58
technologies over time. It's not super
55:01
common for that to happen with a new technology right out of the gate. So,
55:04
that's pretty cool. But it's intended to be because it's
55:08
it's designed to be an order of magnitude bigger than DRAM, but also a
55:13
lot cheaper to sit right in between NAND and DAM. I'm actually I'm super excited
55:18
to see where this takes us in the future. Oh, I realized I should probably
55:21
talk about the deal that we are working on. So, we've been we've been working
55:26
with Lime, makers of Unrade, for
55:31
a year and a half now, I think, going all the way back to Oh, no, cuz Seven
55:35
Gamers 1 CPU was a year and a half ago. Yeah. So, it would have been closer to two
55:40
years now, I guess. Wow. Really? Anyway,
55:43
so we've worked with them on a number of projects. We use their software and I
55:48
like it so much that basically what we're what we're working on is an
55:53
agreement where I would actually take a stake in the company and I would be
55:56
working with them on future development. Um we've seen it in action in terms of
56:02
where it's at today, but I think that there's a lot of stuff that we could
56:06
help with that would make it so that they could really take things to the
56:10
next level. So basically that's where I'm at on that. um nothing is finalized
56:15
so I can't really I can't really give any more detail than that but I just
56:19
thought in the interest of you guys understanding you know sort of what
56:22
we're up to over here I want to make sure that that's clear uh the reason
56:26
that I was comfortable with lime even though I have specifically gone out of
56:30
my way to not buy shares in companies like AMD and Intel is that we don't
56:34
really cover their competitors I mean not just in the last 2 years but even
56:40
you go back a couple years before that we basically don't touch NAS systems.
56:46
Um, and the reason for that is because we just don't really like most of them.
56:51
We don't think they're we don't think they're userfriendly. We don't think they're that easy to use. Uh, which is
56:56
the same thing. That's the main reason that I don't like them. I tried.
57:00
Um, and quite frankly, viewer interest is
57:04
not that high. And a lot of the stuff that we've done with Unrade has really
57:08
been more about the other functionality that it has. So like the uh the
57:12
virtualization and stuff like that. So yeah, that's basically where I'm at on
57:16
that. Um
57:23
so what else do we got? I mean, we got actually a lot of topics this week.
57:27
These are scrap wars. Yeah, I'm I'm down. How are we How are
57:31
we going to do that again? Jump to random part.
57:35
Okay, so I don't know how to do I Oh,
57:40
you know what? I never figured out the audio input.
57:44
So, that's a thing that I didn't do. Um,
57:48
let's go ahead and How did we do it last time? Uh, I held up the laptop to the
57:53
microphone, which was terrible. Um, so
57:58
that might be the only way to do this. Well, it's not the only way, but it's
58:02
the only way. Why can't you listen to it over HDMI right now? Because I don't know either.
58:07
These don't take audio or go right click on it.
58:13
Uh what? Yeah. Go to cam real quick.
58:17
Go to cam. Yeah. Change the audio from built-in/none to
58:21
the recorder that you wanted to come from. Never mind. There they go. See why? What
58:26
would I do without you? What would I do without you?
58:30
Probably a lot of the same stuff. It just wouldn't be as fun. Or as good.
58:33
Yeah. Yeah. Yay. Yay, Luke.
58:37
So, uh, Scrapyard War season 5 episode 1
58:41
is hopefully this works up already over on Float Playing Club.
58:46
Oh, hold on. Well, you got to play back devices.
58:53
Oh, I can't hear that in my monitor, which is super scary because I have no
58:58
idea what kind of volume I'm outputting at. So, uh, I'm sure Twitch chat will be
59:03
telling me if I broke their eard drums. If you plugged it into the computer
59:07
directly, you would hear it.
59:12
Yeah, in theory, this is that's how this is supposed to work, but I forgot which
59:15
one I'm supposed to plug into. So, cuz I'm really really dumb.
59:19
Apparently, they can hear and it's a little bit too loud. Okay. Well, then
59:22
most people are saying it's good, though. The vast majority of people are saying it's good. So, go like a little
59:26
bit down and then So, we'll call that good. So, so you
59:30
just want to skip to So, 45minute episode. Actually, 46 minute episode. Do
59:35
you just want to skip to a random spot? Sure. I don't know. I don't even know
59:38
what's going to come up. Okay. Why are we watching in 360p?
59:42
That's peasant. That's peasant stuff. Peasant stuff.
59:48
I have before. Do you see that playback? You see that lightning fast playback? Hold on. Hold
59:52
on. Stop. Here. Here. Listen for the click. And we click. One of the things too is
59:57
that West Coast North America is like not even a good area for us.
60:01
We're going to be $25 over. I don't even know what we're doing here.
60:05
Me neither. $25 over. Oh, do we not have the preview working
60:09
for such long yet? I guess maybe.
60:13
Interesting. Makes sense. Two drives for 10 bucks.
60:17
I wonder what we're figuring out here. It looks suspenseful.
60:21
Yeah. Yeah, I know what that is. Yeah, I know what's going on. I know what's going on. Calling up um here. Let's see if
60:28
Okay. Okay. Okay. So, that that's enough. So, basically, what what was the
60:32
point of this? So, Scrapyard Wars episode 1 is coming
60:36
out. That right there is coming out on YouTube tomorrow.
60:39
Tomorrow. If you want to see episode two tomorrow,
60:44
you can sign up for Floatplane and you can see that today, episode one today,
60:48
and then watch episode two tomorrow and be totally caught up on the Floatplane
60:52
side of things. Um, also really cool note, this is what I wasn't sure about
60:56
telling you guys last time. We now have
61:00
some load balancing going on with geos. So, a lot of people in Australia and the
61:06
kind of general European area um have
61:09
noticed a lot of improvements. That's because we have some like kind of more
61:14
onloation servers for you guys up and running with some load balancing going
61:18
on. Do you want to say where they are? Yeah, Australia. So, Australia and the
61:23
other ones in France, but it should be able to better serve you than ones in
61:27
Canada. So, yeah, a lot of people from various countries around Europe have had
61:32
a much better experience um than before when everyone was just
61:36
going through Canadian servers. We are going to continue to expand. So, if
61:40
you're in an area that like isn't super great right now, ironically, where we
61:43
are is like not the greatest, it's good. You just watched the playback. It's
61:48
good. Um, it could be slightly better and it will be slightly better because
61:52
we'll eventually have West Coast servers. Um, but yeah, if you're in an
61:56
area and you are consistently not getting super great bandwidth, I'm not
62:00
hearing this very often, almost not at all. But if you are, let me know where
62:03
you're from and we will look into getting that better. To be clear, like the infrastructure
62:07
upgrades we're making right now have nothing to do with capacity for the platform. Um,
62:12
like there's a lot more people on it than you would probably think,
62:17
but it's not like, you know, a billion people or anything like YouTube or
62:21
something. It's more just trying to make user experience better. Yeah.
62:26
I mean, this is the kind of boring stuff
62:29
that takes months to work on and figure out before you want to launch and like
62:34
bring a bunch of other creators. Yeah. And it's frustrating when like we're an international group and we have
62:40
people from North America watching all the footage and they're like everything
62:44
looks great. I can watch 1080p. Everything it's snappy. Hooray. And then someone from Australia is like I have to
62:49
download it. Or like I can sometimes watch it in
62:52
360p. And I'm like that's not fair. People watch us from everywhere. I want
62:56
this to work well for everyone even if it cost us more. So now we're doing that.
62:59
Um so we're actually going to do like a more Floatplane another Floatplane Q&A
63:04
after the WAN Show. if you guys want to kind of hang around for that. Um,
63:08
otherwise, I think we've got a few more pretty good topics for you.
63:11
Um, anyways, one quick, actually, no, we have like nothing. Oh,
63:14
wait, no, we do have more. There's also a little teaser of episode
63:18
two on Floatplane only. So, if you're a Floatplane guy and you haven't been
63:22
there on there in a couple days, jump on there. Check out the teaser for Scrapper
63:25
Wars episode 2. It's it's it's pretty good.
63:30
What? Huh? So, actually, I did not know about
63:33
this. So, this was posted by WM Groom AK on the forum. The original article is
63:38
from RS Technica. And it looks like
63:42
Google gets to keep their trademark even
63:46
though it has turned into a generic term
63:49
for searching the web. So, going back to our notes here, when a brand name
63:54
becomes completely ubiquitous, like thermos, I didn't even know teleprompter
63:58
was a brand name. Me neither. And videotape. Didn't know that was a
64:02
brand name. Oh my goodness. either it can lose its legally protected
64:05
status. So the this process is called genericide and in 2012 Google filed a
64:11
cyber squatting complaint and claimed trademark infringement when a man
64:15
registered 763 domain names that
64:18
combined Google with other words and phrases including googledonaldTrump.com.
64:23
Oh, what a jerk. I see what he was doing there. So, the man filed an appeal
64:26
citing genericide and the court ruled that Google still retains its trademark
64:30
even if the term Google has become known for searching the internet.
64:33
So, how does that actually work though? Um, I have no idea. I mean, John would
64:40
be a more sort of cuz like sensible person to talk about this. Hey,
64:44
hey, hey, John.
64:47
He's coming. He's coming. All right. I'm giving up my seat. I'm giving up my
64:51
seat. Okay. John's coming in to talk about Janera's side. John, generic side.
64:55
All right. This thing, eh? Okay. Hit me or Luke.
65:01
Hit something. Yeah, there you go. Good job. Uide.
65:05
Google almost potentially lost its trademark on Google because like, oh,
65:10
I'll just Google it has become a general term for looking something up on the
65:14
internet. Um, someone sued them over it
65:17
because they tried to take down a bunch of his domains and blah blah blah blah.
65:21
And he was like, "Okay, so you used a lot of pronouns there. Okay, let's see.
65:24
In 2012, Google filed a cyber squatting complaint and claimed trademark
65:28
infringement when a man registered 763
65:31
domain names that combined Google with other words and phrases such as Googled
65:35
Donald Trump.com. The man filed an appeal citing genericide. The court
65:40
ruled that ruled that Google still retains its trademark even if the term
65:44
Google has become known for searching the internet. So, how does that work?
65:47
Um, isn't that literally the even though Google has become known for searching
65:51
the internet? Isn't that the whole point of genericide? Uh, it is. Um, genericide is kind of um
65:58
it's not a hard and fast rule. It's not
66:01
like, oh, after a certain amount of time or a certain number of people say it in
66:05
a generic manner, it becomes generic. It's a little bit squishy. Uh, but they
66:10
do look at things like um and I don't know what happened in this case because
66:13
I just have the one little note in front of me, but they could have looked at something like, oh, has Google taken a
66:17
lot of the steps to protect its trademark? Um, you know, back uh back in
66:21
the day before um um when when
66:24
photocopers were still like a like more
66:28
of a cutting edge thing than they are now, um people would say, "Oh, Xerox
66:32
it," you know, to make a copy even though there were tons of copers not
66:35
made by Xerox, right? So, Xerox actually took out like a big newspaper ad say
66:41
saying or telling people say, "Make a copy. Don't say Xerox because they're
66:46
trying to protect their trademark." So if Google maybe Google showed that they
66:49
were taking um active steps to protect the trademark. So
66:52
maybe it also helps them because Google puts their Google branding on way more
66:56
things than just a browser. They do. Yeah, that might help. I have no idea.
67:00
Yeah, maybe. It's like I said, it's kind of a squishy area of law. I'm not
67:05
surprised that Google won. Um it takes
67:08
it takes like quite Yeah. I mean, just aside from the money they have to spend,
67:12
it takes it takes a lot for a trademark to become generic. So,
67:17
yeah. Yeah. So, there we go. Cool. That's it. I think that's it.
67:20
Okay. Carry on, gentlemen.
67:24
I'm back. The FAA's drone registry requirement got shot down. The original
67:29
article is from TechCrunch. Let's go ahead and pull that up. But basically, a
67:33
federal appeals court shoots down
67:37
the FAA's drone registry registry requirement. It's likely that the FAA
67:43
will appeal this decision or take some other approach uh which seems to be very
67:47
common in the states with stuff like net neutrality being tried like a billion times under a billion different names.
67:52
So drone hobbyist John Taylor argued that the Federal Aviation Administration
67:56
doesn't have jurisdiction over what the law classifies as model aircraft. Um the
68:02
registration database was proposed in 2015 to address growing drone ownership
68:06
in the US which has brought with it privacy and safety concerns. Um, so
68:11
there that's basically where we're at on that. So stay tuned for more and um,
68:17
yeah, blah blah blah blah blah blah blah blah blah blah blah blah. Okay, winners
68:22
of the Casemod World Series 2017
68:25
posted by Rod Rosenberg on the forum. Man, I hate that guy.
68:29
Was it Rod? Yeah, I don't I don't know. I don't know if Rod won.
68:32
Did he post his own winning?
68:35
Let's see. No. No, it doesn't look like it. Was he in it or was he a judge? Yeah.
68:39
Where? I don't uh I have a feeling he has something to do
68:42
with this. I'm just trying to find like a Oh, okay. No, there's
68:47
there's no uh no BS mods. Maybe he says something in the link.
68:53
Yeah, I can't click on it though.
68:57
Okay, so the first winner was uh
69:01
Oh, no. I think he just thinks it's cool. That's cool. Vega by S Russia.
69:08
Yes, Russia. No, no, not FPS Russia. That's a different thing. But good try.
69:12
Good try. Thank you, Nick. Um, it looks cool.
69:15
That's apparently made out of a Mastercase 5.
69:19
It looks like they just chopped it like
69:23
pretty early on. Completely unrecognizable.
69:27
Well, no. The the the very bit at the front, this bit is recognizable. And
69:30
then that's it. Then it stops.
69:34
Under goes a complete structural change. Yeah, we noticed the complete structural
69:38
change. We did notice that. Pretty pretty intense. Yeah.
69:42
Like, hey, here's a case. We'll we'll just take the top quarter of it. You can
69:45
take the rest. We don't need it. Not a big deal.
69:49
That's awesome. What is that based on? Those are pretty cool. Like a stacker or something.
69:53
Futuristic ideas and metal uh modded 30.
69:58
Okay. Yeah, I can see that. Cool. PSU on the side. Just put the PSU
70:02
over there. Seems to be kind of open like you can see the graphics card. You
70:06
can see You're pretty neat. That's actually really pretty cool. It looks really good. Yeah.
70:09
All right, and that's pretty much it. Thank you for tuning in to the WAN Show.
70:12
We will see you again next week. Same bat time, same bat channel.
70:21
Outro. Oh, yeah. I'm really not firing on all
70:24
cylinders today. That's cool.
70:28
Actually, I am. I'm just in other places.
70:34
What's up? Someone's like, "Finally, a good W show episode." Good job, guys.
70:39
I'm surprised. That was decent. We're so
70:42
hungry. Yeah, last week sucked, but I thought the one before that was pretty good.
70:45
Yeah. Oh. Oh, that's
70:49
the afterparty for last week was wonderful.