WEBVTT

00:00:00.080 --> 00:00:07.200
is it just me or is dr lisa sue turning into the master of dunking on the

00:00:05.839 --> 00:00:13.040
competition last time she got on stage she dropped a

00:00:09.360 --> 00:00:16.000
ryzen shaped bomb on Intel and now she's

00:00:13.040 --> 00:00:20.320
at it again this time taking on jensen leatherman huang

00:00:17.760 --> 00:00:25.359
and from what she showed us today jensen might want to keep his oven on so he can

00:00:23.039 --> 00:00:30.960
pump out some more gpus before the new Radeon 6000 series hit store shelves on

00:00:28.240 --> 00:00:35.680
november 18th that's right get back in the kitchen

00:00:32.239 --> 00:00:37.360
it's not sexist anymore cause like she's

00:00:35.680 --> 00:00:41.440
telling him to get back in the kitchen

00:00:39.520 --> 00:00:45.360
just like i should have sent this segway to our sponsor back to the kitchen glass

00:00:43.600 --> 00:00:49.039
wire helps you keep track of the weird stuff that's connecting to your pc even

00:00:47.440 --> 00:00:53.520
when you're not using it if a strange device joins your Wi-Fi you'll be

00:00:50.640 --> 00:00:57.879
notified right away get 25 off by using code Linus at the link below

00:01:04.239 --> 00:01:10.400
let's get this out of the way up front we wrote this script before the big

00:01:08.080 --> 00:01:16.640
reveal so we didn't know pricing for the new cards yet so however excited i am

00:01:13.520 --> 00:01:19.280
right now uh turn it up or down a few

00:01:16.640 --> 00:01:24.479
notches depending on how much lower or higher price the new cards are compared

00:01:21.439 --> 00:01:25.360
to the rtx 3000 series okay

00:01:24.479 --> 00:01:32.240
cool we're going to pin a comment below with the pricing that dr sue announced today

00:01:29.040 --> 00:01:35.200
now our dna 2 has been rumored it's been

00:01:32.240 --> 00:01:41.119
speculated on heck i've even played on the new architecture in the xbox but now

00:01:38.000 --> 00:01:43.200
we finally got the details for the pc

00:01:41.119 --> 00:01:48.399
add-in cards and the first thing we know is that AMD is using the same seven

00:01:45.759 --> 00:01:54.320
nanometer process as their previous Radeon 5000 series cards now given that

00:01:51.600 --> 00:01:59.040
NVIDIA had a performance lead before they moved to their new 8 nanometer node

00:01:56.720 --> 00:02:03.840
with ampere and considering the performance they gained by doing so AMD

00:02:01.520 --> 00:02:08.080
staying on their old process might not sound like it bodes well for the new

00:02:05.680 --> 00:02:14.560
cards but they claim that compared to the Radeon 5000 series they've pulled

00:02:10.239 --> 00:02:17.280
off up to 200 better performance and 50

00:02:14.560 --> 00:02:21.520
better performance per watt which mostly just means that the power draw is the

00:02:19.040 --> 00:02:26.080
same but still considering that the higher end rtx 3000 series from NVIDIA

00:02:24.239 --> 00:02:29.599
is chugging down power like it's trying to drain a lake not too shabby had they

00:02:28.239 --> 00:02:34.560
accomplished all that well unfortunately AMD didn't go into

00:02:32.080 --> 00:02:38.560
much depth but they did say that they tuned their core design for frequency

00:02:36.720 --> 00:02:42.480
and power efficiency just like what the hell else would you tune it for you're

00:02:40.480 --> 00:02:46.959
not tuning it for crappy clock speeds and crummy power consumption so that's

00:02:45.120 --> 00:02:51.680
all marketing speak but they did provide a couple preliminary details if we drill

00:02:49.680 --> 00:02:56.000
a touch deeper the pipeline has been completely rebalanced with a focus on

00:02:53.680 --> 00:02:59.760
optimizing data paths that could mean another update to the way that work

00:02:57.760 --> 00:03:02.400
groups are handled in parallel or it could mean something entirely different

00:03:01.200 --> 00:03:07.200
we're going to have to get our hands on them to know for sure all we can say for

00:03:04.879 --> 00:03:12.319
now is that AMD assures us that it's been designed for efficiency first

00:03:09.680 --> 00:03:16.800
just like our water bottles lttstore.com they keep it ice cold now efficiency is

00:03:14.959 --> 00:03:20.720
good but what other goodies do we have with the new architecture for one we're

00:03:18.720 --> 00:03:25.280
getting full ray tracing support in the form of ray accelerators built into

00:03:23.200 --> 00:03:30.319
every compute unit which not only allows the new gpus to run any standard ray

00:03:27.599 --> 00:03:35.040
trace title to date but also enables new functionality with fidelityfx AMD's open

00:03:33.280 --> 00:03:40.000
source image processing toolkit including a denoiser for ray traced

00:03:37.200 --> 00:03:44.239
content and acceleration for the ambient occlusion and screen space reflection

00:03:41.760 --> 00:03:48.400
features that AMD is already using AMD says that their ray accelerators are 10

00:03:46.879 --> 00:03:52.640
times faster than ray tracing in software though we don't really have a

00:03:50.640 --> 00:03:57.599
baseline for that since AMD never enabled a ray tracing API for rdna1

00:03:56.080 --> 00:04:02.000
i have a good feeling about it though from looking at gtx cards running ray

00:03:59.760 --> 00:04:06.720
tracing in software those features tie into AMD's overarching support for

00:04:04.080 --> 00:04:11.439
directx 12 ultimate which also includes things like variable rate shaders that

00:04:08.799 --> 00:04:15.120
dynamically reduce complexity for less frequently updated portions of the

00:04:12.879 --> 00:04:19.440
screen mesh shaders that combine multiple shading passes at once and of

00:04:17.359 --> 00:04:23.600
course direct storage support which is baked into both of the new rdna 2

00:04:22.079 --> 00:04:27.440
powered consoles sort of playstation implementation is

00:04:25.759 --> 00:04:31.520
not quite direct storage which is a microsoft tech

00:04:28.880 --> 00:04:36.720
as a result then AMD claims that because the consoles will be optimizing for

00:04:33.680 --> 00:04:38.880
exactly these features cross-platform pc

00:04:36.720 --> 00:04:42.639
games will be tuned for them pretty much by default it's obviously not quite that

00:04:41.280 --> 00:04:46.479
simple but my hopes are still pretty high AMD also

00:04:44.880 --> 00:04:50.800
offered us a glimpse at a feature they call super resolution which for all

00:04:49.040 --> 00:04:53.680
intents and purposes sounds like their version of dlss

00:04:52.639 --> 00:04:58.639
except not using quite the same method

00:04:56.639 --> 00:05:02.160
now AMD was hesitant to get into the details of it just yet so make sure that

00:05:00.560 --> 00:05:06.320
you're subscribed so you don't miss our review with all the dirt

00:05:04.639 --> 00:05:11.600
finally there's something that AMD is calling infinity cash which i swear is

00:05:09.600 --> 00:05:16.800
not a cryptocurrency it does sound like one though doesn't it instead infinity

00:05:13.680 --> 00:05:20.080
cash consists of 128 megabytes of

00:05:16.800 --> 00:05:22.800
extremely fast on-die memory that just

00:05:20.080 --> 00:05:27.680
like ryzen's cash keeps the GPU fed during memory intensive operations which

00:05:24.800 --> 00:05:33.199
AMD specifically says minimizes dram bottlenecks from AMD slides it looks

00:05:29.919 --> 00:05:35.600
like this alone accounts for 40 of our

00:05:33.199 --> 00:05:40.320
dna 2's overall performance improvement and in theory it should allow rd a2 to

00:05:38.080 --> 00:05:47.360
maintain performance at high resolutions despite the 256-bit gddr6

00:05:44.160 --> 00:05:50.080
not x memory that AMD is using on all of

00:05:47.360 --> 00:05:53.440
its Radeon 6000 gpus and

00:05:50.960 --> 00:06:00.400
which gpus are there well got a couple for you the rx 6800 and 6800 xt are the

00:05:57.840 --> 00:06:05.600
two gpus that AMD is prepping to try and propel themselves back to enthusiast

00:06:02.400 --> 00:06:07.600
status with the rx 6800 completely

00:06:05.600 --> 00:06:12.319
dominating tomorrow's you mean yesterdays it launches tomorrow oh right

00:06:10.240 --> 00:06:17.120
completely dominating tomorrow's rtx 3070 in performance if AMD slides are to

00:06:15.039 --> 00:06:23.199
be believed bearing in mind though that the 3070 is roughly equivalent to an rtx

00:06:20.400 --> 00:06:26.880
2080 ti it really is no wonder NVIDIA carpet bombed the new cycle if these

00:06:24.960 --> 00:06:31.600
numbers are true and we will of course confirm them before the simultaneous

00:06:28.639 --> 00:06:35.520
launch of the 6800 and 6800 xt on november 18th

00:06:32.960 --> 00:06:41.120
i am officially telling you dear viewer to wait

00:06:36.800 --> 00:06:43.199
for rx 6800 before buying a 3070. and if

00:06:41.120 --> 00:06:49.360
you're one of the many waiting on an rtx 3080 the fact that the rx 6800 xt is

00:06:47.039 --> 00:06:54.400
expected to trade blows with it while drawing slightly less power may be

00:06:51.440 --> 00:06:57.520
enough to give you pause as well but wait what's that the performance

00:06:56.319 --> 00:07:03.919
slides say rage mode smart access memory what the heck are

00:07:00.960 --> 00:07:08.479
these okay well rage mode is basically an easy mode way to bump up the power

00:07:06.240 --> 00:07:12.000
target and fan curves in a single click although AMD did admit that it's

00:07:10.319 --> 00:07:16.080
probably only going to net you one to two percent better performance

00:07:13.840 --> 00:07:20.720
as for smart memory access well here's where things get really interesting AMD

00:07:18.560 --> 00:07:26.720
is leveraging their platform control with ryzen 5000 and 500 series

00:07:23.599 --> 00:07:29.919
motherboards by allowing those cpus to

00:07:26.720 --> 00:07:33.440
directly address the new Radeon 6000

00:07:29.919 --> 00:07:36.400
series is memory all 16 gigs of it now

00:07:33.440 --> 00:07:41.840
usually this has to be done in 256 megabyte chunks in the form of the pci

00:07:38.880 --> 00:07:46.319
express base address register or bar which means that ryzen plus Radeon

00:07:44.319 --> 00:07:52.080
allows for a level of optimization that simply isn't possible on a traditional

00:07:49.360 --> 00:07:55.039
system using previous gen ryzen or Intel cpus

00:07:53.280 --> 00:07:59.919
now currently this is optimized primarily for games and while not all

00:07:58.000 --> 00:08:04.240
games will see a benefit performance uplift according to AMD should be

00:08:01.520 --> 00:08:09.280
somewhere between 6 and 10 for those that do just from having a 5th

00:08:06.720 --> 00:08:13.520
gen ryzen CPU now i guess we've really got a reason to upgrade the lan center

00:08:11.280 --> 00:08:17.360
although i do have some discomfort about proprietary memory sharing nonsense and

00:08:16.160 --> 00:08:22.240
whether developers will actually implement it and all that but again

00:08:20.080 --> 00:08:27.840
stay tuned for the full review oh yeah and one more thing who could

00:08:24.000 --> 00:08:31.280
forget the big boy the rx 6900 xt

00:08:27.840 --> 00:08:33.760
nice an 80 compute unit Radeon GPU with

00:08:31.280 --> 00:08:41.360
performance that is allegedly on par with the 350 watt rtx 3090 yet with a

00:08:38.320 --> 00:08:43.880
total graphics power of 300 watts and a

00:08:41.360 --> 00:08:50.240
power supply requirement of just 650 watts that happens to be the same

00:08:46.640 --> 00:08:52.720
power requirements as the rx 6800 xt and

00:08:50.240 --> 00:08:56.399
the same core clocks clearly this is a really really well-bended piece of

00:08:54.640 --> 00:08:59.760
silicon if that is to be believed and maybe that's why AMD is giving

00:08:57.839 --> 00:09:02.720
themselves another three weeks to release it hopefully they don't have to

00:09:01.200 --> 00:09:06.480
push it back any further to resolve supply shortages

00:09:04.399 --> 00:09:10.160
i'm sorry excuse me i guess we won't know any more than this

00:09:08.320 --> 00:09:14.800
until we have the cards in our hands and we've seen them on store shelves which

00:09:12.000 --> 00:09:20.880
is another three weeks out but at that time we are going to have a very

00:09:17.360 --> 00:09:23.200
good idea of how ryzen 5000 and rtx 3070

00:09:20.880 --> 00:09:27.839
are gonna perform so we're gonna have plenty of comparison points for our

00:09:26.560 --> 00:09:32.240
review just like you're gonna have plenty of

00:09:29.760 --> 00:09:35.920
message from our sponsor ting ting does mobile phone service

00:09:33.839 --> 00:09:40.240
differently there's no contracts no overage fees and no other carrier tricks

00:09:38.480 --> 00:09:43.920
you just pay a fair price for the talk text and data that you use every month

00:09:42.320 --> 00:09:47.440
and it's especially great if you're stuck at home using wi-fi instead of

00:09:45.760 --> 00:09:50.800
mobile data why would you be paying for it if you're not using it ting doesn't

00:09:49.200 --> 00:09:54.160
think you should have to and they give you complete control over your account

00:09:52.480 --> 00:09:59.680
you can set alerts and caps for each device in your account to keep your usage in check and you get nationwide

00:09:57.360 --> 00:10:03.600
lte coverage in the us using t-mobile and sprint almost any phone will work

00:10:02.000 --> 00:10:07.600
with ting and you can check your phone's compatibility at

00:10:05.560 --> 00:10:11.839
Linus.ting.com and you'll even get a bonus credit for signing up

00:10:09.920 --> 00:10:16.160
thanks for watching guys go check our launch reaction video for the ryzen 5000

00:10:14.399 --> 00:10:22.800
series to get an idea of what to look forward to before november's out but

00:10:19.279 --> 00:10:24.959
dang text hover has not let me down

00:10:22.800 --> 00:10:29.519
i mean iphone 12 is pretty boring but like pc stuff yeah we finally got some

00:10:27.600 --> 00:10:32.720
cool stuff thanks AMD and NVIDIA did Intel do anything

00:10:31.279 --> 00:10:35.040
noteworthy okay