AMD Radeon Rx6000 series

AMD Radeon Rx6000 series

 - Is it just me, or is Dr. Lisa Su turning into the master ofdunking on the competition? Last time she got on stage she dropped a Ryzen-shaped bomb on Intel, and now she's at it again, this time taking onJensen Leatherman Huang, and from what she showed us today, Jensen might wanna keep his oven on so he can pump out some more GPUs before the new Radeon 6000 series hits store shelves on November 18th. That's right, get back in the kitchen. It's not sexist anymore, 'cause, like, she's tellin' him toget back in the kitchen. Just like I shoulda sentthis segue to our sponsor back to the kitchen. GlassWire helps you keeptrack of the weird stuff that's connecting to your PC, even when you're not using it. If a strange device joins your WiFi, you'll be notified right away. Get 25% off by using codeLinus at the link below. (upbeat music) Let's get this out of the way up front: we wrote this scriptbefore the big reveal, so we didn't know pricingfor the new cards yet, so however excited I am right now, turn it up or down a few notches depending on how muchlower or higher price the new cards are comparedto the RTX 3000 series. Okay? Cool. We're gonna pin a comment below with the pricing thatDr. Su announced today. Now, RDNA 2 has been rumored, it's been speculated on, heck, I even played on thenew architecture in the Xbox, but now we finally got the details for the PC add-in cards, and the first thing we know is that AMD is using the same seven-nanometer process as their previous Radeon5000 series cards. Now, given that NVIDIAhad a performance lead before they moved to their new eight-nanometer node with Ampere, and considering the performancethey gained by doing so, AMD staying on their old process might not sound like itbodes well for the new cards. But they claim that, comparedto the Radeon 5000 series, they've pulled off up to200% better performance and 50% better performance per watt, which mostly just means thatthe power draw is the same. But still, considering that the higher-end RTX 3000 series from NVIDIA is chugging down power likeit's trying to drain a lake, not too shabby. How'd they accomplish all that? Well, unfortunately, AMDdidn't go into much depth, but they did say that theytuned their core design for frequency and power efficiency, which is, like, what the hellelse would you tune it for?

         You're not tuning itfor crappy clock speeds and crummy power consumption. So that's all marketing speak, but they did provide acouple preliminary details if we drill a touch deeper. The pipeline has beencompletely rebalanced with a focus on optimizing data paths. That could mean another update to the way that workgroupsare handled in parallel, or it could mean somethingentirely different. We're gonna have to get ourhands on them to know for sure. All we can say for nowis that AMD assures us that it's been designedfor efficiency first. Just like our water bottles. LTTStore.com. They keep it ice cold. Now, efficiency is good,but what other goodies do we have with the new architecture? For one, we're gettingfull ray-tracing support in the form of ray accelerators built into every compute unit, which not only allows the new GPUs to run any standardray-traced title to date, but also enables newfunctionality with FidelityFX, AMD's open source imageprocessing toolkit, including a denoiserfor ray-traced content and acceleration for the ambient occlusion and screen space reflection features that AMD's already using. AMD says that their ray accelerators are 10 times faster thanray tracing in software, though we don't reallyhave a baseline for that since AMD never enabled aray tracing API for RDNA 1. I have a good feeling about it though from looking at GTX cardsrunning ray tracing in software. Those features tie intoAMD's overarching support for DirectX XII Ultimate, which also includes thingslike variable rate shaders that dynamically reduce complexity for less frequently updatedportions of the screen, mesh shaders that combinemultiple shading passes at once, and of course DirectStoragesupport which is baked into both of the new RDNA2-powered consoles, sort of. The PlayStation implementationis not quite DirectStorage, which is a Microsoft tech. As a result then, AMD claimsthat because the consoles will be optimizing forexactly these features, cross-platform PC gameswill be tuned for them pretty much by default.

         It's obviously not quite that simple, but my hopes are still pretty high. AMD also offered us a glimpse at a feature they call Super Resolution, which, for all intents and purposes, sounds like their version of DLSS, except not using quite the same method. Now, AMD was hesitant to get into the details of it just yet, so make sure that you're subscribed so you don't miss ourreview with all the dirt. Finally, there's something that AMD is calling Infinity Cache, which I swear is not a cryptocurrency. It does sound like one though, doesn't it? Instead, Infinity Cache consists of 128 megabytes ofextremely fast on-die memory that, just like Ryzen's cache, keeps the GPU fed duringmemory-intensive operations, which AMD specifically saysminimizes DRAM bottlenecks. From AMD's slides, itlooks like this alone accounts for 40% of RDNA 2'soverall performance improvement and in theory it should allow RDNA 2 to maintain performanceat high resolutions despite the 256-bit GDDR6 not-X memory that AMD's using on allof its Radeon 6000 GPUs. And which GPUs are there? Well, got a couple for ya: the RX 6800 and 6800XT are the two GPUs that AMD is prepping to try and propel themselvesback to enthusiast status, with the RX 6800 completely dominating... Tomorrow's? You mean yesterday's. It launches tomorrow. Oh, right. Completely dominating tomorrow'sRTX 3070 in performance, if AMD's slides are to be believed, bearing in mind though that the 3070 is roughly equivalent to an RTX 2080 Ti. It really is no wonder NVIDIAcarpet bombed the news cycle. If these numbers are true, and we will of course confirm them before the simultaneous launch of the 6800 and 6800 XT on November 18th, I am officially telling you, dear viewer, to wait for RX 6800 before buying a 3070. And if you're one of themany waiting on an RTX 3080, the fact that the RX 6800 XT is expected to trade blows with it while drawing slightly less power may be enough to give you pause as well. But wait, what's that the performance slides say? Rage mode? Smart Access Memory?

         What the heck are these? Okay, well, Rage mode isbasically an easy mode way to bump up the power target and fan curves in a single click, although AMD admit that it's probably only gonna net you 1 to2% better performance. As for Smart Memory Access, well, here's where thingsget really interesting. AMD is leveraging their platform control with Ryzen 5000 and500 series motherboards by allowing those CPUs to directly address the newRadeon 6000 series' memory, all 16 gigs of it. Now, usually this has to bedone in 256-megabyte chunks in the form of the PCIExpress base address register, or BAR, which means that Ryzen plus Radeon allows for a level of optimization that simply isn't possibleon a traditional system using previous-gen Ryzen or Intel CPUs. Now, currently this isoptimized primarily for games, and while not all gameswill see a benefit, performance uplift according to AMD should be somewhere between6 and 10% for those that do, just from having a fifth-gen Ryzen CPU. Now I guess we've really got a reason to upgrade the LAN center, although I do have some discomfort about proprietary memory sharing nonsense and whether developerswill actually implement it and all that, but again, stay tuned for the full review. Oh yeah, and one more thing: who could forget thebig boy, the RX 6900 XT? Nice. An 80-compute unit Radeon GPU with performance that is allegedly on par with the 350-watt RTX 3090, yet with a total graphicspower of 300 watts and a power supplyrequirement of just 650 watts. That happens to be thesame power requirements as the RX 6800 XT andthe same core clocks. Clearly this is a really, really well-binned piece of silicon, if that is to be believed, and maybe that's whyAMD is giving themselves another three weeks to release it. Hopefully they don't haveto push it back any further to resolve supply shortages. (coughing) I'm sorry. Excuse me. I guess we won't know any more than this until we have the cards in our hands and we've seen them on store shelves, which is another three weeks out, but at that time we aregonna have a very good idea of how Ryzen 5000 and RTX3070 are gonna perform, so we're gonna have plenty of comparison points for our review. 

        Just like you're gonna have plenty of message from our sponsor, Ting. Ting does mobile phoneservice differently. There's no contracts, no overage fees and no other carrier tricks. You just pay a fairprice for the talk, text and data that you use every month, and it's especially greatif you're stuck at home using WiFi instead of mobile data. Why would you be paying forit if you're not using it? Ting doesn't think you should have to. And they give you completecontrol over your account. You can set alerts and capsfor each device on your account to keep your usage in check, and you get nationwideLTE coverage in the U.S. using T-Mobile and Sprint. Almost any phone will work with Ting, and you can check yourphone's compatibility at linus.ting.com. And you'll even get a bonuscredit for signing up. Thanks for watching, guys. Go check our launch reaction video for the Ryzen 5000 series to get an idea of what to look forward to before November's out. Dang! Techtober has not let me down. I mean, iPhone 12 was pretty boring, but PC stuff, yeah! We finally got some cool stuff! Thanks, AMD. And NVIDIA. Did Intel do anything noteworthy? Okay.