AMD put on one hell of a show - showstopper when it revealed all the details aboutits next - gen processorsearlier this month . And it ’s done it again with its full - reveal of its next - gen graphics notice today . This may very well be the generation of AMD ironware that in conclusion put it cervix - and - cervix with its competitors , not just slightly behind in a tightlipped second .
But not content to just talk about its coming nontextual matter cards , which have some interesting architectural changes , AMD also announced a shipload of raw software system and features that strengthen its ecosystem — and maybe even volunteer a little inducement to go all AMD on your next build .
To start , here ’s the good material :

Screenshot: AMD
Radeon RX 6900 crosstalk : 80 compute units ; 2015 MHz game clock ( 2250 MHz boost ) ; 128 MB infinity cache ; 16 GB GDDR6 memory ; 300W TBP ; $ 1,000
Radeon RX 6800 XT : 72 compute units ; 2015 MHz game clock ( 2250 Mc boost ) ; 128 MB infinity cache ; 16 GB GDDR6 remembering ; 300W TBP ; $ 650
Radeon RX 6800 : 60 compute units ; 1815 MHz biz clock ( 2105 MHz boost ) 128 MB infinity cache ; 16 GB GDDR6 memory ; 250W TBP ; $ 580

Screenshot: AMD
Both the RX 6800 XT and 6800 will be uncommitted starting November 18 , with the RX 6900 XT follow on December 8 .
Before I start up comparing the RX 6000 cards to the RTX 3000 cards , let ’s tattle architecture and software . Other than take a crap the compute units 30 % more index - effective , and increasing the frequence by 30 % ( which is no minuscule feat in itself ) , with its RDNA2 architecture , one of the big change AMD is ushering in with its RDNA2 computer architecture is what it calls Infinity Cache . It ’s based on what AMD has done with high - denseness L3 caches in its Zen intersection line and adapted for its new GPUs . basically , this is a small second of extra storage space , 128 MB , that the GPU can tap into on top of the already uncommitted bandwidth on the card itself , devote up to a 0.9x ability cost increase and up to a 2.17x bandwidth rise , according to AMD .
These new GPUs will also have computer hardware - accelerated shaft of light tracing , and support for variable rate blending , mesh shaders , and sampler feedback — all the same advanced graphical feature of speech Nvidia supports with its RTX 3000 series cards . In addition to supportingDirectX12 Ultimate , which makes it sluttish for game developers to port their plot between PC and cabinet ( important for AMD to have since its GPUs are in the new PS5 and Xbox Series X / S ) , AMD will also support DirectStorage API in the future tense , which it says will improve how debauched games load from SSDs . Combined with PCIe 4.0 store and a motherboard that corroborate the new standard , it sound like it ’s a decent improvement , but AMD did not go into contingent about how much firm it could be .

Screenshot: AMD
AMD is bringing back Rage Mode , too , or one - click overclocking , which harkens back to the ATI GPU day . All novel Radeon graphics carte will come with a software program packet that allows you to overclock your identity card with a single computer mouse click . It ’s fundamentally an auto - overclocking feature , so if you do n’t want to mess around in your BIOS or any other tools , you do n’t have to . This feature will be useable at launching .
Lastly , AMD is introducing what it anticipate Smart Access Memory . If you build a scheme with an AMD Ryzen 5000 processor , plus an AMD - compatible motherboard , plus an AMD Radeon 6000 graphics menu , your CPU will be able-bodied to access all of the GPU ’s 16 GB of memory and frailty versa . This feature will need to be enabled on the motherboard chipset , but it allows the CPU and GPU to talk like a shot to one another . This allows your system to rely less on its RAM for various labor , which should improve loading time according to AMD .
All that said , the RX 6900 XT , RX 6800 XT , and RX 6800 are positioned to seriously compete withNvidia ’s RTX 3000 - seriesline - up , even without pay much attention to the gameplay stats in AMD ’s presentation . First and foremost , there are some language differences in the elbow room both AMD and Nvidia break down the specs of their graphics cards . CUDA cores are n’t lineal eq to work out units , and pedestal filaria are n’t needs the same affair as game filaree , it can be a minuscule challenging to equate both troupe ’ add-in by numbers alone . If you ’ve been succeed both companies for at least the last two twelvemonth , then this will probably be old intelligence to you .

Screenshot: AMD
AMD introduce its biz clock concept with the declaration of its RX 5000 - series . Instead of foreground the sheer gloomy relative frequency of its nontextual matter scorecard , it rather now highlights the low-pitched frequency users can expect to get while gaming . It ’s kind of a merchandising tactic more than anything else , but it ’s significant to note that biz clock and substructure clock are not the same thing , while boost clock has the same meaning for both AMD and Nvidia . So just keep those thing in idea when comparing basis clock and plot clock .
Compute units ( or stream processors ) versus CUDA Cores gets a petty more cunning . Because every GPU manufacturer defines its computer architecture in its own mode , there really is n’t a comparison between these two . Even if you get down to roughly the same issue of “ burden , ” literal performance can take issue a little bit or a lot . Which is why gaming benchmark or other execution measures are of the essence when compare two bill of fare from different industry .
However , AMD says its raw Radeon RX 6800 XT will get tantamount or better execution in most games than Nvidia ’s RTX 3080 due to its RDNA 2 computer architecture and other features outlined above . That ’s at 4 honey oil and 1440p , too . Same with its other 6000 - serial GPUs . And with Rage Mode and Smart Access Memory change state on , AMD sound out you ’ll get up to an extra 8 % frame pace bump . AMD did n’t meet on what itsray trace performancewill count like , but it could very well equal what Nvidia has been able-bodied to do . With the toll of these identity card all undercutting Nvidia ’s rivals , I think I ’m even more excited to get my hands on these notice than I was the RTX 3080 and RTX 3070 .

GPUs
Daily Newsletter
Get the dear technical school , science , and culture news show in your inbox day by day .
News from the future , give up to your nowadays .
You May Also Like














![]()