E3 has n’t been just for games for a while . While the TV games are certainly the focus , society like AMD are prompt us that what we spiel on is sometimes crucial to what we play . Today the company herald a whole line of CPUs and GPUs intended to make gaming better , and the affair they all had in common ? They run on 7 nm .
Do you screw how revolutionary 7 micromillimeter is ? Do you hump WHAT 7 Land of Enchantment is ? Heck , do you even know what nm is ? ( It ’s the abbreviation for nanometer . ) Processor makers like Apple , Qualcomm , and Intel have been talking a lot about nanometre over the last few eld . The nanometers specifically touch on to gate distance of the junction transistor . The smaller the logic gate duration the more transistors you’re able to correspond on a dice . As the ability to cringe logic gate length every year , and get the accompanying hike and performance and push efficiency , has become more difficult , every nanometre has initiate to count . So Apple crows about the 7 nm CPU in the iPhone XS , andIntel talks a lotabout the 10 nm parts in its 10th - generation of CPUs . And AMD babble about 7 micromillimetre in its raw CPUs and GPUs . The troupe has been call for a while to add 7 nm to desktops and laptops we apply day by day , and with Zen 2 and RDNA , it ’s render on that promise .
To be clean , this is n’t the first time AMD relinquish a 7 nm part . Back in January , it found the Radeon VII , a 7 nm GPU base on its older Vega architecture . The new 5700 and 5700XT are based on the brand young RDNA computer architecture that will also be appearing in the next - gen PlayStation and Xbox consoles set to be released next class . ( While many producer and observers use RDNA and Navi interchangeably , AMD has told me it prefers the great unwashed use Navi when speak of specific GPUs and RDNA when tattle about the overall architecture . ) But when talking about the 5700s AMD still spend a lot of time speak about 7 micromillimeter , because , at least from the benchmarks it showed off , this material is seriously fast .
Photo: Alex Cranz (Gizmodo)
AMD cites the move to 7 millimicron as a major intellect why the 5700s should , theoretically , use less power to get full performance than their Nvidia contender . In slides comparing performance , the $ 450 RX5700XT was play off against the $ 600 Nvidia RTX 2070 ( though the 2070 can be found for as low as $ 480 ) , and the $ 380 RX5700 was pit against the $ 350 RTX 2060 . Notably , the AMD GPUs do not do ray trace . The Nvidia ace do .
According to AMD , the RX5700XT has a die that ’s half the size last generation ’s Vega 64 , and that it has 14 percent estimable overall performance while using 25 percent less power . Simon Ng , one of RDNA ’s project director , told me it was a 1.5 times honorable performance per W .
And the slide that AMD presented to outline its performance equate to its rival 2070 paint a picture it is impressively tight . The same can be said for the RX5700 versus the 2060 .
Photo: Alex Cranz (Gizmodo)
The RX 5700 , in exceptional , seems to be a better bargain than its rival , with double - finger’s breadth best performance on four of the ten games tested , and single finger improvements on the other six . All while being only $ 30 more . If you’re able to survive without the beam decipher backup , the RTX 2060 provides , the RX 5700 seems like a good deal . The RX 5700XT is also a good deal when you consider its public presentation is on equation with the RTX 2070 , but it costs $ 150 less .
AMD also detail a few bell and whistle coming to the 5700 line , despite a few mocking references it made towards Nvidia , which loves to harp on the doorbell and whistles when announce component .
There ’s Fidelity FX , its resolution toNvidia DLSS . DLSS uses neuronal connection and rich learning to meliorate performance and graphics on Turing - found GPUs from Nvidia . Fidelity FX should better performance and graphics along similar lines , but by being an opened source set of tool available to anyone .
Image: (AMD)
Another tool mean to improve graphics is Radeon Image Sharpening . RIS will let programmers have sharper images in background signal asset . Ideally , so things look sharp and less blocky than in many current game . RIS is basically a super lightweight post - processing shader . Usually , such a shader would be tax , but accord to AMD , turning on RIS will make a minimal slowdown . On average it should be less than a 2 - per centum operation come to versus playing the same biz with RIS off .
lastly , there ’s Radeon Anti - Lag . RAL is intended to trim down the stave that comes from a GPU rendering what ’s on the cover . In a demonstration of Borderlands 3,AMD show up an betterment of about 15ms with RAL turn on . 15ms is about the duration of metre a single frame is onscreen . So that ’s about a 1fps improvement . Admittedly , that ’s not necessary for most of us , but crucial for competitive gamers who ask every last physique they can get .
Today AMD also spoke more about the third generation of Ryzen processor it foretell at Computex in May . The third generation of Ryzen CPUs might fuddle you a little because they ’re free-base on a novel computer architecture called Zen 2 — which is n’t the second genesis of Zen , but the third .
Five processors were announced at Computex . have ’s just heel them out and get through the alphabet soup as speedily as possible . They ’re the $ 500 12 - core , 24 - train of thought , 105W Ryzen 9 3900X , the $ 400 8 - core , 16 - yarn , 95W Ryzen7 3800X , the $ 330 8C/16 T 65W Ryzen 7 3700X , the $ 250 6C/12 triiodothyronine 95W Ryzen 5 3600X , and the $ 200 6C/12 T 65W Ryzen 5 3600 . Those ware roughly liken to the i9 , i7 , and i5 kB - series CPUs from Intel , though the Intel product have a high-pitched MSRP , usually around $ 100 more .
of course , AMD claims its processors are much faster than Intel ’s while using less force . It claim the 3900X has 14 percent sound single thread carrying out than Intel ’s i9 - 9920X and 6 percent upright multithread performance while requiring only 105W of power versus the i9 - 9920X ’s 165W necessary . This is because Zen 2 , like RDNA , is based on a 7 millimicron process . Which means datum ask less energy to move a shorter distance quicker than the larger 14 nm central processing unit Intel ’s current desktop CPUs use .
It ’s also because AMD has double the size of the L3 cache . That ’s unspoilt because L3 hoard is where the most crucial and immediate data is stored that your processor needs to record . When it fill up , the data gets give up over to the RAM . read data from RAM is fast , but not about as fast as from the L3 cache . The smaller stash of the second gen of Ryzen processors lead to dull single core public presentation and overall meantime in L3 cache memory intensive work like video biz . By doubling the cache , AMD is desire it can beat Intel , which has traditionally had much full execution on games .
The newfangled Ryzen processors will mostly be available July 7 , 2019 . The only one not being released next month is the surprise Ryzen 9 3950X , which is due in September 2019 . That part was foretell today by AMD . It ’s another 105W CPU . Only instead of 12 nub and 24 thread — the most encounter in a background CPU mean for gaming , the 3950X has 16 core and 32 threads . Intel has nothing like it available at that size of it or power requisite .
And that ’s kind of the full point . As you ’ll mark in the above benchmark slides , even by AMD ’s own title , it is not always winning against the competition when the comparability is pith to core or compute whole to calculate unit . Intel and Nvidia have been dominant in the CPU and GPU fields respectively because they make infernal good parts . AMD ’s benefit is that it can give user a fate more raw power at a much lower price . The Ryzen 3950X is emblematic of that .
But as we ’ve seen repeatedly when reviewing AMD product , it ’s not just about the raw power . kinship with software system makers matter . Nvidia and Intel have good documentation and , at least in the PC infinite , wider adoption . So carry to see more from AMD about its partners in the package and hardware humanity . And especially , expect to hear more about how it ’s in about all of the next generation of cabinet .
Correction 7/11/19 10:22am EST : This web log antecedently said nanometers refers to die size . That is incorrect . It refers to the gate distance of transistors . We ’ve corrected the text to reverberate that and profoundly apologize for the misplay .
12:39pm EST : While Nvidia list the 2070 GPU as usable for $ 600 , Nvidia arrive at out to note the 2070 can be regain for as low as $ 480 from many retail merchant . We ’ve update to showcase the spacious range of toll .
Na’vi
Daily Newsletter
Get the best technical school , science , and refinement news in your inbox daily .
News from the future , delivered to your nowadays .
Please choose your desired newssheet and put in your email to upgrade your inbox .
You May Also Like
