Unlike Cinebench, 720p@360hz is an actual real world scenario if somewhat niche
Would be nice to see AMD closing the gap in high fps gaming because that's pretty much the last bastion where Intel is strictly superior.
Nobody tests $700+ GPUs at 720p mate...i have not seen Techspot cracking out Quake 3 Arena for their RTX 3090 review.
Intel is around 5% ahead at 1440p and negligible at 4k due to the gpu bottleneck, there are some outliers which show a much bigger gap however.
Latency is the most important metric for improving Ryzen's gaming performance at this moment and Zen 3 should fix that.
10900K and 10700K
Engineering samples have reached 4.9 GHz
I expect better clocks, an improvement in this area is mandatory. But if so, why would AMD not clock the skus as high as possible? With the allowed and possible overclock being a lone core that may go 100MHz higher?
AMD Ryzen 9 5900X 12 Core & Ryzen 7 5800X 8 Core Zen 3 CPUs Could Potentially Launch As Early As October 20th
Unlike Cinebench, 720p@360hz is an actual real world scenario if somewhat niche
I totally agree. My point was that with per-core voltage regulation and stated engineering sample clocks it should reach ~ 4.9-5.0 on the best cores. Definitely not all cores we're lucky if all cores can sustain ~4.4 (based on the cpu-z leak I'm not sold)Matisse can't reliably reach it's max boost speed on all cores, even with top-notch ambient cooling (read: no chillers or anything else involved). I doubt we'll see 4.9 GHz all-core from Vermeer on any SKU. Feel free to prove me wrong AMD, but I am skeptical.
I totally agree. My point was that with per-core voltage regulation and stated engineering sample clocks it should reach ~ 4.9-5.0 on the best cores. Definitely not all cores we're lucky if all cores can sustain ~4.4 (based on the cpu-z leak I'm not sold)
ST result is undeniably strong but the MT ones (after we factor in ipc increase) reveal that for whatever reason this sample did not boost very high. Might be a cooling limitation however. We will see.
ST result is undeniably strong but the MT ones (after we factor in ipc increase) reveal that for whatever reason this sample did not boost very high. Might be a cooling limitation however. We will see.
ST result is undeniably strong but the MT ones (after we factor in ipc increase) reveal that for whatever reason this sample did not boost very high. Might be a cooling limitation however. We will see.
Yes, someone would buy a 3090 for 4K or 8K as you suggest, but testing at these resolutions, will only show that a 3300X has the same performance as a 10900K (-1,5% at 4K, same perf at 8K) The majority of the Tech press declared 10900K as the Gaming King and I would imagine that is based on their own experience and tests, but surely was not 4K average fps testing...Is this sarcasm? Are there actually people testing a 3090 with a 10900k at 720p? That is something like 1/36 of the 8k resolution that you are supposed be buying the 3090 for. What kind of frame rates are we talking there? You should not see 720p except in some cheap 13 inch laptop or something.
I am of the opinion that such low resolution test are testing raw memory latency, and that is about it. You might as well run some synthetic memory testers. Although, there is a possibility that it is partially the different cache architectures between AMD and Intel. Running through the scene data may cause the cache to cycle more in Zen 2. I think there was something about a single core only being able to access 8 MB of the 16 MB before the latency increased; I don’t have time to look that up now. AMD will certainly have worked on this in Zen 3. Even if Zen 3‘s raw latency is still higher, which I expect it will be, Zen 3 average latency due to massive cache improvements may be lower. AMD would probably be able to achieve similar raw latency with a monolithic APU, but not an MCM.
This is the crux of the problem: AMD needed to have stockpiled these chips way in advance or else the demand for them will outpace supply. TSMC N7 is in high demand and pretty much every AMD product line uses it.We already have websites who tested RTX 3080 from 1080p to 4K on 3900XT and OCed 10900K platforms. At lowest (1080p) OCed 10900K is around 10% faster, at 1440p it's ~7.5% faster and finally at 4K there is virtually no difference. Zen3 will have no problems offering better performance than intel parts at low(er) resolutions while most likely being GPU bound at 4K+. There will be no reason to buy intel after October 20th, unless there are supply issues on AMD side.
Actually quite a big leap in IPC over Tiger Lake in this one benchmark only. 1185G7 scores approximately 600pts here, almost exactly 20% over 1065G7.It was the time for AMD after 10+ years to push IPC and clock over intel
Now we gona wait how long intel need with the tiger lake to get the desktop/server footing
I believe AMD has more dedicated resources for SMT, but could this be the symptom of a stronger core leaving lesser resources for SMT ala Intel? I can't wait to see the SMT scaling tests from The Stilt.That could be the case. It could also be that Zen 3 makes significant improvements in the core, to caches, to scheduling, prefetch, you name it, improving core effectiveness at the cost of SMT scaling.
Expected CPU clocks for Zen 3, for example stock R5 3600X or XT version has 4.2ghz All core Turbo.
Wow! That's quite bold. I'll have to commend you for sticking your neck out, though.Zen3 will have no problems offering better performance than intel parts at low(er) resolutions while most likely being GPU bound at 4K+. There will be no reason to buy intel after October 20th, unless there are supply issues on AMD side.
Yes, someone would buy a 3090 for 4K or 8K as you suggest, but testing at these resolutions, will only show that a 3300X has the same performance as a 10900K (-1,5% at 4K, same perf at 8K) The majority of the Tech press declared 10900K as the Gaming King and I would imagine that is based on their own experience and tests, but surely was not 4K average fps testing...
I mentioned the new GPUs mainly to highlight that 1% of the miniscule advantage Zen3 will have vs 10900K at this test, will also be due to PCI-Expess 16X gen difference.
720p testing shows some indications when GPU is not the limit and is useful if you know what to take from it (depending on the game/level used etc). Also why the argument regarding latency/cache/CCX etc, for whatever reason an architecture is better at something than another, in the end it is better. It is like saying that the Rendering superiority of the Ryzen does not count because mainly it is from SMT vs HT performance difference, does it change the outcome that Ryzen is better at Rendering? Also i would imagine that most of the game engines are more optimized for Skylake than Ryzen architecture, in the future with the new consoles and with Intel stepping away from Skylake this will change also.
Wow! That's quite bold. I'll have to commend you for sticking your neck out, though.
This is the crux of the problem: AMD needed to have stockpiled these chips way in advance or else the demand for them will outpace supply. TSMC N7 is in high demand and pretty much every AMD product line uses it.
Wow! That's quite bold. I'll have to commend you for sticking your neck out, though.