Friday, March 3rd 2017
AMD Responds to Ryzen's Lower Than Expected 1080p Performance
The folks at PC Perspective have shared a statement from AMD in response to their question as to why AMD's Ryzen processors show lower than expected performance at 1080p resolution (despite posting good high-resolution, high-detail frame rates). Essentially, AMD is reinforcing the need for developers to optimize their games' performance to AMD's CPUs (claiming that these have only been properly tuned to Intel's architecture). AMD also puts weight behind the fact they have sent about 300 developer kits already, so that content creators can get accustomed to AMD's Ryzen, and expect this number to increase to about a thousand developers in the 2017 time-frame. AMD is expecting gaming performance to only increase from its launch-day level. Read AMD's statement after the break.
AMD's John Taylor had this to say:
"As we presented at Ryzen Tech Day, we are supporting 300+ developer kits with game development studios to optimize current and future game releases for the all-new Ryzen CPU. We are on track for 1000+ developer systems in 2017. For example, Bethesda at GDC yesterday announced its strategic relationship with AMD to optimize for Ryzen CPUs, primarily through Vulkan low-level API optimizations, for a new generation of games, DLC and VR experiences.
Oxide Games also provided a public statement today on the significant performance uplift observed when optimizing for the 8-core, 16-thread Ryzen 7 CPU design - optimizations not yet reflected in Ashes of the Singularity benchmarking. Creative Assembly, developers of the Total War series, made a similar statement today related to upcoming Ryzen optimizations.
CPU benchmarking deficits to the competition in certain games at 1080p resolution can be attributed to the development and optimization of the game uniquely to Intel platforms - until now. Even without optimizations in place, Ryzen delivers high, smooth frame rates on all "CPU-bound" games, as well as overall smooth frame rates and great experiences in GPU-bound gaming and VR. With developers taking advantage of Ryzen architecture and the extra cores and threads, we expect benchmarks to only get better, and enable Ryzen excel at next generation gaming experiences as well.
Game performance will be optimized for Ryzen and continue to improve from at-launch frame rate scores."
Two game developers also chimed in.
Oxide Games, creators of the Nitrous game engine that powers Ashes of the Singularity:
"Oxide games is incredibly excited with what we are seeing from the Ryzen CPU. Using our Nitrous game engine, we are working to scale our existing and future game title performance to take full advantage of Ryzen and its 8-core, 16-thread architecture, and the results thus far are impressive. These optimizations are not yet available for Ryzen benchmarking. However, expect updates soon to enhance the performance of games like Ashes of the Singularity on Ryzen CPUs, as well as our future game releases." - Brad Wardell, CEO Stardock and Oxide
And Creative Assembly, the creators of the Total War Series and, more recently, Halo Wars 2:
"Creative Assembly is committed to reviewing and optimizing its games on the all-new Ryzen CPU. While current third-party testing doesn't reflect this yet, our joint optimization program with AMD means that we are looking at options to deliver performance optimization updates in the future to provide better performance on Ryzen CPUs moving forward. "
Source:
PC Perspective
"As we presented at Ryzen Tech Day, we are supporting 300+ developer kits with game development studios to optimize current and future game releases for the all-new Ryzen CPU. We are on track for 1000+ developer systems in 2017. For example, Bethesda at GDC yesterday announced its strategic relationship with AMD to optimize for Ryzen CPUs, primarily through Vulkan low-level API optimizations, for a new generation of games, DLC and VR experiences.
Oxide Games also provided a public statement today on the significant performance uplift observed when optimizing for the 8-core, 16-thread Ryzen 7 CPU design - optimizations not yet reflected in Ashes of the Singularity benchmarking. Creative Assembly, developers of the Total War series, made a similar statement today related to upcoming Ryzen optimizations.
CPU benchmarking deficits to the competition in certain games at 1080p resolution can be attributed to the development and optimization of the game uniquely to Intel platforms - until now. Even without optimizations in place, Ryzen delivers high, smooth frame rates on all "CPU-bound" games, as well as overall smooth frame rates and great experiences in GPU-bound gaming and VR. With developers taking advantage of Ryzen architecture and the extra cores and threads, we expect benchmarks to only get better, and enable Ryzen excel at next generation gaming experiences as well.
Game performance will be optimized for Ryzen and continue to improve from at-launch frame rate scores."
Two game developers also chimed in.
Oxide Games, creators of the Nitrous game engine that powers Ashes of the Singularity:
"Oxide games is incredibly excited with what we are seeing from the Ryzen CPU. Using our Nitrous game engine, we are working to scale our existing and future game title performance to take full advantage of Ryzen and its 8-core, 16-thread architecture, and the results thus far are impressive. These optimizations are not yet available for Ryzen benchmarking. However, expect updates soon to enhance the performance of games like Ashes of the Singularity on Ryzen CPUs, as well as our future game releases." - Brad Wardell, CEO Stardock and Oxide
And Creative Assembly, the creators of the Total War Series and, more recently, Halo Wars 2:
"Creative Assembly is committed to reviewing and optimizing its games on the all-new Ryzen CPU. While current third-party testing doesn't reflect this yet, our joint optimization program with AMD means that we are looking at options to deliver performance optimization updates in the future to provide better performance on Ryzen CPUs moving forward. "
126 Comments on AMD Responds to Ryzen's Lower Than Expected 1080p Performance
Your a smart dude, I like to think. Why even look at 1700 vs 7700k if the primary reason is gaming. That mindset I don't understand.
The IPC difference is 6-8% give or take at clock. Thats going to continue down the line 7, 5, 3. I personally don't think they have room to have higher clocks. Its GLOFD. That's why they are priced they way they are, bang for the buck.
You don't know me or my usage so back up.
And check the badges for an example.
I've been having the same i5 argument a while and it doesn't change,play old dx11 era games go i5, see mostly new era games, don't get an i3.
You seen gtaV running on crossfire 480s on many i3 or i5 pcs, I expect my present PC would beat them both clocked at 5ghz running 4k as I am, ultra settings .
So there ya go my use cases this eve on one page and my use cases state only i7 or ryzen is a step up, I'm not going to put cash in a clowns pocket though ,no.
6 cpus
1.39%
7 cpus
0.00%
8 cpus
0.24%
It looks like ~1.63% of the users on steam have cpus with more than 6 cores, which would normally be called enthusiast, however I'm afraid that includes all the 6 and 8 core FX-es and also some older parts like the 6 core phenoms. It looks like the enthusiast market at least for gaming is very close to 0. Maybe 0.5%.
Will it be worth to pay double for maybe 10% better gaming performance?
Hopefully by then they'll also resolve the glitches the platform has.
I've been so tempted to pull the trigger on both options, but I'm gonna wait for the dust to settle and revisit the options in a couple of months.... patience is a virtue after all.
For a company as small as AMD, yes I am comparing it to their previous architecture and it is significantly better. It is a huge generalational leap and better than they were telling us it was going to be. Gaming performance, at settings people will actually play at is negilible, unlike the FX line which suffers from terrible minimum frame rates and high frame latency. If you want to base a purchasing decision off of game performance at setting you will not be playing at, be my guest, to each their own. AMD's biggest problem is not gamers buying 8 core CPU's, the rest of the Ryzen lineup has yet to be released. Also, things need to be put into perspective, AMD does not have the R&D budget intel has, but yet look at what they have accomplished. Is it really fair to say that Ryzen just sucks at gaming because it doesn't match a quad core CPU clocked at 4.2GHz base that has its architecture based as far back as nehalem, when they first introduced the Itegrated mem controller? This is the first revision of a from scratch design that is already much better off than they were with bulldozer. If it is not for you, great, that doesn't mean it is not for anyone.
For my ten pence worth on the actual issue,I think the change to a write back cache is affecting the max frame throughput of the core in some game engines optimised to use write through caches.
AMD's slightly inferior single core capabilities are not the problem. Compilers that can direct the workload to a single core or multiple cores, for that matter, that are not working in an efficient manner is the problem.
Oxide is actually working on getting the workload to distribute to multiple cores as Ashes of Singularity performed like crap on Ryzen in spite of AMD's apparent multi core superiority, or equivalency, depending on which camp you fall into.
This is similar to games being optimized for Nvidia hardware resulting in AMD having to create post release drivers to overcome the advantage Nvidia had. Game developers have been optimizing for Intel CPUs for a long time since, quite frankly, there was no reason to put much effort into optimizing for AMD's FX line after AMD quit updating it after year one.
Between AMD's efforts and developer efforts the delta between Intel and AMD will narrow. However, as we see in the GPU realm, there will be developers "persuaded" to not make an effort to optimize for AMD and the burden will be on AMD to find workarounds to close the performance gap.
However, as AMD's market share grows it will take more persuasion to coerce developers to ignore AMD optimizations in the future.
Gaming was never gonna be the catalyst for AMD to grow market share in the 8 core market. It's the content creators that are going to drive that and some of them are gamers. As their numbers grow, they will become a big enough market for the developer holdouts to accept AMD CPUs as a viable market to optimize for.
Time will fix this issue and the performance of these new Zen CPU's will only get faster over time.
I myself eye'n a 1600X or non-X if there is one.I didn't see those posts. I'm in the same situation but I don't see much benefit to going full 8c at the moment for my use. My position is i'll see how this plays out (things get ironed out) the next couple of months.
I am just gonna leave this here,
all credit to this reddit: www.reddit.com/r/Amd/comments/5xcc0r/jokers_bench_showing_7700k_cpu_bottleneck/
BTW, don't anger the 7700K fanboys. How dare you insinuate that it can cause fps drops or stuttering?!
And development should move towards utilizing more cores?
Quad core is so old hat now, AMD are on the right road.
Also, this will hopefully force Intel to pull up their socks... improvement between 6700k and 7700k(for example) was quite sad and far from worth the price/cost.
That's mean 20% Delta behind Kapy-Lake in single core performance.
*These information from AMD not from my pocket*
Coffee-Lake should as usual bring 5% in IPC 10% in IGPU 15% in total , that's mean Ryzen will be 25~30% Delta behind Coffee-Lake single core performance and that's huge deal in games.
Coffee-Lake also will bring 6C/12T and
24~30 PCI-E3.0 lanes to mainstream, plus the IGPU will support 4K HD10, Dolby Vision [H.265 ,Vp9 4K 12bit Encode and Decode].
AMD did great job with Ryzen but they have a lot of work to do with Zen2.
First they have to close the IPC gab between Zen and Coffee-Lake (10% higher IPC will be really good news).
Second they have to close the gap between Zen and Coffee-Lake in clock speed (10% will be amazing)
[That's will but Zen2 just 8~10% Delta behind Coffee-Lake in single core performance] .
Third they have to add the missing features from Zen architecture like AVX256, 28~40 PCI-E 3.0 lanes.
Fourth they have to fix memory bandwidth issue from what AMD said it's look like they will not add Quad channel memory anytime soon, but they can increase the bandwidth buy supporting higher RAM speed out of the box 3200 MHz with two Dimm slots and 2699 MHz with Four Dimm slots that's will bring huge improvements to the performance plus support RAM OC up to 4000 MHz.
AMD have to do that next year to close the gap because in 2019 Intel will release Ice-Lake with new architecture using 10nm, we will see at least 10% higher IPC than Coffee-Lake, that's will but Ryzen 40% Delta behind Ice-Lake in single core performance, just in two years With support for DDR5 and DDR4, PCI-Express 4.0, and many other features, not to mention Optan X will be available for consumers.
In 2019 AMD should release Zen3 on 7nm FinFET not on 14nm FinFET and add all the features from Ice-Lake and keep the 5~8% Delta behind Intel in single core performance.
AMD also have two years to work with games developer to support Ryzen 8C/16SMT, 6C/12SMT maybe also 12C/24SMT to make sure First generation Ryzen will not be end up 40% behind Intel Ice-Lake in games, that's possible specialy if Xbox Scorpio will end up using Ryzen CPU that's will help AMD a lot in optimization issues.
[if AMD will not bring down the 40% Delta gab in performance between Ryzen and Ice-Lake, most of the people will jump in Intel train and will never go back to AMD train even if they will offer Quantum computer for free for each person bought Ryzen in 2017-2018].
We know also Tiger-Lake in 2021 will be the last Intel (Cor I7,I5,I3) architecture after that they will move to new architecture from the ground, AMD have a lot of work to do to keep up with Intel.
AMD APU's before end of the year should give us big example about how AMD will fix the RAM bandwidth issues, if they will use HBM2 to feed the CPU and GPU they should do that with Zen 2 as L4 cash to avoid all the problems from RAM speed, timing and channels.
If they will use DDR4 Dual channel memory to feed the APU, like what they did with old APU's and DDR3 single channel and dual channel it will be epic fail to AMD in performance. Buy the way AMD APU's still better than Intel IGPU but both suffered from RAM bandwidth limitations, Intel still has the problem.
Coming months will give us some answers about next Zen architecture. [Just hope it's good not bad].
Previously published prices suggest ~260$ for top line 6 core part, ~200 for the 4 core 8 thread and ~150$ for the 4 core 4 thread.
I'm not sure what to make of it for now, I think, non GPU bound, gaming price/performance ratio vs intel will still be good, probably with 60% of the price you will buy 75% of performance, with few exceptions for games properly optimized for MT.
Nevertheless for pure best gaming performance intel will still be the best choice.
Let's see what the real pricing will be and how the things will unfold.
Anyway, in general without discussing specifically about Ryzen, when benchmarking games on various CPUs, I think review sites should consider using lesser GPUs for better consumer information.
With 1080 GTX , probably 7700 will be on top, with 1060 GTX probably all of them will be clumped together and maybe even the lowest i5, the 7400 or the lowest Ryzen, will provide adequate performance.
I've never seen any games optimized for Intel, as a matter of fact games commonly contain some of the worst CPU code. The reason why Intel wins is their prefetcher is better at handling crappy code.Imagine if AMD spent this kind of resources on designing a god prefetcher for their CPU…During gaming all the CPUs will boost, AMD is not far behind in clock speed, if it's not ahead.
Ryzen 7 1800 X boost beyond 4.0 GHz
i7-7700K boosts to 4.5 GHz
i5-7600K boosts to 4.2 GHz
i7-6950X boosts to 4.0 GHz
i7-6900K boosts to 4.0 GHz
i7-6800K boosts to 3.8 GHz
Yet all of these Intel CPUs have marginal differences in games, while Ryzen struggles in a number of games. Something tells me that it's not just a lack of clock speed. We already know there is little gains for Intel beyond 4.0 GHz, so AMD would have to do something with their prefetcher.
-----
Well, we all knew this were going to happen. AMD did a decent job by building a more superscalar processor, but they didn't prioritize building a proper front end/prefetcher. Their prefetcher is worse the one in Sandy Bridge, and considering that most of the improvements from Sandy Bridge to Kaby Lake is in the prefetcher, they have some serious catching up do to.
The efficiency of the prefetcher matters a lot for some workloads, including gaming. And when it comes to cache misses, increasing the clock frequency wouldn't help mitigate the performance penalty.
It's not like this "problem" is going to blow over. It might not matter to a GTX 1060, but when Ryzen is too slow to saturate a GTX 1080, things are only going to get worse with GTX 1080 Ti, Volta, etc. For buyers of GTX 1070 or higher the first generation Ryzen is simply too slow. For gaming a i7-6800K is a better deal, even if Ryzen 7 1800X beats it in some workloads.