AMD's Future Chips & SoC's: News, Info & Rumours.

Page 86 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

jdwii

Splendid
At this point i'm pretty darn sure its coming out at computex at least their 8 core models.

I personally don't care much for their higher core count models gonna stick with 8 cores for a little while longer.
 

digitalgriffin

Distinguished
Jan 29, 2008
421
63
18,870
3


It's time Intel got a kick in their complacency for delivering mediocre speed increases with no price breaks for over 10 years now. I'm not anti Intel. I'm pro competition and pro value driven.

I also welcome Intel to the GPU fight. If they can deliver value, I'll buy them. But as a whole I'm more looking forward to the downward price pressure this will put on the GPU market as a whole.
 
Reactions: rigg42
Probably you have already read that.
https://www.tomshardware.com/news/amd-epyc-radeon-frontier-exascale-supercomputer,39275.html

The fact that both AMD CPUs and AMD GPUs have been selected to power most powerful supercomputer in the world, is a clear evidence that upcoming AMD products are really competitive.
To be fair, for those applications you care more about power per core rather then pure number crunching power, as you are aiming to install the maximum amount of cores possible.
 

InvalidError

Titan
Moderator
To be fair, for those applications you care more about power per core rather then pure number crunching power, as you are aiming to install the maximum amount of cores possible.
It is more about total throughput per watt than just core count. Having twice as many cores for the same performance and power budget is pointless since you now have twice as many threads to manage, more than twice as much inter-process communication overhead to optimize and it is this overhead that ultimately dictates how many threads the algorithm can hypothetically scale to. (I say hypothetical because the practical limit where incremental cost far exceeds incremental performance will come long before that.)
 
Reactions: digitalgriffin

Aspiring techie

Reputable
Mar 24, 2015
816
5
5,365
118
Yeah, 5 GHz sounds a bit high. But, 4.7 doesn't sound unreasonable.
14nm to 12nm saw a 200 MHz improvement. GloFo 12nm to TSMC 7nm should see a greater improvement, so 4.6-4.8 turbo sounds pretty reasonable.
Either way, a 15% increase in IPC plus a max turbo of around 4.7 GHz would be enough to win the single thread performance crown.

I do hope we're all wrong, though. 5GHz would be enough to keep Intel at bay for quite a while.
We'll just have to wait and see.
 
Reactions: prtskg and rigg42

rigg42

Upstanding
Oct 17, 2018
286
89
290
6
100% believe this unlike the other 5ghz 16 core rumors lol

https://www.tomshardware.com/news/amd-ryzen-3000-series-16-core-cpu-specs,39304.html

I also beleive the 4.5ghz turbo speeds motherboard manufacturers received from 4 core samples.

Max turbo will be 4.5-4.7 and max OC will probably be 4.4 on most chips. I expect a 10% IPC increase on average.

All we know is quad core engineering samples are hitting as high as 4.5 boost. We don't even know if that's all core or single core boost or how recent any of the samples are. I wouldn't base too much off this info. The clock speed on 1st gen Ryzen engineering samples were like 700mhz slower than final clock speeds. The TSMC 7nm process is capable of producing 5ghz clock speed cpus if designed with enough L1 cache from what I understand. We should know more in 3 weeks. Can't wait! Getting bored with the guessing game. With best case scenario of %30 IPC gain I'm guessing 15% minimum. IPC is hard to gauge since its load dependent. Intel only has an IPC advantage now in certain scenarios. It just happens that one of those scenarios is most current and older video games.
 

Aspiring techie

Reputable
Mar 24, 2015
816
5
5,365
118
Just some napkin math. If Zen 2 ends up hitting 5 GHz with a 15% IPC increase, then the theoretical single thread performance would increase by 37%.
If it hits only 4.7 GHz with a 10% IPC increase, then theoretical ST performance increase would be 29%.
Of course, this doesn't translate directly into real world performance, but in either of those scenarios, it should be enough to beat Intel at gaming and pretty much everything else.
It's a good time for AMD's CPU division.
 
Reactions: rigg42

rigg42

Upstanding
Oct 17, 2018
286
89
290
6
Just some napkin math. If Zen 2 ends up hitting 5 GHz with a 15% IPC increase, then the theoretical single thread performance would increase by 37%.
If it hits only 4.7 GHz with a 10% IPC increase, then theoretical ST performance increase would be 29%.
Of course, this doesn't translate directly into real world performance, but in either of those scenarios, it should be enough to beat Intel at gaming and pretty much everything else.
It's a good time for AMD's CPU division.
Just wait until AAA developers start coding for the new ryzen based game consoles. Anybody who decided to buy 4 thread and 6 thread intel CPU's for the better game performance is going to be regretting that. The 4c/4t quad cores are already losing to the 2600 slightly in a lot of games with an 800mhz clock speed advantage. Even in games where it does beat it its usually so insignificant it's not really an argument. Sorry folks the 4c/4t CPU is dead for gaming in 2019. The 6c/6t Intel chip only fairs slightly better clock for clock. It does kind of destroy current ryzen with a clock speed advantage though.
 

InvalidError

Titan
Moderator
Just wait until AAA developers start coding for the new ryzen based game consoles. Anybody who decided to buy 4 thread and 6 thread intel CPU's for the better game performance is going to be regretting that.
It'll be quite a few more years before game developers get up to speed with next-gen consoles to the point of it becoming a concern on the PC side. Plenty of time for AMD to solidly establish 8C16T as mainstream and possibly force Intel to follow unless Intel is taking a break from retail CPUs as seems to be the case in the leaked roadmaps.
 
Reactions: rigg42

Eximo

Titan
Herald
I agree, takes a bit longer for the industry to stop working on older platforms. Always the lowest common denominator for consoles. If they are going to continue PS4 and XBoxOne support for several more years, they will certainly target that CPU requirement for any non-exclusives and let the updated GPUs in the new units drive 4K or higher FPS, and higher settings.

One would hope they are already coding for upcoming consoles. They will need a few launch titles after all. Partners get the expected performance limits of the new hardware and likely engineering/development samples to test with. Given how much information is out there about next gen hardware, safe to say someone has it in hand.
 
Reactions: rigg42
Just wait until AAA developers start coding for the new ryzen based game consoles.
I'm pretty sure we all heard the same exact statements when the PS4/XB1 came out. You don't code SW that way; you make threads as necessary and let the OS manage them. Honestly, OS calls aside, the majority of the functional code is more or less identical between consoles and PCs.
 
Reactions: Mandark

InvalidError

Titan
Moderator
I'm pretty sure we all heard the same exact statements when the PS4/XB1 came out.
We do have considerably more games that scale at least up to 4C8T/6C6T today, thanks to per-thread performance having more or less flat-lined for the past eight years and very little likelihood of it getting any better at any point in the future. If software developers need to get more performance, their only option is more thread-level parallelism even if it means having to write code in a counter-intuitive way to achieve that.
 
Reactions: rigg42

Mandark

Distinguished
Sep 13, 2002
1,517
166
19,990
13
I'm pretty sure we all heard the same exact statements when the PS4/XB1 came out. You don't code SW that way; you make threads as necessary and let the OS manage them. Honestly, OS calls aside, the majority of the functional code is more or less identical between consoles and PCs.
sounds like you know a thing or two about software development. yes, I agree. the new consoles will however give us really smooth 4k and all that stuff, higher sustained frame rates, etc.. etc.. and yes, the code is most likely almost identical for consoles and pcs, they are fundamentally the same thing (in a different package) for crying out loud... lol

I marvel at my lowly Xbox One (first gen) and how magnificent games like Red Dead Redemption run and look, and that hardware isn't that impressive at all.
 
Some context of the speeds and gaps between the i7 8700K (I still consider this a great CPU, BTW) and the engy sample:

https://cpu.userbenchmark.com/Compare/AMD-Eng-Sample--2D3212BGMCWH2-3734-N-vs-Intel-Core-i7-8700K/m697865vs3937

So it's clocked at ~3.4Ghz and the 8700K is 28% ahead in single core speed at ~4.7Ghz. I'd say that is pretty damn impressive, as the page says it didn't go beyond 3.4Ghz.

EDIT: This is what I'm talking about (napkin math):
  • 137 / 107 - 1 = 0.280 -> 28.0%
  • 4.7 / 3.4 - 1 = 0.382 -> 38.2%
The 10 core part is going to boost up to ~5Ghz on its own, so it's going to be above the i7 8700K in single core speed. Well, for whatever they test in that place, that is.

MORE EDIT: The big fat assumption I'm surprised no one has pointed out is about how linear (or not) the test score in that site is. And that score looks oddly similar to CB scores. I'll find more information against other engy samples from the Ryzen 2K gen era and extrapolate a bit more information.

2700X @3.5Ghz: https://cpu.userbenchmark.com/UserRun/16847690

  • 107 / 101 - 1 = 0.059 = 5.9% (or 1 - 101 / 107 = 0.056 = 5.6%)
  • 3.5 / 3.4 - 1 = 0.029 = 2.9%
Interesting numbers and looks like there is a slight IPC increase (~7%). The ~30% jump in performance looks more real now with these tidbits of information if they'll have > 4.7Ghz turbo speeds. Now, the 2700X was using slower RAM, so that could also play against it in the benchmark.

But hey! I still like what I'm seeing.

Cheers!
 
Last edited:
We do have considerably more games that scale at least up to 4C8T/6C6T today, thanks to per-thread performance having more or less flat-lined for the past eight years and very little likelihood of it getting any better at any point in the future. If software developers need to get more performance, their only option is more thread-level parallelism even if it means having to write code in a counter-intuitive way to achieve that.
The only real difference today as far as software design goes is the primary rendering thread is no longer the only thread that can do all the rendering; you can offload that to multiple different threads now as of DX12/Vulkan. (Note: DX11 allowed this under very constrained circumstances).

And as I've long noted, your maximum parallelism is limited by Amdhal's Law. There's only so much you can do by adding more threads, and there's a point where the necessary thread scheduling/memory management starts to become an issue for non-trivial tasks.

sounds like you know a thing or two about software development. yes, I agree. the new consoles will however give us really smooth 4k and all that stuff, higher sustained frame rates, etc.. etc.. and yes, the code is most likely almost identical for consoles and pcs, they are fundamentally the same thing (in a different package) for crying out loud... lol

I marvel at my lowly Xbox One (first gen) and how magnificent games like Red Dead Redemption run and look, and that hardware isn't that impressive at all.
I doubt we'll get smooth 4k, simply because the GPUs won't be able to push it reliably. 4k/60 is probably out of reach for anything that's computationally intensive. 4k/30 is doable.

Also remember that software on consoles can make a lot of assumptions and optimizations that PCs can't, due to having one HW spec. That's why no one really expected the PS4 Pro/XB1 X, but those were necessary due to their CPUs being cripples (as I continue to note: The PS3 has a more powerful CPU then either current gen console). This generation was really hamstrung by using early gen APUs as their main processors.
 

ASK THE COMMUNITY