- Mar 3, 2017
- 1,622
- 5,892
- 136
Core-to-core results can be different with 128c or 8c.No, he said "core to core"!!
Sure, but if AMD was better core to core with their 128 core product, which likely will not even clock as high as their 96 core product then their 8 core product will be even stronger compared to a different 8 core product. So the 40% should be the MINIMUM especially in the desktop market.Core-to-core results can be different with 128c or 8c.
If their epyc chips are 40% faster then that should translate nicely to their Ryzen chips...40% faster Epycs and 20% faster Ryzens would not mske sense.I'd be more likely to believe the 40% uplift for Epyc chips, they're trying to throttle whatever hail mary intel is trying to cook up after Emerald Rapids. If we get half that for Ryzen, I'd be thrilled (and lining up for a 9900x)
There are probably 16 * 8c CCDs in that 128c SKU. Imagine the traffic. Besides the server IOD is not reused from Genoa unlike the client IOD.40% faster Epycs and 20% faster Ryzens would not mske sense.
There are probably 16 * 8c CCDs in that 128c SKU. Imagine the traffic. Besides the server IOD is not reused from Genoa unlike the client IOD.
If their epyc chips are 40% faster then that should translate nicely to their Ryzen chips...40% faster Epycs and 20% faster Ryzens would not mske sense.
It'd make sense if zen 5 can't scale with clock and this is a very big if.If their epyc chips are 40% faster then that should translate nicely to their Ryzen chips...40% faster Epycs and 20% faster Ryzens would not mske sense.
No.and probably a different approach to solving various perf scaling problems (like different coherency protocols), etc.
Server and DT do not share the CCD.I could very easily see them giving us some safety-scissors binned silicon and keeping the very best for ginormous chiplet blobs.
They do not care anymore.they're trying to throttle whatever hail mary intel is trying to cook up after Emerald Rapids.
I am also skeptical of the high claims applying to desktop where clock rates are already absurd. But there is no business sense in nerfing a halo product. Even if it is hard to acquire the brand reputation is worth it to have a desktop part pushed to the limits.Well, from a business perspective it would be pretty reasonable to hold back that sort of a jump lest they completely ruin their currently saleable product line. There's a giant high-margin market to be captured with Epyc, and they wouldn't necessarily want to deprive themselves of supply there... I could very easily see them giving us some safety-scissors binned silicon and keeping the very best for ginormous chiplet blobs.
That's fabric power.Which, I think idle power is the main aspect that ARM was holding over x86, with Intel making significant inroads in their most recent chips?
Not changing much until Medusa.Hoping AMD can make inroads there.
It's the exact same cIOD.Any news on the iGPU going in the non-APU Zen 5 (Zen 4 has a 2CU part, does it include video processing block)?
No, no, yes, a bit.Will we be getting new chipsets for Zen 5? Anything else new? It'll still be DDR5 and PCIe 5? Just faster DDR5?
I don't think Emerald Rapid's successor Granite Rapids is much of a competition to Zen5 in Servers.I'd be more likely to believe the 40% uplift for Epyc chips, they're trying to throttle whatever hail mary intel is trying to cook up after Emerald Rapids. If we get half that for Ryzen, I'd be thrilled (and lining up for a 9900x)
Server and DT do not share the CCD.
no commentIs that new for Zen 5, or did the CCDs diverge before?
They have something, but it isn't ready yet.Yeaaaa let's just say IDC needs something far meaner than LNC to compete.
BTW, what happened @adroc_thurston and @Kepler_L2 that caused you to change your predictions ~10% higher than just a few weeks ago?:
ARL-S and Zen 5 Desktop Speculation
RULE RULE This spreadsheet is unmoderated. No spamming and please respect what others have written. I will stop making these public speculation spreadsheets once someone breaks the rule. That's all, have fun. (By the way, the tabs are at the bottom.)docs.google.com
That google sheet was put up on March 9:
Zen 5 Hype Train taking off from the station?
I didn't see the explanation, but among other things posting exact numbers is never a good idea. There are ways of tracing certain information, but more important, numbers are meaningless without context. That is why I told people to take the Arrow Lake slide for what it is worth.He has explained his predictions already. Scroll up.
Rule the power infrastructure of a modern country, perhaps.Our resident Prince Harry expects Arrow Lake to hit 7 GHz!
Intel will RULE!!!
I am sorry, this is not specifically about Zen 5, but AMD literally did this just a few years ago. Apple did it with the M1 vs both x86 and ARM. Why is it so hard to believe? Because Intel isn't doing it? Intel has been stuck and spinning their wheels to get anywhere. Don't get me wrong, they ARE working on some stuff, but they were far too reliant on node improvements and that absolutely kicked them in the rear end.if Zen 5 is indeed >40% uplift in ST SPECint (which is usually very close to Geekbench ST and the average IPC AMD reports) then this is quite a unique achievement.
The industry standard is about 20% IPC uplift for a new generation (usually on a new process with plenty more transistors).
See above for the first part.I am also skeptical of the high claims applying to desktop where clock rates are already absurd. But there is no business sense in nerfing a halo product. Even if it is hard to acquire the brand reputation is worth it to have a desktop part pushed to the limits.
It's the reason why some gamers with a measly 4070 will call Radeon people 'the poors' despite those with a 7900 XTX having much better performance. Like the RTX 4090 the 8950X(3D) doesn't have to be available in large numbers or be affordable it simply has to exist to get prestige sales further down the stack.
yeah i know but they'll be too little too late as usualThey have something, but it isn't ready yet.
So 5.8 to 6 GHz turbo frequency? Means a slight bump compared to Zen4 at 5.7 GHz (for 7950X).MLID mentioned 5.8 to 6 GHz top speed for Zen 5 in his latest video.
Not really a lot of newsworthy info in the video otherwise.
So 5.8 to 6 GHz turbo frequency? Means a slight bump compared to Zen4 at 5.7 GHz (for 7950X).
No clock regression after all then, as previously was suggested by some in this thread?
Ok, so now we’re at 40+ % IPC and 2-5% frequency increase on top, which adds up to around 45-50% perf increase in total.
I really hope AMD will be able to deliver this, because the expectations are set high for sure now! Don’t want to get disappointed.
Wasn't it explained at the time that the 15% was spec in rate n/T aka socket performance? So Turin has More Core with More IPC per core at the same power and about the same memory bandwidth. Which in the modern world of highend compute is actually crazy crazy good.No one said 40% IPC, that s the perf improvement, although that seems quite inflated in respect of the slide provided by MLID even if 10-15% better IPC is a voluntary underestimation by AMD.
Ok, so now we’re at 40+ % IPC and 2-5% frequency increase on top, which adds up to around 45-50% perf increase in total.
I really hope AMD will be able to deliver this, because the expectations are set high for sure now! Don’t want to get disappointed.
Wasn't it explained at the time that the 15% was spec in rate n/T aka socket performance? So Turin has More Core with More IPC per core at the same power and about the same memory bandwidth. Which in the modern world of highend compute is actually crazy crazy good.