Tuesday, September 24th 2019

AMD Could Release Next Generation EPYC CPUs with Four-Way SMT

AMD has completed design phase of its "Zen 3" architecture and rumors are already appearing about its details. This time, Hardwareluxx has reported that AMD could bake a four-way simultaneous multithreading technology in its Zen 3 core to enable more performance and boost parallel processing power of its data center CPUs. Expected to arrive sometime in 2020, Zen 3 server CPUs, codenamed "MILAN", are expected to bring many architectural improvements and make use of TSMC's 7nm+ Extreme Ultra Violet lithography that brings as much as 20% increase in transistor density.

Perhaps the biggest change we could see is the addition of four-way SMT that should allow a CPU to have four virtual threads per core that will improve parallel processing power and enable data center users to run more virtual machines than ever before. Four-way SMT will theoretically boost performance by dividing micro-ops into four smaller groups so that each thread could execute part of the operation, thus making the execution time much shorter. This being only one application of four-way SMT, we can expect AMD to leverage this feature in a way that is most practical and brings the best performance possible.
AMD isn't the first to implement this kind of solution to its processors. IBM has been making CPUs based on POWER ISA for years now that feature four or even eight-way SMT and they are one of the key reasons why POWER CPUs are so powerful. Nonetheless, we can hope to see more details about Zen 3 core design decisions as we approach 2020 and launch of Milan CPUs. Source: Hardwareluxx
Add your own comment

159 Comments on AMD Could Release Next Generation EPYC CPUs with Four-Way SMT

#151
GoldenX
Camm, post: 4124276, member: 110377"
We've seen this happen with both Dolphin & RPCS3 in the past, emudevs targeting Nvidia in particular, and thinking its broken on AMD's side, where as its a non-reference implementation of OpenGL from Nvidia that is the cause, and would daresay would be a huge contributing factor here.
Try to use the AMD drivers on a project, face those problems yourself, then come back here and tell me. They are AWFUL. Intel drivers are FAR superior, and we joke about their GPUs all the time.
It's easy to blame the devs when the only thing you do is install a .exe.
Posted on Reply
#152
Camm
GoldenX, post: 4124281, member: 160319"
Try to use the AMD drivers on a project, face those problems yourself, then come back here and tell me. They are AWFUL. Intel drivers are FAR superior, and we joke about their GPUs all the time.
It's easy to blame the devs when the only thing you do is install a .exe.
This has been documented before.I'm pretty sure Dolphin when implementing ubershaders called this (amongst other issues) out. And please, no sanctimonous bullshit on the religious wars of emudevs.
Posted on Reply
#153
GoldenX
It simple, Nvidia gets better features and performance first, because their OpenGL drivers are just better, there's no conspiracy behind it.
Even an UHD620 doesn't get stuck on draw calls, unlike a Navi.

https://community.amd.com/thread/206176
Check it yourself, even PCSX2 is tired of AMDs bullshit.
Posted on Reply
#154
Camm
GoldenX, post: 4124290, member: 160319"
It simple, Nvidia gets better features and performance first, because their OpenGL drivers are just better, there's no conspiracy behind it.
Even an UHD620 doesn't get stuck on draw calls, unlike a Navi.

https://community.amd.com/thread/206176
Check it yourself, even PCSX2 is tired of AMDs bullshit.
The very first post I said AMD's Windows OpenGL implementation suck, but you keep pushing that out to both linux and vulkan, where they really don't (and now including PCSX2 which yes, references AMD's shit Windows OpenGL Performance). Man, I really would hate to have you testing my code when you can't even make that distinction. I'm out.
Posted on Reply
#155
GoldenX
Camm, post: 4124291, member: 110377"
The very first post I said AMD's Windows OpenGL implementation suck, but you keep pushing that out to both linux and vulkan, where they really don't (and now including PCSX2 which yes, references AMD's shit Windows OpenGL Performance). Man, I really would hate to have you testing my code when you can't even make that distinction. I'm out.
Get a GCN1 card (7750, 7870, 7970, 270, whatever), test amdgpu and radv. Tell me how it goes.
Posted on Reply
#156
Camm
GoldenX, post: 4124295, member: 160319"
Get a GCN1 card (7750, 7870, 7970, 270, whatever), test amdgpu and radv. Tell me how it goes.
Learn the difference between linux and windows. Tell me how you go.
Posted on Reply
#157
GoldenX
Camm, post: 4124297, member: 110377"
Learn the difference between linux and windows. Tell me how you go.
I do that constantly, there's no magic on the Linux drivers too. Test it yourself, put a distro on a usb stick, boot steam and use them. Check RPCS3, Dolphin, etc on both AMDGPU and RADV. See their performance, their bugs, the crashes and memory leaks.
AMD has a nice front for Windows Direct3D gamers, the rest of it's software is rotting.
Posted on Reply
#158
efikkan
Guys, aren't we quite a bit off topic here?
While I can attest to AMD's OpenGL support as being garbage after using it for development for years, their Linux support is actually worse than their Windows support. But this deserves a thread of its own, as I don't see how this relates to Epyc CPUs with 4-way SMT.
Posted on Reply
#159
red-ray
Aldain, post: 4121702, member: 170164"
Are you F stupid?? When did intel implement 4 way smt??
In the Intel Xeon Phi 7290 (Knights Landing)

Posted on Reply
Add your own comment