Don
And Zen4C still does AVX-512 (to great effect):
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
AMD 16 core chiplets fly under the radar
- Thread starter humbug
- Start date
Would be nice, probably not. An 8(full power) chiplet +16(lower power) chiplet would be great.
Don
Nope still 8+8 afaik
Rumours are that the Zen5c cores might make an appearance in some APUs though:
AMD Ryzen 8000 "Strix Point" APU said to feature 4xZen5 with 8xZen5c cores, "Strix Halo" using 16xZen5 - VideoCardz.com
AMD Strix Halo similar to a desktop CPU with more graphics-capable IO die The upcoming AMD next-gen APU series, based on the Zen5 architecture, is rumored to adopt a hybrid design, combining Zen5 and Zen5c codes. However, the specific configuration of these chips has not been officially...
videocardz.com
(Interesting also a big upgrade to the Graphics side of the APUs as well - 16 CU and 40 CU parts - although can't see the point without some embedded infinity cache / HBM on package)
No, big IPC gain tho, 20% minimum, possibly more.
Performance should be up 30% per core, that counts for MT too.
Score about 49,000 in R23.
I prefer than than having more slower cores.
Last edited:
I can see a case for both.
Last edited:
Nope still 8+8 afaik
Rumours are that the Zen5c cores might make an appearance in some APUs though:
AMD Ryzen 8000 "Strix Point" APU said to feature 4xZen5 with 8xZen5c cores, "Strix Halo" using 16xZen5 - VideoCardz.com
AMD Strix Halo similar to a desktop CPU with more graphics-capable IO die The upcoming AMD next-gen APU series, based on the Zen5 architecture, is rumored to adopt a hybrid design, combining Zen5 and Zen5c codes. However, the specific configuration of these chips has not been officially...
(Interesting also a big upgrade to the Graphics side of the APUs as well - 16 CU and 40 CU parts - although can't see the point without some embedded infinity cache / HBM on package)
I cant wait to get hold of those chips.
Last edited:
Soldato
I've got a 2400G system connected to the TV, and it's good for some old games and great for emulation. I've been waiting for AMD to do something when it comes to APUs on the desktop. In the handheld/mini-PC space the integrated graphics available are awesome. I'd happily go a mobo/CPU and DDR5 upgrade if performance is in-line with their handheld offerings and the price isn't comically high (which it likely will be).
Soldato
Intel went the E core route because they couldn't cram more P cores into their CPUs, not because of some awesome design advancement. AMD doesn't have that limitation, plus Bergamo is designed for specific use cases, so the C cores have had some things stripped out and the cache reduced.
The C-class cores offer no benefit to desktop or workstations, so there's no need to mix and match. AMD are perfectly capable of downclocking cores as required, they have no need to replicate Intel's crutch.
The C-class cores offer no benefit to desktop or workstations, so there's no need to mix and match. AMD are perfectly capable of downclocking cores as required, they have no need to replicate Intel's crutch.
The argument is but but but laptops and power consumption, yes, sort of, to a limited extent.
Intel are actually being quite clever in that they want a performance lead over AMD in terns of per-core performance (games / some 2006 code Adobe apps) and MT performance.
The Ring Bus only support 8 to 10 cores, that's ok you only need 8, so you make these 8 really fat cores for games and Adobe apps and for everything else you can tag on an infinite number of these little cores.
The little cores have much less transistors and limited performance so if windows is doing something in the background it can use those cores at 6 watts instead of 12 for the big ones and it makes no difference.
Cool, ok, its not the same thing as those ARM chips that you see in mobiles, those little cores are RISC, they are striped out instruction sets, they can't do everything the big cores can and the power consumption of those is measured in tiny fractions of a watt, its really not the same thing at all, Intel little cores are full CISC cores, just crappier slower versions of the P-Cores.
Now here is the thing about all of this, the P-Cores are 4X the size of the E-Cores, impressive right? Well those P-Cores are also 3X the size of AMD's P-Cores, AMD big cores are far closer in size to Intel's little cores, and for those massive 3X the size P-cores Intel gain......................... 10% IPC over AMD's cores, what's more AMD's P-Cores use no more power during these low stress you don't even know its happening tasks, so the Intel battery life in pure idle is no better than AMD's while working AMD's battery life is better.
What Intel are doing is clever, exactly the sort of innovation you would expect from Pat Gelsinger, but completely ineffective. I have no doubt Pat thought it would be his second Sandy Bridge moment, no. Not this time buddy.
Last edited:
Soldato
I was under the incorrect impression that the C-class cores actually stripped some things out so they're dedicated to specific tasks, such as Cloud stuff. It appears I was wrong and the only difference is the C-class cores have less cache and a physically lower power envelope and clock speed, which reduces their size (and no vias for V-cache).
If that's the case then yes I can see a benefit (to an extent) of going a mixed chiplet approach akin to Intel's Hybrid and schedule the C-class cores to running small tasks and have the full-fat cores run the grunt stuff. But like I said before, since full-fat cores can be downclocked and powered off perfectly fine on their own, it's probably overcomplicating things to mix and match full and C-cores. If AMD want to do super low power laptops and devices then just use C chiplets exclusively, don't muddy the waters (and the Windows scheduler) by mixing chiplets.
If that's the case then yes I can see a benefit (to an extent) of going a mixed chiplet approach akin to Intel's Hybrid and schedule the C-class cores to running small tasks and have the full-fat cores run the grunt stuff. But like I said before, since full-fat cores can be downclocked and powered off perfectly fine on their own, it's probably overcomplicating things to mix and match full and C-cores. If AMD want to do super low power laptops and devices then just use C chiplets exclusively, don't muddy the waters (and the Windows scheduler) by mixing chiplets.
Last edited:
You cant just go flying chips in under the radar... its rude. Looks like it goes brum brum! I was was in the market for an Epyc chip right now you could be damn sure it would need to go brum brum... I still find what they are doing here and how they approached it pretty spot on.
Last edited:
The rear mounted brum approach.
Exactly, I wonder if they knew when they named Epyc, just how epic it would become? - What we are seeing now is the fruits of that initial way of thinking like 8 years ago or something. You don't just come up with this **** overnight.
Associate
I think it was infinity fabric which unlocked it all. I remember reading an article about the zen developments and that the IF was a game changer design which could continue to apply to a lot more stuff.
Yea 100% the fabric is a massive contributor and is years ahead of what anybody else has basically.
Permabanned
- Joined
- 10 Nov 2005
- Posts
- 2,553
awesome!
Associate
Yup, i think it has the potential to become an interconnect between devices too. For example cpu, gpu, storage, etc could all talk independently to each other via I.F. Ie you wouldn’t need to round trip via the cpu
Resurrecting this thread.
I was having a look at the potential for having the equivalent of 16 P cores in the nearish future. Intel don't seem to have anything planned.
Some AMD CPUs are "in theory" 16 core, but in use are x2 CCDs of 8 cores, with the CCDs seemingly not talking to each other very effectively, meaning that gaming is apparently worse on a 7950 than a 7800.
I did look at info on the AMD Zen4c, which does look like having 16 cores on a single CCD. However, what I then read was that AMD are doing this by cramming x2 CCX of 8 cores onto a single CCD, with lower cache per core.
Has anyone read of a genuine 16 core (on a single board) in the pipeline?
For that matter, does anyone really understand why many of the synthetic benchmarks can make full use of the 16 cores in current AMD 7950s, whilst most games seem to fail miserably? If so, is there a potential fix?
I was having a look at the potential for having the equivalent of 16 P cores in the nearish future. Intel don't seem to have anything planned.
Some AMD CPUs are "in theory" 16 core, but in use are x2 CCDs of 8 cores, with the CCDs seemingly not talking to each other very effectively, meaning that gaming is apparently worse on a 7950 than a 7800.
I did look at info on the AMD Zen4c, which does look like having 16 cores on a single CCD. However, what I then read was that AMD are doing this by cramming x2 CCX of 8 cores onto a single CCD, with lower cache per core.
Has anyone read of a genuine 16 core (on a single board) in the pipeline?
For that matter, does anyone really understand why many of the synthetic benchmarks can make full use of the 16 cores in current AMD 7950s, whilst most games seem to fail miserably? If so, is there a potential fix?
Intel have their Xeon W chips and they are refreshing those shortly.
At some point I expect AMD to release a 7950X3D2 with extra cache on both CCDs.