Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

First 96-Core AMD Zen 4 Threadripper Tests Show Utter Domination over Intel

extremetech.com

Post reply on HN
Print view View original

41–50 of 108 posts

Re: First 96-Core AMD Zen 4 Threadripper Tests Show Utter Domination over Intel

#41

Post by Dork1234 »

wingmanjd wrote: post #4

"One big disparity between the three machines was the Zen 4 machine was outfitted with 512GB of memory, while its competition had just 128GB each" This feels like cheating to me.

Larger sticks (or more sticks) of RAM are generally slower than smaller sticks. It might not be possible to duplicate same speed ram and capacity on the Intel board. I would hope they duplicated the test with the same speed of RAM before matching capacity. The results are so one sided, I doubt either capacity or speed matters.

Re: First 96-Core AMD Zen 4 Threadripper Tests Show Utter Domination over Intel

#42

Post by coder543 »

Racing0461 wrote: post #37
dralley wrote: post #17

Earlier quoted context omitted.

I doubt 2.5ghz vs 4.2ghz is going to make any real difference to "responsiveness". The more pressing concern would be making sure the kernel actually schedules foreground workloads promptly.

Maybe i'm missing why it doesn't matter. Ryzen runs at 4.5 base clock, TR Pro runs at 2.5 base clock, upon booting up nothing is happening so each cpu runs at their base clock. I open something like visual studio that isn't too intensive so each cpu doesn't realize it needs to boost. Won't the ryzen with 4.5 base clock feel snappier than the 2.5 one when vs loads? When a pretty big task (compiling, rendering) activat…

No. None of what you're writing is accurate. No processor made in the last decade will ever idle at the base clock. Any modern processor idles far, far lower than that, approximately 1Ghz, maybe even lower. Idling at multiple gigahertz would be a huge waste of energy. The minimum clock speed is not correlated to the "base clock" at all.

It takes tens of milliseconds to reach the boost clock: https://stackoverflow.com/a/64254459

The base clock is entirely irrelevant to everything you are talking about. Please reference my earlier comment for the meaning of "base clock": https://news.ycombinator.com/item?id=37980293

Re: First 96-Core AMD Zen 4 Threadripper Tests Show Utter Domination over Intel

#43

Post by userbinator »

IMHO, increasingly inefficient software notwithstanding, CPU performance has been well beyond acceptable since a long time ago.

Thus, the real question is how buggy these new CPUs will be. https://news.ycombinator.com/item?id=37812556

Re: First 96-Core AMD Zen 4 Threadripper Tests Show Utter Domination over Intel

#44

Post by hedora »

Has anyone done a perf / watt scatterplot of the current AMD and Apple offerings?

I'd guess this wins in absolute performance (due to the TDP), but I've heard AMD is pretty competitive on energy efficiency too these days. I'm typing this on an M2, which is simultaneously cooler and faster than my old Intel laptop.

Wondering if it's time to upgrade my decade-old Intel Linux desktop to an AMD yet.

Re: First 96-Core AMD Zen 4 Threadripper Tests Show Utter Domination over Intel

#45

Post by Racing0461 »

coder543 wrote: post #42
Racing0461 wrote: post #37

Earlier quoted context omitted.

Maybe i'm missing why it doesn't matter. Ryzen runs at 4.5 base clock, TR Pro runs at 2.5 base clock, upon booting up nothing is happening so each cpu runs at their base clock. I open something like visual studio that isn't too intensive so each cpu doesn't realize it needs to boost. Won't the ryzen with 4.5 base clock feel snappier than the 2.5 one when vs loads? When a pretty big task (compiling, rendering) activat…

No. None of what you're writing is accurate. No processor made in the last decade will ever idle at the base clock. Any modern processor idles far, far lower than that, approximately 1Ghz, maybe even lower. Idling at multiple gigahertz would be a huge waste of energy. The minimum clock speed is not correlated to the "base clock" at all. It takes tens of milliseconds to reach the boost clock: https://stackoverflow.com…

Interesting, i'l l have to do more research here since the reason i went with my current setup is based on this thinking.

Re: First 96-Core AMD Zen 4 Threadripper Tests Show Utter Domination over Intel

#46

Post by hetman »

userbinator wrote: post #43

IMHO, increasingly inefficient software notwithstanding, CPU performance has been well beyond acceptable since a long time ago. Thus, the real question is how buggy these new CPUs will be. https://news.ycombinator.com/item?id=37812556

I don't think someone doing video editing or 3D rendering is likely to agree with you here...

Re: First 96-Core AMD Zen 4 Threadripper Tests Show Utter Domination over Intel

#47

Post by nineteen999 »

xcv123 wrote: post #39
mrichman wrote: post #25

What use cases is this CPU ideal for? It's not a GPU, so gaming, ML model training, and crypto mining are out.

Production 3D rendering is still done on CPU at VFX studios.

Exactly, no guarantees that your scene will even fit in GPU VRAM given the complexity of some VFX shots.

Re: First 96-Core AMD Zen 4 Threadripper Tests Show Utter Domination over Intel

#48

Post by asfdgianionio »

ahahahahah wrote: post #20
_joel wrote: post #9

Earlier quoted context omitted.

They're testing CPU performance, not memory capacity. They need that much for the channels.

It just seems so incredibly sloppy to not have that be consistent. Do they expect all of their readers to then audit the code of the benchmarking that they are doing to ensure that it cannot possibly be bound by this?

There is no need to start out every article with remedial lessons. It is reasonable for an enthusiast magazine to assume familiarity with how computers work. The average ExtremeTech reader should already know what memory is and how increasing the amount of memory changes (or doesn't change) performance.

It is not possible to give all the systems the same memory configuration. They don't support the same memory and don't have the same number of channels. And I expect giving all the systems 512 GiB of memory would only widen the lead because larger DIMMs are usually slower.

>Do they expect all of their readers to then audit the code of the benchmarking that they are doing to ensure that it cannot possibly be bound by this?

Cinebench is hardly obscure. The average ExtremeTech reader will already know what it is and what it measures.

Re: First 96-Core AMD Zen 4 Threadripper Tests Show Utter Domination over Intel

#49

Post by bitwize »

mrichman wrote: post #25

What use cases is this CPU ideal for? It's not a GPU, so gaming, ML model training, and crypto mining are out.

These things can chew through compiles so fast I/O to disk will be the bottleneck, rather than actual compilation time. Great for Rust hackers.

Re: First 96-Core AMD Zen 4 Threadripper Tests Show Utter Domination over Intel

#50

Post by CobaltFire »

userbinator wrote: post #43

IMHO, increasingly inefficient software notwithstanding, CPU performance has been well beyond acceptable since a long time ago. Thus, the real question is how buggy these new CPUs will be. https://news.ycombinator.com/item?id=37812556

Tell that to my flight simulators. DCS has gotten better with the recent MT release, but the determinative requirements of that coupled with multiplayer servers in the 50+ population and all the AI brings most systems to their knees. Especially in VR…
Post reply on HN