AMD Epyc 7742 2P Rome Server Review

[HEXUS]

More of just about everything that matters in the datacenter space.

Read more.

[Korrorra]

Wish cpu's could get away from the old x86 leftover and create something new. Why not just have a x64 default. Or x128.

[3dcandy]

Wish cpu's could get away from the old x86 leftover and create something new. Why not just have a x64 default. Or x128.

Simply because it would break so much stuff.....
But yes - I agree!

[DanceswithUnix]

Wish cpu's could get away from the old x86 leftover and create something new. Why not just have a x64 default. Or x128.

I think that is one of the reasons there is so much interest in RISC-V.

Making a cpu go this fast is hard enough that dealing with the historical baggage of x86 doesn't really add that much to the problem. Previous estimates were about 5% of die size. Decoding AMD64 instructions is already pathologically hard so adding the old x86 junk in there doesn't really matter. Using a modern ISA like RISC-V would make lots of the decode logic go away, and maybe get a small performance improvement, but not much. It would probably win most if you went for a lower performance CPU where the core size would benefit a lot more, and then compared to AMD64 you could cram a lot more high-ish performance cores on a cheap-ish die.

But there is truth to what you say. I presume modern PCs still honor the A20 gate from 286 days, adding a slight delay to the addressing path and limiting clock speeds, but I doubt more than a handful of people would use that for their DOS himem.sys driver these days.

Simply because it would break so much stuff.....
But yes - I agree!

From the last time I tried running the original Civilization game (which I think expected 16 bit Windows 3.1) on a modern PC, it's already broken.

[MLyons]

Bet Intel PR are having a wonderful morning

[Yoyoyo69]

Bet Intel PR are having a wonderful morning

Very true.

I really do wonder what slanderous trash they come up with this time.

One day, they're bound to surprise us all and actually be honest.... or say nothing, surely it has to happen?

[Tabbykatze]

Bet Intel PR are having a wonderful morning

Agreed.

Hexus not reporting on the SWAPGS vulnerability, you didn't report on the ZombieLoad one either?

[Zak33]

Bet Intel PR are having a wonderful morning

Very true.

I really do wonder what slanderous trash they come up with this time.

One day, they're bound to surprise us all and actually be honest.... or say nothing, surely it has to happen?

best guess?

"Call Dev... bring a bigger fridge, this needs to be cooler"

[DanceswithUnix]

Nice, would be interesting to see it do some real benchmark loads and virtualisation loads as I can see these being VM monsters.

I think for the conclusions, AI is usually in the form of plug in cards making PCIe 4 a better optimisation than the AVX512 instructions.

There is a silly typo on page 3, I presume those NVMe modules are 4TB, not 4GB.

Can't say I'm too surprised that Intel were slow in getting a machine together to compare this against

[Quartz]

I'm not seeing any tests on how Rome handles a mix of VMs, surely a must for any datacentre. Likewise I'd like to see MS Terminal Server tested.

[3dcandy]

I think that is one of the reasons there is so much interest in RISC-V.

Making a cpu go this fast is hard enough that dealing with the historical baggage of x86 doesn't really add that much to the problem. Previous estimates were about 5% of die size. Decoding AMD64 instructions is already pathologically hard so adding the old x86 junk in there doesn't really matter. Using a modern ISA like RISC-V would make lots of the decode logic go away, and maybe get a small performance improvement, but not much. It would probably win most if you went for a lower performance CPU where the core size would benefit a lot more, and then compared to AMD64 you could cram a lot more high-ish performance cores on a cheap-ish die.

But there is truth to what you say. I presume modern PCs still honor the A20 gate from 286 days, adding a slight delay to the addressing path and limiting clock speeds, but I doubt more than a handful of people would use that for their DOS himem.sys driver these days.



From the last time I tried running the original Civilization game (which I think expected 16 bit Windows 3.1) on a modern PC, it's already broken.

Games though huh - ade shrugs shoulders

[Tabbykatze]

I'm not seeing any tests on how Rome handles a mix of VMs, surely a must for any datacentre. Likewise I'd like to see MS Terminal Server tested.

The closest I've seen is STH have a partner that does KVM virtualisation load metrics that they are able to get obfuscated data for:

https://www.servethehome.com/amd-epy...-a-knockout/8/