AMD Talks Up Zen 4 AVX-512, Genoa, Siena & More At Financial Analyst Day

[phoronix]

Phoronix: AMD Talks Up Zen 4 AVX-512, Genoa, Siena & More At Financial Analyst Day

AMD today hosted their 2022 Financial Analyst Day where they made some new disclosures and firmed up past product road-map plans...

https://www.phoronix.com/scan.php?page=news_item&px=AMD-Financial-Day-2022

[Linuxxx]

So, will AVX-512 support mean all instructions are going to be implemented or only the ones benefitting AI/ML workloads?

And will it be similar to first gen Zen 1 where AVX2 was realized as 2 × 128 bits wide registers, so maybe AVX-512 as 2 × 256 bits?

[photom]

Great news, btw. I'll only buy another piece of AMD hardware if the (linux) software is ready on release date - I have been fooled thrice and I still buy AMD stuff, perhaps I should use Intel/Linux or AMD/Windows

[tildearrow]

For EPYC 7004 "Genoa", AMD is talking greater than 75% improvement for enterprise Java performance compared to EPYC 7003 series.

Minecraft

Genoa-X was confirmed as Genoa with 3D V-Cache. Genoa-X will offer more than 1GB of 3D V-Cache per CPU/socket.

Wow! At this rate in 10 to 15 years we'll have the entire system memory (or at least 1/4 of it) in processor cache.

[EvilHowl]

I'm really interested on the AMD Instinct MI300, which is basically an incredibly big and powerful APU with lots of memory bandwidth. Just imagine having that as a consumer product.

[cb88]

Wow! At this rate in 10 to 15 years we'll have the entire system memory (or at least 1/4 of it) in processor cache.

These systems are typically gonna have 512-2TB of ram... it doesn't make much sense to buy a CPU that costs as much as a car and spend coffee maker money on the ram. They certainly aren't going to have a mere 16GB... even one stick in these systems will likely be 128GB+

[birdie]

Minecraft



Wow! At this rate in 10 to 15 years we'll have the entire system memory (or at least 1/4 of it) in processor cache.

Most likely not going to happen ever. The attached memory is very special, limited (in terms of space) and very expensive. Apple promises 24GB of it as far as I remember, but on PC you can have 128GB or is it 256 with DDR5 now? I don't remember. Servers can spot a lot more of course. Literally terabytes of RAM for multisocket configurations.

[ms178]

So, will AVX-512 support mean all instructions are going to be implemented or only the ones benefitting AI/ML workloads?

And will it be similar to first gen Zen 1 where AVX2 was realized as 2 × 128 bits wide registers, so maybe AVX-512 as 2 × 256 bits?

According to the Youtuber Moore's Law is Dead it was said to be comparable in performance to IceLake-X at same thread count / clock speeds (see the slide in the video here: https://youtu.be/6yFn85I5PbY?t=395).

[mumar1]

I'm really interested on the AMD Instinct MI300, which is basically an incredibly big and powerful APU with lots of memory bandwidth. Just imagine having that as a consumer product.

If your pockets are deep enough you will have it as a "consumer product" next year ;-)

[zamroni111]

3nm euv will be very expensive foundry unless cheaper and more efficient euv laser generation is invented.
Low profit margin products, e.g. consumer cpu and gpu, might not be profitable enough to use it.
i won't be surprised if even apple don't use it for a and m soc.