Hacker News Reader

Comment by bitL | Hacker News Reader

They commit patches to OpenBLAS which is still 2-3x slower on Zen 2 than MKL on Intel hardware. Intel has better software engineers for high-performance math it seems.

View on HN

> Intel has better software engineers for high-performance math it seems.

Intel has software engineers for pretty much every application. AMD has very few software engineers.

Yeah, that's quite unfortunate, however as AMD seems to be able to earn some $ in the following ~2 years they might go on SWE hiring spree...

Or drive a truck full of money to Intel HQ to pay them to do it.

Probably need to be a lot of money though.

I think much of that will flow into their GPU department. They have the node advantage compared to Nvidia (TSMC 7nm vs. TSMC 12nm) but can only match Nvidia in power consumption and performance. Once Nvidia get's to 7nm with their new architecture they will increase the gap again (unless RDNA2 which is rumored for H1 2020 brings a big efficiency and performance improvement - and not only hardware accelerated RT).

AOCL (https://developer.amd.com/amd-aocl/ ) is also a thing that exists.

But yes, they are generally way behind in this area, and in my domain (genomics) this is a serious barrier to adoption. I've been waiting for years for good-enough AMD-optimized linear algebra libraries, but distribution of Linux binaries statically linked to Intel MKL is still the obvious best choice as of November 2019, and that's a shame.

MATLAB (which I think uses Intel MKL) showing faster performance for Intel than AMD but apparently due the above issues of the MKL libraries not utilising AVX on AMD processors? Is this correct?

https://www.extremetech.com/computing/302546-amd-threadrippe...

Yes, it's correct. But it's easy to fix, just do this: export MKL_DEBUG_CPU_TYPE=5

Depending on the benchmark, I've seen up to 6.6x improvement in matlab on the AMD Rome/Zen2 chips

Reply on news.ycombinator.com