That would be my opinion, but I have no evidence to back it up.cma6 wrote:Joseph:
Then you believe that one is getting a better search--perhaps much better search--(at the same depth) with many more threads at work?
Intel’s 28-Core Xeon Platinum 8176
Moderators: hgm, Rebel, chrisw
-
Modern Times
- Posts: 3550
- Joined: Thu Jun 07, 2012 11:02 pm
Re: Intel’s 28-Core Xeon Platinum 8176
-
Yar
- Posts: 298
- Joined: Wed Mar 08, 2006 9:00 pm
- Location: Moscow, Russia
Re: Intel’s 28-Core Xeon Platinum 8176
AMD EPYC 7601 SiSoft Sandra and Cinebench 2P Benchmarks and Comparison to Intel Platinum 8180Vinvin wrote:The 8176 runs 28 cores @ 2.1 GHz but the 8168 is probably faster for chess : 24 cores @ 2.7 GHz.
https://en.wikichip.org/wiki/intel/xeon_platinum/8168
https://en.wikichip.org/wiki/intel/xeon_platinum/8176
http://wccftech.com/amd-epyc-7601-benchmarks-analysis/
-
jdart
- Posts: 4367
- Joined: Fri Mar 10, 2006 5:23 am
- Location: http://www.arasanchess.org
Re: Intel’s 28-Core Xeon Platinum 8176
I don't think Cinebench is a very reasonable workload for these chips.
As an engineer most of the apps I have seen deployed on high-end chips are database servers or high-volume web servers. Some people also use them for 3D rendering. But for most of these applications, 28 cores would be overkill.
I also have seen an app that did inventory optimization for a huge retailer, running overnight. And a few other big things like that.
But a lot of things can be run on a big network of cheap machines rather than one or a few big machines. You need the big machine when you can make good use of low latency shared RAM, as in chess.
--Jon
As an engineer most of the apps I have seen deployed on high-end chips are database servers or high-volume web servers. Some people also use them for 3D rendering. But for most of these applications, 28 cores would be overkill.
I also have seen an app that did inventory optimization for a huge retailer, running overnight. And a few other big things like that.
But a lot of things can be run on a big network of cheap machines rather than one or a few big machines. You need the big machine when you can make good use of low latency shared RAM, as in chess.
--Jon