By: hobold (hobold.delete@this.vectorizer.org), October 31, 2020 5:54 am
Room: Moderated Discussions
Jukka Larja (roskakori2006.delete@this.gmail.com) on October 30, 2020 9:04 pm wrote:
[...]
> Anyway, the fact that even AVX (non-2) on many CPUs offers twice the bandwidth, yet gathers no interest
> from game developers, just points more directly that SIMD extensions aren't interesting in the field.
In my absolutely not humble opinion, there needs to be a syntax more convenient than the existing intrinsics. Something like GCC's generic vector facility: GCC vector extensions.
If I have to (syntactically) write a function call even for a simple arithmetic operation, and if I have to manually fuse multiply-adds, then the code is so much hassle to maintain, that I will do it only for functions that _never_ change after deployment. That would be basic linear algebra, and not much else.
As far as I know there is work being done to include something like std::simd in future C++ standards. That might bring enough practical usability for all those SIMD ISAs.
[...]
> Anyway, the fact that even AVX (non-2) on many CPUs offers twice the bandwidth, yet gathers no interest
> from game developers, just points more directly that SIMD extensions aren't interesting in the field.
In my absolutely not humble opinion, there needs to be a syntax more convenient than the existing intrinsics. Something like GCC's generic vector facility: GCC vector extensions.
If I have to (syntactically) write a function call even for a simple arithmetic operation, and if I have to manually fuse multiply-adds, then the code is so much hassle to maintain, that I will do it only for functions that _never_ change after deployment. That would be basic linear algebra, and not much else.
As far as I know there is work being done to include something like std::simd in future C++ standards. That might bring enough practical usability for all those SIMD ISAs.
Topic | Posted By | Date |
---|---|---|
Expiry of x86-64 patents | Beastian | 2019/04/19 08:05 AM |
Expiry of x86-64 patents | Gian-Carlo Pascutto | 2019/04/19 08:46 AM |
Expiry of x86-64 patents | Beastian | 2019/04/19 09:06 AM |
Expiry of x86-64 patents | Jukka Larja | 2019/04/19 09:44 AM |
Expiry of x86-64 patents | Gian-Carlo Pascutto | 2019/04/19 10:12 AM |
Expiry of x86-64 patents | Jukka Larja | 2019/04/19 11:41 AM |
Expiry of x86-64 patents | Robert Williams | 2019/04/19 12:18 PM |
Expiry of x86-64 patents | Gian-Carlo Pascutto | 2019/04/19 01:35 PM |
Expiry of x86-64 patents | IntelUser2000 | 2020/10/30 01:17 AM |
Expiry of x86-64 patents | Jukka Larja | 2020/10/30 06:49 AM |
Expiry of x86-64 patents | me | 2020/10/30 08:47 AM |
Expiry of x86-64 patents | Jukka Larja | 2020/10/30 08:52 AM |
Expiry of x86-64 patents | Mark Roulo | 2020/10/30 09:21 AM |
Expiry of x86-64 patents | Jukka Larja | 2020/10/30 10:29 AM |
Expiry of x86-64 patents | Mark Roulo | 2020/10/30 10:42 AM |
Expiry of x86-64 patents | Jukka Larja | 2020/10/30 08:04 PM |
SIMD syntax | hobold | 2020/10/31 05:54 AM |
SIMD syntax | Jukka Larja | 2020/10/31 08:14 AM |
SIMD syntax | hobold | 2020/11/01 07:22 AM |
SIMD syntax | Jukka Larja | 2020/11/01 10:11 AM |
SIMD syntax | hobold | 2020/11/02 04:33 AM |
Expiry of x86-64 patents | me | 2020/10/31 02:01 PM |
Expiry of x86-64 patents | Jukka Larja | 2020/10/31 08:23 PM |
Expiry of x86-64 patents | Foo_ | 2020/11/01 03:48 AM |
Expiry of x86-64 patents | Jukka Larja | 2020/11/01 06:01 AM |
Expiry of x86-64 patents | Adrian | 2020/10/30 11:02 AM |
Expiry of x86-64 patents | Bigos | 2020/10/30 12:20 PM |
Expiry of x86-64 patents | Geoff Langdale | 2019/04/19 01:52 PM |
Expiry of x86-64 patents | Jukka Larja | 2019/04/19 08:38 PM |
Expiry of x86-64 patents | Yuhong Bao | 2019/04/20 01:35 PM |
Expiry of x86-64 patents | Doug S | 2019/04/19 09:40 AM |
Expiry of x86-64 patents | Beastian | 2019/04/19 10:10 AM |
Expiry of x86-64 patents | Robert Williams | 2019/04/20 07:15 AM |
Expiry of x86-64 patents | Robert Williams | 2020/10/28 05:42 AM |
Expiry of x86-64 patents | anyone | 2019/04/20 06:11 AM |
Expiry of x86-64 patents | Groo | 2019/04/20 06:29 AM |
Expiry of x86-64 patents | wumpus | 2019/04/20 07:32 AM |
Expiry of x86-64 patents | blaine | 2020/10/30 11:03 AM |
Expiry of x86-64 patents | David Kanter | 2020/10/30 07:59 PM |
Intel vs AMD patents | Yuhong Bao | 2019/04/20 01:32 PM |
Intel vs AMD patents | Beastian | 2019/04/20 02:35 PM |
Expiry of x86-64 patents | Travis Downs | 2019/04/20 06:24 PM |
Expiry of x86-64 patents | none | 2019/04/21 06:36 AM |
Expiry of x86-64 patents | somebody | 2019/11/27 09:44 AM |
Expiry of x86-64 patents | Anon3 | 2019/11/27 04:16 PM |
Expiry of x86-64 patents | Travis Downs | 2019/11/27 05:17 PM |
Expiry of x86-64 patents | Montaray Jack | 2019/11/27 11:03 PM |
Expiry of x86-64 patents | none | 2019/11/28 12:57 AM |
Expiry of x86-64 patents | dmcq | 2019/11/28 10:20 AM |
Expiry of x86-64 patents | Montaray Jack | 2019/11/29 04:00 AM |