Search
Search titles only
By:
Search titles only
By:
Log in
Register
Search
Search titles only
By:
Search titles only
By:
Menu
Install the app
Install
Home
Featured
Forums
Trending
Search forums
What's new
New posts
Latest activity
Members
Current visitors
JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an
alternative browser
.
N
naukkis
Senior member
Messages
787
Featured content
0
Profile views
453
Reaction score
639
Points
136
Joined
Jun 5, 2002
Last seen
Today at 4:25 PM
Find
Find content
Find all content by naukkis
Find all threads by naukkis
Latest activity
Postings
Featured content
About
Trophies
N
naukkis
replied to the thread
Discussion
Qualcomm Snapdragon Thread
.
128 bit vectors are useful. 256 bit not so much. Doing twice as wide 256bit vector SIMD will benefit single digits for performance at...
Yesterday at 8:10 AM
N
naukkis
replied to the thread
Discussion
Qualcomm Snapdragon Thread
.
Those are so niche use-cases that cpu designed for desktop and mobile should use 128 bit vector SIMD. Wider vectors in x86 is just...
Yesterday at 6:35 AM
N
naukkis
replied to the thread
Discussion
Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
.
From decreasing 14900KS clocks. Those st boost clocks come from voltages that silicon couldn't handle so comparison to those are...
Yesterday at 4:38 AM
N
naukkis
replied to the thread
Discussion
Qualcomm Snapdragon Thread
.
256 bit AVX2 is pretty useless for non-hpc use. Not many programs can use such a wide vectors. Intel E-cores use a similar approach to...
Yesterday at 4:34 AM
N
naukkis
replied to the thread
Discussion
Speculation: Zen 4 (EPYC 4 "Genoa", Ryzen 7000, etc.)
.
I'm pretty sure that they won't. It does not make any sense. Instead they are probably introducing some of their older Zen3 7nm-chips to...
Jul 16, 2024
N
naukkis
replied to the thread
Question
Are scalable vectors the ultimate solution to fixed width SIMD?
.
I haven't named them. Vector cpu load vectors either from continuous memory location or non-continuous indexed memory. After that costly...
Jul 3, 2024
N
naukkis
replied to the thread
Question
Are scalable vectors the ultimate solution to fixed width SIMD?
.
Not everything is worth vectorization but vector ISA does about everything to get most of vectorization. Vector cpus do data reordering...
Jul 3, 2024
N
naukkis
replied to the thread
Question
Are scalable vectors the ultimate solution to fixed width SIMD?
.
RVV is vector ISA not SIMD. You are talking about SIMD arch, vector ISA is totally different. Vector ISA vectors aren't fixed size but...
Jul 3, 2024
N
naukkis
replied to the thread
Question
Are scalable vectors the ultimate solution to fixed width SIMD?
.
The complex addressing mode the worse performing it usually is. If compiler does some optimization like loop unrolling it pretty much...
Jun 30, 2024
N
naukkis
replied to the thread
Question
Are scalable vectors the ultimate solution to fixed width SIMD?
.
Let's see what Intel optimization guide has to say: "The micro-op queue decouples the front end and the out-of order engine. It stays...
Jun 30, 2024
N
naukkis
replied to the thread
Question
Are scalable vectors the ultimate solution to fixed width SIMD?
.
There isn't such a limitations involved. What that immediate size will make is that data localisation is effectively used. If code...
Jun 29, 2024
N
naukkis
replied to the thread
Question
Are scalable vectors the ultimate solution to fixed width SIMD?
.
Those general code optimization rules will stay as long as hardware is page-based. Optimal data access patterns are full pages as those...
Jun 29, 2024
N
naukkis
replied to the thread
Question
Are scalable vectors the ultimate solution to fixed width SIMD?
.
Fastpath only needs to calculate lower address. Whether if ISA supports longer immediates or don't well written code is optimized to...
Jun 29, 2024
N
naukkis
replied to the thread
Question
Are scalable vectors the ultimate solution to fixed width SIMD?
.
That's near pointer from cpu hardware point, operating in 4KB range. RV design is very well done, there's not much which could be done...
Jun 29, 2024
N
naukkis
replied to the thread
Question
Are scalable vectors the ultimate solution to fixed width SIMD?
.
Yeah, those simple predictable addressing modes can be handled pretty much with fixed function logic. But hardware needs massive...
Jun 29, 2024
Loading…
Loading…
Loading…
Loading…
RESOURCES
Advertising
Cookies Policies
Privacy
Term & Conditions
FOLLOW
Facebook
Twitter
Top
Bottom
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
Accept
Learn more…