Welcome to MilkyWay@home

CUDA Application Updated

Message boards : Number crunching : CUDA Application Updated
Message board moderation

To post messages, you must log in.

Previous · 1 · 2

AuthorMessage
Profile GalaxyIce
Avatar

Send message
Joined: 6 Apr 08
Posts: 2018
Credit: 100,142,856
RAC: 0
Message 31733 - Posted: 30 Sep 2009, 6:23:16 UTC - in response to Message 31720.  

And that will continue to scale, the new ATI HD5870 should easily double the performance of a HD4890 at Milkyway. And when the next nvidia generation arrives, I'm quite sure it will do much more than to double the DP performance of a GTX285 ;)

Holy Batman! So I've just spent months building up a nice credit total with my farm of ATIs, and then the new ATIs/nvidias come along and just one of them will get the same credit total in five minutes <wow> ;)



ID: 31733 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Paul D. Buck

Send message
Joined: 12 Apr 08
Posts: 621
Credit: 161,934,067
RAC: 0
Message 31738 - Posted: 30 Sep 2009, 7:30:31 UTC - in response to Message 31733.  

And that will continue to scale, the new ATI HD5870 should easily double the performance of a HD4890 at Milkyway. And when the next nvidia generation arrives, I'm quite sure it will do much more than to double the DP performance of a GTX285 ;)

Holy Batman! So I've just spent months building up a nice credit total with my farm of ATIs, and then the new ATIs/nvidias come along and just one of them will get the same credit total in five minutes <wow> ;)

As it has always been...

I remember having a single core and dreaming of the day when I would get dual cores, then quads, now I have mostly i7s and dream of i9s with 12 instead of 8 CPUs ... 3 PCI slots instead of 2 so I can add more GPUs even if they are not the fastest GPUs around ... it will still be more cores ...

Of course, by then I will be dreaming of 16 cores ... :)
ID: 31738 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile verstapp
Avatar

Send message
Joined: 26 Jan 09
Posts: 589
Credit: 497,834,261
RAC: 0
Message 31745 - Posted: 30 Sep 2009, 11:40:58 UTC
Last modified: 30 Sep 2009, 11:43:35 UTC

Well I remember when I had my Radio Shack Model 1 clone, Z-80 @ 1.0MHz, and a friend turned up one day with a soldering iron and knife, cut a couple of tracks and soldered a couple of connections, and suddenly it was running at 1.5MHz. Gee, that was fast. Also, no bios overclocking in those days.
Cheers,

PeterV

.
ID: 31745 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Thamir Ghaslan

Send message
Joined: 31 Mar 08
Posts: 61
Credit: 18,325,284
RAC: 0
Message 31751 - Posted: 30 Sep 2009, 13:45:29 UTC - in response to Message 31720.  

That's why other projects are "whining" about the overpay at MW. For most of them it's just impossible to utilize the hardware to this extent. It's not just about being lazy programmers - the algorithm and the problem itself don't allow it. They could never achieve the same Flop/s even if they made an ATI app.

MrS

That is very true.
The MW algorithm is really perfect for a GPU. Take vast amounts of parallelism (millions of threads), no branching (except you want to call a loop with a counter checked each iteration a branch), a very compute intense algorithm with only a few memory accesses, minimal communication between the threads (the values are just added in the end), and what you get is virtually the peak performance of a given GPU for the instruction mix of the algorithm. It's not all about multiply-adds, so you won't get exactly peak performance. But the v0.20 has cut all the overhead down to a minimum so you really arrive within 10% of what is theoretically possible with the algorithm's instruction mix. That's better than any current CPU achieves, even relative to its peak performance.

And that will continue to scale, the new ATI HD5870 should easily double the performance of a HD4890 at Milkyway. And when the next nvidia generation arrives, I'm quite sure it will do much more than to double the DP performance of a GTX285 ;)


Far be it that I'd like to start a flame war.

Maybe this analogy is appropriate, maybe not. But lets suppose I'm running Crysis and get 10 FPS (Credit or WU), and I'm running another game and get 100 FPS. For the sake of argument lets suppose in both scenarios the CPU is 0% and the GPU is at 100%.

My point? Not sure exactly, just making an analogy.

I can imagine a boinc future where GPU users will cry foul for quantum computer owners running quantuamilly optimzed apps. :P

ID: 31751 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Paul D. Buck

Send message
Joined: 12 Apr 08
Posts: 621
Credit: 161,934,067
RAC: 0
Message 31761 - Posted: 30 Sep 2009, 17:34:58 UTC - in response to Message 31751.  

Maybe this analogy is appropriate, maybe not. But lets suppose I'm running Crysis and get 10 FPS (Credit or WU), and I'm running another game and get 100 FPS. For the sake of argument lets suppose in both scenarios the CPU is 0% and the GPU is at 100%.

100 feet per second ... yeah that will be pretty hard to catch up to ... do you think at that speed it will still be able to miss the walls?
ID: 31761 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ExtraTerrestrial Apes
Avatar

Send message
Joined: 1 Sep 08
Posts: 204
Credit: 219,354,537
RAC: 0
Message 31764 - Posted: 30 Sep 2009, 19:21:52 UTC - in response to Message 31751.  

Far be it that I'd like to start a flame war.


Yeah, I didn't think so either. But it just happens far too easily on these forums.

MrS
Scanning for our furry friends since Jan 2002
ID: 31764 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile verstapp
Avatar

Send message
Joined: 26 Jan 09
Posts: 589
Credit: 497,834,261
RAC: 0
Message 31771 - Posted: 30 Sep 2009, 20:39:40 UTC

>CPU is 0%
Attach to a cpu-only project - it stops your cpu getting bored when you gpu is busy.
Cheers,

PeterV

.
ID: 31771 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 · 2

Message boards : Number crunching : CUDA Application Updated

©2024 Astroinformatics Group