Random Musings About the Value of CPUs vs CUDA

Message boards : Number crunching : Random Musings About the Value of CPUs vs CUDA
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 . . . 11 · Next

AuthorMessage
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51464
Credit: 1,018,363,574
RAC: 1,004
United States
Message 846700 - Posted: 30 Dec 2008, 2:44:55 UTC - in response to Message 846695.  


But then again this cuda app is really buggy so its not working quite right...


Yepp, and someone made a fool of himself releasing that crap into the open.

The question is whose gonna take the blame... the berkley boyz or Nvidia ...



3r......I had postulated that nvidia had donated to the project to force the new app through........to bolster the sales of new vid cards.....and I am sure a few went out an bought some...

I was told by certain sources that this was not the case....but I am not sure who to believe.....
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 846700 · Report as offensive
Profile Crunch3r
Volunteer tester
Avatar

Send message
Joined: 15 Apr 99
Posts: 1546
Credit: 3,438,823
RAC: 0
Germany
Message 846704 - Posted: 30 Dec 2008, 2:50:46 UTC - in response to Message 846700.  
Last modified: 30 Dec 2008, 2:55:33 UTC


3r......I had postulated that nvidia had donated to the project to force the new app through........to bolster the sales of new vid cards.....and I am sure a few went out an bought some...

I was told by certain sources that this was not the case....but I am not sure who to believe.....


I don't think that nvidia donated nor that it was their initial intend to port the app.

But I do strongly believe that the administration thought that this new app would quickly attract lots of new users (which seems to work) to squeeze money out of them...

not sure if scam is the proper translation for what i have in mind...

/edit
trickery seems to fit better

Join BOINC United now!
ID: 846704 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51464
Credit: 1,018,363,574
RAC: 1,004
United States
Message 846710 - Posted: 30 Dec 2008, 2:56:55 UTC - in response to Message 846704.  


3r......I had postulated that nvidia had donated to the project to force the new app through........to bolster the sales of new vid cards.....and I am sure a few went out an bought some...

I was told by certain sources that this was not the case....but I am not sure who to believe.....


I don't think that nvidia donated nor that it was their initial intend to port the app.

But I do strongly believe that the administration thought that this new app would quickly attract lots of new users (which seems to work) to squeeze money out of them...

not sure if scam is the proper translation for what i have in mind...



Attract new users??
I'll bet not more than a dozen...
Most of those posting here about it have been here since dirt was a new thing......


"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 846710 · Report as offensive
Profile Crunch3r
Volunteer tester
Avatar

Send message
Joined: 15 Apr 99
Posts: 1546
Credit: 3,438,823
RAC: 0
Germany
Message 846717 - Posted: 30 Dec 2008, 3:02:48 UTC - in response to Message 846710.  
Last modified: 30 Dec 2008, 3:06:35 UTC


Attract new users??
I'll bet not more than a dozen...
Most of those posting here about it have been here since dirt was a new thing......



I wouldn't count on that dirty vocal dozen (most of them won't post here... you know .. the gamer kids... see pic below lol...) :p



I've seen some postings on the bigger online news sites here in germany about it... could be quite a few more than a dozen ;)

Join BOINC United now!
ID: 846717 · Report as offensive
Profile Voyager
Volunteer tester
Avatar

Send message
Joined: 2 Nov 99
Posts: 602
Credit: 3,264,813
RAC: 0
United States
Message 846746 - Posted: 30 Dec 2008, 4:27:03 UTC

Is it dumb not to use cuda? I see gpus for less money than a memory upgrade ,$59 with what sounds like more improvment. I know nothing about graphic cards ,but am I getting this right?This was a lowend gpu 512mb 455mz core.


So don't run out and buy one?(actually I let my fingers run to new egg)
I'd still like to know if this is a good way to go, money wise.
ID: 846746 · Report as offensive
Profile Francois Piednoel
Avatar

Send message
Joined: 14 Jun 00
Posts: 898
Credit: 5,969,361
RAC: 0
United States
Message 846748 - Posted: 30 Dec 2008, 4:31:26 UTC - in response to Message 846717.  
Last modified: 30 Dec 2008, 4:43:46 UTC


Attract new users??
I'll bet not more than a dozen...
Most of those posting here about it have been here since dirt was a new thing......



I wouldn't count on that dirty vocal dozen (most of them won't post here... you know .. the gamer kids... see pic below lol...) :p



I've seen some postings on the bigger online news sites here in germany about it... could be quite a few more than a dozen ;)


if (Fanboy=1) goto Label1


With Homogenious processors coming in the pipe, i don 't think the Cuda stuff will go very far. From an enabling point of view, it is awesome that NV spend all of this time on it, it will be very easy to recompile and run with Homogenious processors with Caches ... :)
The cache part is the important part, algorythm rarely touch only one time the data (Especially Seti) ... and branches are very important.

I am focusing on other stuff than SETI right now, but I think i am fairly happy of what I am seeing, will clean the table later with the hardware, as I always do ;-)

PS: This is my very personal opinion.

:Label1
Fanboy = always right ... then Happy?
lol
ID: 846748 · Report as offensive
Cosmic_Ocean
Avatar

Send message
Joined: 23 Dec 00
Posts: 3027
Credit: 13,516,867
RAC: 13
United States
Message 846753 - Posted: 30 Dec 2008, 4:39:58 UTC - in response to Message 846590.  

I have just discovered the joys of optimized clients since CUDA. The whole thing made me thing am I getting the most out of my computers eating up electricity.
I have since swapped to the latest 'custom' apps for seti and AP on vista and 64b Linux.
I now have 4 cpu's and 1 gpu working on my main box and a further 3 cores chugging away on other computers, waiting to see the increase in points.
The SSSE3(?) optimized AP client makes a huge difference on my Quad core AMD

SSSE3 is Intel-only. AMD users are stuck with a max of SSE3 (only two S's).
Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)
ID: 846753 · Report as offensive
Profile ML1
Volunteer moderator
Volunteer tester

Send message
Joined: 25 Nov 01
Posts: 16616
Credit: 7,508,002
RAC: 20
United Kingdom
Message 846848 - Posted: 30 Dec 2008, 13:04:27 UTC - in response to Message 846677.  

after trying the CUDA stuff, it pretty well sucks if you run it on the celeron with a GT280 ... I thought the processor did not matter ...
one more marketing hype ... The CPU still does most of the work on the CUDA stuff ...

pffffff...

Who?

Usual FUD-slinging I see?

So... You've just confirmed that the Intel Celerons are crippled with too little on-chip cache and strangled by a very old FSB bottleneck.

Now try a useful test?...

Or even openly demonstrate (and release) something that is of help to everyone?


Or are you just going to play the Marketing Hype game yet again (Yawn...).

Just my opinion as ever.

Regards,
Martin

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)
ID: 846848 · Report as offensive
Profile ML1
Volunteer moderator
Volunteer tester

Send message
Joined: 25 Nov 01
Posts: 16616
Credit: 7,508,002
RAC: 20
United Kingdom
Message 846849 - Posted: 30 Dec 2008, 13:07:42 UTC - in response to Message 846748.  

... I am focusing on other stuff than SETI right now, but I think i am fairly happy of what I am seeing, will clean the table later with the hardware, as I always do ;-)

[...]

lol

As you always never do for SETI.

Yawn yawn yawn...

Cheers,
Martin

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)
ID: 846849 · Report as offensive
KWSN Sir Clark
Volunteer tester

Send message
Joined: 17 Aug 02
Posts: 139
Credit: 1,002,493
RAC: 8
United Kingdom
Message 846850 - Posted: 30 Dec 2008, 13:10:19 UTC
Last modified: 30 Dec 2008, 13:10:52 UTC

From my observations so far CUDA seems to be about level with my 1.86GHz Core2Duo time wise....nothing like the up to 10x claimed...crunching with an factory OC'd 8800GT

Rather disappointed by the claims made by SETI.
I wonder what their test rigs were
ID: 846850 · Report as offensive
Cosmic_Ocean
Avatar

Send message
Joined: 23 Dec 00
Posts: 3027
Credit: 13,516,867
RAC: 13
United States
Message 846853 - Posted: 30 Dec 2008, 13:19:43 UTC - in response to Message 846850.  

...

I wonder what their test rigs were

From my understanding, there....were no test rigs. I'm pretty sure the 3-10x speed-up was an educated guess with some suggestion from nVidia.

By the way, I have an 8800GT that I have yet to put to work. I'll wait for it to be worthwhile. I run opti's, but for a very general sense, they just about are "set it and forget it"...for a few months at least.

Right now I see that CUDA has too much micro-managing and baby-sitting to be useful.

I don't really care about the crunch times that it would take though. If it is like having a 5th core on the CPU, then that's 25% more work that can be done in the same amount of time. If it's slower, then obviously that percentage goes down, but it's still an increase nonetheless.

Who knows, I might just opt not to do it at all, which is looking somewhat likely.
Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)
ID: 846853 · Report as offensive
Profile -= Vyper =-
Volunteer tester
Avatar

Send message
Joined: 5 Sep 99
Posts: 1652
Credit: 1,065,191,981
RAC: 2,537
Sweden
Message 846859 - Posted: 30 Dec 2008, 13:31:02 UTC - in response to Message 846850.  
Last modified: 30 Dec 2008, 13:44:49 UTC

From my observations so far CUDA seems to be about level with my 1.86GHz Core2Duo time wise....nothing like the up to 10x claimed...crunching with an factory OC'd 8800GT

Rather disappointed by the claims made by SETI.
I wonder what their test rigs were


Well so far i'm not disappointed.

My trusty old 8800GTX made in 2006 peforms 430 Gflop with Cuda-Z.

Regular 14.x credit wu's are taking 4 minutes and 55 seconds in realtime and of these minutes and seconds the cpu needs 80 seconds to setup the data for the GPU.

It's truly cool to see BoincMgr update with 0.41x% every update and that is one in a second aproximately.

Yes i've overclocked the shadercore slightly but not that much.

But claiming 10x performance is somewhat an overstatement but not if they compare with a regular dualcore computer today, then it's achievable.

Kind Regards Vyper

_________________________________________________________________________
Addicted to SETI crunching!
Founder of GPU Users Group
ID: 846859 · Report as offensive
tfp
Volunteer tester

Send message
Joined: 20 Feb 01
Posts: 104
Credit: 3,137,259
RAC: 0
United States
Message 846865 - Posted: 30 Dec 2008, 13:52:22 UTC - in response to Message 846849.  

Hey Who? any word on an update to the Atom chipset? One that would use a lot less power would be nice.
ID: 846865 · Report as offensive
Ingleside
Volunteer developer

Send message
Joined: 4 Feb 03
Posts: 1546
Credit: 15,832,022
RAC: 13
Norway
Message 846882 - Posted: 30 Dec 2008, 14:45:38 UTC - in response to Message 846850.  

From my observations so far CUDA seems to be about level with my 1.86GHz Core2Duo time wise....nothing like the up to 10x claimed...crunching with an factory OC'd 8800GT

Rather disappointed by the claims made by SETI.
I wonder what their test rigs were

As the press-release mentions, a GTX-280 is roughly 2x faster than a i7-965 (3.2 GHz), and a GTX-280 is roughly 10x faster "than an average dual core consumer CPU (2.66 GHz AMD Phenom 995".

Now, why an "average consumer" should buy the most high-end nvidia-card is another matter...

"I make so many mistakes. But then just think of all the mistakes I don't make, although I might."
ID: 846882 · Report as offensive
KWSN Sir Clark
Volunteer tester

Send message
Joined: 17 Aug 02
Posts: 139
Credit: 1,002,493
RAC: 8
United Kingdom
Message 846910 - Posted: 30 Dec 2008, 15:46:22 UTC

Like Vyper Boinc Manager showed the % rapidly counting down in chunks but only after several hours of crunching so I don't know what's going on.

I've decided to give CUDA another go this time without any other projects vying for CPU time but I've got to get through two Astropulses and several WCG
ID: 846910 · Report as offensive
Profile Francois Piednoel
Avatar

Send message
Joined: 14 Jun 00
Posts: 898
Credit: 5,969,361
RAC: 0
United States
Message 848093 - Posted: 2 Jan 2009, 6:43:11 UTC

ZZZZZZZZZZZZZZzzzzzzzzzzzzzzzzzzzzzzzzzzzz
ID: 848093 · Report as offensive
tfp
Volunteer tester

Send message
Joined: 20 Feb 01
Posts: 104
Credit: 3,137,259
RAC: 0
United States
Message 848095 - Posted: 2 Jan 2009, 6:46:59 UTC - in response to Message 848093.  

ZZZZZZZZZZZZZZzzzzzzzzzzzzzzzzzzzzzzzzzzzz


Asleep at the wheel again?
ID: 848095 · Report as offensive
Profile Francois Piednoel
Avatar

Send message
Joined: 14 Jun 00
Posts: 898
Credit: 5,969,361
RAC: 0
United States
Message 848097 - Posted: 2 Jan 2009, 6:47:41 UTC - in response to Message 846849.  
Last modified: 2 Jan 2009, 6:54:24 UTC

... I am focusing on other stuff than SETI right now, but I think i am fairly happy of what I am seeing, will clean the table later with the hardware, as I always do ;-)

[...]

lol

As you always never do for SETI.

Yawn yawn yawn...

Cheers,
Martin


Really ? remind me what CPU is TOP1 ? (and Who is the performance guy for it?)
Your opinion is always useless and innacurate, as usual, and the lack of people running SETI is proving how annoying you can be in the forum with your buddies ...

well, where is the NV computer in the top 10? top 20? top 30?
At least, I can open my mouth, I got several times my computers on the top 1, what about you? still barking? lol ...

I will be as insolent as you are, help yourself!
I forgot, stop trying to make people think you are smart, smart people do not spend their life pissing off people for the pleasure.

PS: This is my personal opinion.
ID: 848097 · Report as offensive
Profile tullio
Volunteer tester

Send message
Joined: 9 Apr 04
Posts: 8646
Credit: 2,930,782
RAC: 1
Italy
Message 848111 - Posted: 2 Jan 2009, 8:22:13 UTC - in response to Message 848097.  

Webster Merriam is a good dictionary.
Tullio
ID: 848111 · Report as offensive
Profile Francois Piednoel
Avatar

Send message
Joined: 14 Jun 00
Posts: 898
Credit: 5,969,361
RAC: 0
United States
Message 848236 - Posted: 2 Jan 2009, 16:24:15 UTC - in response to Message 846882.  
Last modified: 2 Jan 2009, 16:28:35 UTC

From my observations so far CUDA seems to be about level with my 1.86GHz Core2Duo time wise....nothing like the up to 10x claimed...crunching with an factory OC'd 8800GT

Rather disappointed by the claims made by SETI.
I wonder what their test rigs were

As the press-release mentions, a GTX-280 is roughly 2x faster than a i7-965 (3.2 GHz), and a GTX-280 is roughly 10x faster "than an average dual core consumer CPU (2.66 GHz AMD Phenom 995".

Now, why an "average consumer" should buy the most high-end nvidia-card is another matter...


If this was true, where is the NV result on SETI?

3653198 | tasks 1 Anonymous 10,885.28 3,121,065 GenuineIntel
Intel(R) Xeon(R) CPU E5462 @ 2.80GHz [x86 Family 6 Model 23 Stepping 6] Darwin
9.6.0
3666921 | tasks 2 awit 10,471.77 1,356,226 GenuineIntel
Intel(R) Xeon(R) CPU X5482 @ 3.20GHz [x86 Family 6 Model 23 Stepping 6] Darwin
9.5.0
4015522 | tasks 3 Helli 9,532.27 4,605,948 GenuineIntel
Intel(R) Core(TM) i7 CPU 920 @ 4.22GHz [Intel64 Family 6 Model 26 Stepping 4] Microsoft Windows Vista
Ultimate x64 Editon, Service Pack 1, (06.00.6001.00)
4529500 | tasks 4 Tomohisa Miya 9,151.88 1,224,324 GenuineIntel
Intel(R) Xeon(R) CPU E5462 @ 2.80GHz [x86 Family 6 Model 23 Stepping 6] Darwin
9.6.0
4097989 | tasks 5 Anonymous 8,931.75 3,676,601 GenuineIntel
Intel(R) Xeon(R) CPU X5365 @ 3.00GHz [x86 Family 6 Model 15 Stepping 7] Darwin
8.11.1
4641997 | tasks 6 Arthur L. Smith 8,628.33 588,982 GenuineIntel
Intel(R) Xeon(R) CPU X5492 @ 3.40GHz [Intel64 Family 6 Model 23 Stepping 10] Microsoft Windows Vista
Business x64 Editon, Service Pack 1, (06.00.6001.00)
2254633 | tasks 7 ugfbill 8,461.21 5,053,593 GenuineIntel
Intel(R) Xeon(R) CPU X5482 @ 3.20GHz [x86 Family 6 Model 23 Stepping 6] Darwin
9.6.0
4558594 | tasks 8 AT3 Schuricht 8,457.64 1,022,612 GenuineIntel
Intel(R) Xeon(R) CPU E5420 @ 2.50GHz [Intel64 Family 6 Model 23 Stepping 6] Microsoft Windows Server 2008
Enterprise x64 Editon, Service Pack 1, (06.00.6001.00)
4537613 | tasks 9 steffff 8,427.75 1,209,717 GenuineIntel
Intel(R) Xeon(R) CPU X5460 @ 3.16GHz [Intel64 Family 6 Model 23 Stepping 6] Microsoft Windows Server 2008
Enterprise x64 Editon, Service Pack 1, (06.00.6001.00)
4564582 | tasks 10 Andy Lee Robinson 8,427.14 975,992 GenuineIntel
Intel(R) Xeon(R) CPU E5410 @ 2.33GHz [Family 6 Model 23 Stepping 6] Linux
2.6.18-92.1.13.el5
4515289 | tasks 11 Wild6-NJ 8,280.62 1,214,190 GenuineIntel
Intel(R) Xeon(R) CPU E5430 @ 2.66GHz [Intel64 Family 6 Model 23 Stepping 6] Microsoft Windows Server 2008
Standard x64 Editon, Service Pack 1, (06.00.6001.00)
4536563 | tasks 12 steffff 8,275.79 1,157,953 GenuineIntel
Intel(R) Xeon(R) CPU X5460 @ 3.16GHz [Intel64 Family 6 Model 23 Stepping 6] Microsoft Windows Server 2008
Enterprise x64 Editon, Service Pack 1, (06.00.6001.00)
692737 | tasks 13 dcappello 8,241.46 2,836,531 GenuineIntel
Intel(R) Xeon(R) CPU X5482 @ 3.20GHz [x86 Family 6 Model 23 Stepping 6] Darwin
9.6.0
4657209 | tasks 14 Gandelf 8,170.05 460,587 GenuineIntel
Intel(R) Xeon(R) CPU E5450 @ 3.00GHz [EM64T Family 6 Model 23 Stepping 6] Microsoft Windows Server 2003 "R2"
Enterprise Server x64 Editon, Service Pack 2, (05.02.3790.00)
4641147 | tasks 15 john deneer 7,825.93 531,765 GenuineIntel
Intel(R) Xeon(R) CPU E5462 @ 2.80GHz [x86 Family 6 Model 23 Stepping 6] Darwin
9.5.0
4221516 | tasks 16 Beren 7,793.70 2,838,146 GenuineIntel
Intel(R) Xeon(R) CPU E5462 @ 2.80GHz [x86 Family 6 Model 23 Stepping 6] Darwin
9.6.0
4181753 | tasks 17 Rob Neff (N7ORY) 7,574.77 2,029,646 GenuineIntel
Intel(R) Xeon(R) CPU E5410 @ 2.33GHz [x86 Family 6 Model 23 Stepping 6] Microsoft Windows Server 2003 "R2"
Standard Server x86 Editon, Service Pack 2, (05.02.3790.00)
4676380 | tasks 18 Gandelf 7,537.86 321,176 GenuineIntel
Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz [Intel64 Family 6 Model 26 Stepping 4] Microsoft Windows Vista
Ultimate x64 Editon, Service Pack 1, (06.00.6001.00)
1239220 | tasks 19 Eric-Extreme-Goldberg 7,452.97 2,887,810 GenuineIntel
Intel(R) Xeon(R) CPU X5365 @ 3.00GHz [x86 Family 6 Model 15 Stepping 11] Darwin
9.6.0
3440890 | tasks 20 Anonymous 7,390.89 4,321,335 GenuineIntel
Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz [Intel64 Family 6 Model 26 Stepping 4] Microsoft Windows Vista
Business x64 Editon, Service Pack 1, (06.00.6001.00)

no NV stuff in the top 10 ...

It is all marketing BS.
I really hope NV debug their SETI stuff, like this, i will be giving them a punishment in 2009 :), it will take me an hour to port to my own special project ... it is going to be bloody. I did not think NV would try to port SETI, it is extremely cache sensitive ... I am going to be dancing in my cubicle one more time ... The more they enable for CUDA, the more I ll have fun.

Ps: This is my personal opinion.
ID: 848236 · Report as offensive
Previous · 1 · 2 · 3 · 4 · 5 . . . 11 · Next

Message boards : Number crunching : Random Musings About the Value of CPUs vs CUDA


 
©2022 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.