Why use CPU on SETi@home?

Message boards : Number crunching : Why use CPU on SETi@home?
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · Next

AuthorMessage
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1682569 - Posted: 22 May 2015, 12:46:08 UTC - in response to Message 1682545.  

I'm using two of my laptops at home for crunching. An old Pentium M. Which is overclocked from 1.5GHz to 2.4GHz & runs 100% 24/7. I have actually been trying to kill the things for the past 2 or 3 years, but it just keep going!
Them my newer i3-390m I am currently only using the GPU. As I can run either run 4 threads on the CPU at 100% or I can use the GPU at 100%, but the cooling in it is not sufficient to run both at 100%.

you got me an idea going on 4 my fathers PentiumM laptop T41...btw, how did you change the multiplier?

That is why you should use TThrottle!
This way either CPU gets heat from GPU on laptop...or GPU gets it heat from CPU...
running both CPU & GPU is to hot...but using the TThrottle you equalize the top temps...& it doesn't overheat! ;)

but you also, have a faster data release with your GPU, than with your CPU...did you choose GPU crunching beacuase of that fact?

For the Pentium M overclocking I did a jumper trick on the CPU to increase the FSB from 100MHZ to 133MHz. Then used a software tool that let me overclock the FSB further. I think I still have the links for that info somewhere at home, but I am current at work.

Since I actually use the i3 laptop it works out better if I don't use the CPU for crunching. It has weird lag issues when the CPU is fully loaded. Which I don't see on any other system I use for crunching. Plus I prefer hardware cooling solutions. So if I wanted to run both I would likely just engineer something to help cooler it better. I do have a 50w peltier around somewhere from the days of overclocking old slot1 celerons.
When I get a new laptop I'll retire the i3 to do nothing but run BOINC on the CPU. Running 4 CPU threads gets more work done in a day than the GPU alone.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1682569 · Report as offensive
Profile Akio
Avatar

Send message
Joined: 18 May 11
Posts: 375
Credit: 32,129,242
RAC: 0
United States
Message 1682576 - Posted: 22 May 2015, 13:05:55 UTC - in response to Message 1682034.  

My 2 i5 rigs' CPU's make up around 20% of their total RAC alone without fuss so why not use them?


I have to concur with my good buddy Wiggo on this one. Although my CPU may crunch slower than my 780 TI's, it still accounts for a good bit of the workload. But on a personal note, my Voyager rig was built for S@H specifically, so it would be silly not to put my i7 processor to use...otherwise it's just overkill for internet surfing and mIRC usage. (Yes, I still frequent good 'ol IRC, hehe)
ID: 1682576 · Report as offensive
KLiK
Volunteer tester

Send message
Joined: 31 Mar 14
Posts: 1304
Credit: 22,994,597
RAC: 60
Croatia
Message 1682588 - Posted: 22 May 2015, 13:44:50 UTC - in response to Message 1682565.  
Last modified: 22 May 2015, 13:49:28 UTC

Just looked at your machines you have 2 GPU's that are decent the others , Quadro's not that fast and your Raidon mm looks a bit old

Are you using the IGPU's ?

You would Be better off buying a few morw GPU's like the 730 and put them in the Xenon than running 5 computers and save electricty .And get rid of the quadro on the I 5 and put a 700 series GPU or 2

My current machine has done 4,863,405 in nine months and up until the end of January it was only working with 1 GPU and only 50% of the cores going

You don't need TThrottle

Bionic can do the same thing in the pref and the Nvida cards can use there own programs and all you have to do is turn the GPU fan up but instead of running it at 85% just turn one core off and turn up the fan on the GPU you will get a much higher R.A.C and what ever cores you have running can run 100% of the time and so can GPU's

Well actually, Quadro2000 is faster than 730GT...both runing a single job, get them on 90% of GPU capacity! But can't mess with it cause in the office coputer, that is BitLocker locked...
Radeon 34xx is too old!
Few days ago a new intelHD & nVidia 820M was added... ;)

TThottle is put only on laptops...belive me, it's better that than finding a machine reboted...or shut down 'cause of the heat! ;)

my GPUs are running single job...gets these GPUs up to 90% of their compute capacity...so I don't bother to use whole 100%!
if it was less than 80% I would think about it...
but Quadro2000 has a cooler of their own...& 730GT is heatsink only...but never gets over 45-50C...both in desktops! ;)

I'm using two of my laptops at home for crunching. An old Pentium M. Which is overclocked from 1.5GHz to 2.4GHz & runs 100% 24/7. I have actually been trying to kill the things for the past 2 or 3 years, but it just keep going!
Them my newer i3-390m I am currently only using the GPU. As I can run either run 4 threads on the CPU at 100% or I can use the GPU at 100%, but the cooling in it is not sufficient to run both at 100%.

you got me an idea going on 4 my fathers PentiumM laptop T41...btw, how did you change the multiplier?

That is why you should use TThrottle!
This way either CPU gets heat from GPU on laptop...or GPU gets it heat from CPU...
running both CPU & GPU is to hot...but using the TThrottle you equalize the top temps...& it doesn't overheat! ;)

but you also, have a faster data release with your GPU, than with your CPU...did you choose GPU crunching beacuase of that fact?

For the Pentium M overclocking I did a jumper trick on the CPU to increase the FSB from 100MHZ to 133MHz. Then used a software tool that let me overclock the FSB further. I think I still have the links for that info somewhere at home, but I am current at work.

Since I actually use the i3 laptop it works out better if I don't use the CPU for crunching. It has weird lag issues when the CPU is fully loaded. Which I don't see on any other system I use for crunching. Plus I prefer hardware cooling solutions. So if I wanted to run both I would likely just engineer something to help cooler it better. I do have a 50w peltier around somewhere from the days of overclocking old slot1 celerons.
When I get a new laptop I'll retire the i3 to do nothing but run BOINC on the CPU. Running 4 CPU threads gets more work done in a day than the GPU alone.

don't know if I havce a jumper for that on laptop? but will check! ;)
please send to inbox any info you have!

that would be noisy! :D
maybe if you remove the keyboard & screen you can run it with big cooler & fans? put laptop on a docking station... ;)


non-profit org. Play4Life in Zagreb, Croatia, EU
ID: 1682588 · Report as offensive
Darth Beaver Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Avatar

Send message
Joined: 20 Aug 99
Posts: 6728
Credit: 21,443,075
RAC: 3
Australia
Message 1682596 - Posted: 22 May 2015, 14:18:49 UTC - in response to Message 1682588.  
Last modified: 22 May 2015, 14:22:32 UTC

Well actually, Quadro2000 is faster than 730GT...both runing a single job, get them on 90% of GPU capacity


well nvida web site says the quadro has 192 cuda cores and if it's faster than the 730 then the 730 must be the crap version there are 3 versions , 1 with 96 cuda cores 1 with 384 and 64 bit and the good one has 384 cuda cores and ddr5

My 650's have ddr5 and 384 cuda cores

my GPu's run at 99% with 3 units and only 45% with one unit so you wouldn't be able to do 2 units on the Quadro or the 730 without the time blowing out to much

So some GPU's arn't as fast as CPU multi core chips unless your going to by a good card and why there is no reason to get rid of the CPU type units . My AMD FX will do more than both those GPU cards i reackon .
ID: 1682596 · Report as offensive
spitfire_mk_2
Avatar

Send message
Joined: 14 Apr 00
Posts: 563
Credit: 27,306,885
RAC: 0
United States
Message 1682662 - Posted: 22 May 2015, 17:03:44 UTC - in response to Message 1681725.  

Given that "most" (if not almost all) projects on BOINC use CPU for their calculations...the question comes to my mind why use it on SETi@home also?

Especially, when GPU are so awesome in calculating at SETi@home:
1. they don't clog up the GPU that it's unusable - like other projects...so you can crunch almost all the time (exceptions are in gaming or 3D CAD programs, in my examples)
2. they are so much powerful than CPUs (but not in all areas, just in some)

Example:
my SETi@home grid consists of 4 computers, in which only 2 are crunching on SETi@home (other don't have enough minimum requirements for calculating, either by RAM or GPU version)...in those 2 computers there are 3 GPUs, not the latest kind, but they crunch 24/7...but the put me in top 5% contributators on SETi@home!

So my question comes to:
Why use SETi@home on CPUs at all?
Why not switch SETi@home on GPUs all together?
;)

When S@H came out, there was NO GPU CRUNCHING.
ID: 1682662 · Report as offensive
KLiK
Volunteer tester

Send message
Joined: 31 Mar 14
Posts: 1304
Credit: 22,994,597
RAC: 60
Croatia
Message 1683216 - Posted: 23 May 2015, 10:11:03 UTC - in response to Message 1682596.  
Last modified: 23 May 2015, 10:16:32 UTC

@spitfire_mk2
When SETi@home came out, you didn't have multi-core machines...nor a 3D graphics CUDA or OpenCL to do calculations...just a beginning of OpenGL!
also, you didn't have DSL...or optics...or Solar sails...
and we didn't land on the asteroid...nor did we have NEO-WISE...
we also didn't know that stars move closer & away from our Sun...we never seen Magnetar...nor did we see a Black hole horizon...nor did we search for Pulsar on SETi@home...

your meaning is what? that SETi@home shouldn't evolve?

Well actually, Quadro2000 is faster than 730GT...both runing a single job, get them on 90% of GPU capacity


well nvida web site says the quadro has 192 cuda cores and if it's faster than the 730 then the 730 must be the crap version there are 3 versions , 1 with 96 cuda cores 1 with 384 and 64 bit and the good one has 384 cuda cores and ddr5

My 650's have ddr5 and 384 cuda cores

my GPu's run at 99% with 3 units and only 45% with one unit so you wouldn't be able to do 2 units on the Quadro or the 730 without the time blowing out to much

So some GPU's arn't as fast as CPU multi core chips unless your going to by a good card and why there is no reason to get rid of the CPU type units . My AMD FX will do more than both those GPU cards i reackon .

it's this one: http://www.techpowerup.com/gpudb/b2007/asus-gt-630-silent-2-gb.html
only rebranded to 730 GT...

did you check that out? does it really outperform the GPUs?


non-profit org. Play4Life in Zagreb, Croatia, EU
ID: 1683216 · Report as offensive
Darth Beaver Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Avatar

Send message
Joined: 20 Aug 99
Posts: 6728
Credit: 21,443,075
RAC: 3
Australia
Message 1683266 - Posted: 23 May 2015, 14:20:49 UTC - in response to Message 1683216.  

it's this one: http://www.techpowerup.com/gpudb/b2007/asus-gt-630-silent-2-gb.html
only rebranded to 730 GT...


so it's a rebranded 630

http://www.geforce.com/hardware/desktop-gpus/geforce-gt-730/specifications

which means it's the second one on this list . NO FAN !! it's crap

did you check that out? does it really outperform the GPUs?


Simple maths your card is doing units from aproxx 35 min to 78 mins and you only do 1 unit . so my AMD does them in aproxx the same times and it will do 8 at a time so yes my AMD Chip will do more than your GPU it is the 4 ghz version and that is not over clocked that is the speed out of the box
My GPU does 1 unit in aprox 15 min on average it's doing 3 at a time now and takes from 29 min to 1 hr still faster than yours does .
ID: 1683266 · Report as offensive
Darth Beaver Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Avatar

Send message
Joined: 20 Aug 99
Posts: 6728
Credit: 21,443,075
RAC: 3
Australia
Message 1683275 - Posted: 23 May 2015, 14:48:33 UTC

Klik say you buy a dual xenon chips system and the xenon chip is the ones with 16 threads per chip that means a system like that will have 32 units going at one time and that's without even adding a GPU . Even just one xenon chip with 16 threads will out perform most single GPU's
ID: 1683275 · Report as offensive
KLiK
Volunteer tester

Send message
Joined: 31 Mar 14
Posts: 1304
Credit: 22,994,597
RAC: 60
Croatia
Message 1683336 - Posted: 23 May 2015, 16:42:26 UTC

1. this is not a pissing contest...
2. it's the fastest I could find on server PCIe x8!

& it depends on which Xeon do u buy... ;)


non-profit org. Play4Life in Zagreb, Croatia, EU
ID: 1683336 · Report as offensive
Darth Beaver Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Avatar

Send message
Joined: 20 Aug 99
Posts: 6728
Credit: 21,443,075
RAC: 3
Australia
Message 1683509 - Posted: 24 May 2015, 0:22:26 UTC - in response to Message 1683336.  
Last modified: 24 May 2015, 0:32:26 UTC

1. this is not a pissing contest...


No it's not your right but you statement about only using Gpu's is wrong to out do a Dual Xenon system without a Gpu you need mutiple Gpu's and they will have to be very high end ones and you will need more than 1 more like about 8 , 4 X 8 = 32 . And to do 4 units on a GPU you need the very high end version which will set you back something along the lines of $400 plus ozzie per GPU , big bucks

There is a friend that has only 3 GPU's on his computers and they are Xenon with 12 threads and he is currently getting 90,000 as a rac and the GPU's are GTX 750 2 on 1 machine and 1 in another machine and he did not pay much for them second hand for the computers . His GPU's are new and 1 has just Stuffed up so he is back to 0 GPU on that machine .
So for a comparison

His dual cored Xenon with 24 cores and 2 750 GPU's is getting 38,000

That's double what i'm getting he has GTX 750 i have GTX 650 yet 18,000 more ! and that is not just because he has better GPU's they are not 100% faster than my GTX 650's more like 30% faster , 50% max yet he is getting 38,000 i'm getting aprox 18,300 , his CPU Xenon's run at 2.67 ghz my AMD runs at 4 ghz.
It just boils down to how many units at 1 time he can do i can do 8 he does 24 on the CPU's

Edit : my friend is being checky he must have brought a couple of 750 as both machines have 2 now but it doesn't change the example
ID: 1683509 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13736
Credit: 208,696,464
RAC: 304
Australia
Message 1683525 - Posted: 24 May 2015, 1:35:56 UTC - in response to Message 1681769.  

NONE of my computers has a graphics card. I'm not a gamer, I don't do anything that NEEDS a GPU or other graphics card, so I'm not going to spend the money for one, no matter how inexpensive they are.

That's pretty much the reason for using the CPU- probably 90% (or more) of desktop PC systems sold don't have an addon video card. They used to have onboard graphics, these days it's on-die graphics.
And ever for those of us that do have addon video cards; If you've got it (a CPU that is), you might as well make use of it.
Grant
Darwin NT
ID: 1683525 · Report as offensive
Profile betreger Project Donor
Avatar

Send message
Joined: 29 Jun 99
Posts: 11361
Credit: 29,581,041
RAC: 66
United States
Message 1683530 - Posted: 24 May 2015, 2:16:33 UTC - in response to Message 1683525.  

NONE of my computers has a graphics card. I'm not a gamer, I don't do anything that NEEDS a GPU or other graphics card, so I'm not going to spend the money for one, no matter how inexpensive they are.

That's pretty much the reason for using the CPU- probably 90% (or more) of desktop PC systems sold don't have an addon video card. They used to have onboard graphics, these days it's on-die graphics.
And ever for those of us that do have addon video cards; If you've got it (a CPU that is), you might as well make use of it.

Very true if you have a resource use it. As an important aside VLARs are only CPU doable and they very well could contain what we are looking for.
To reiterate what Grant said crunch on with what you have, it is a big hay stack we are searching thru.
ID: 1683530 · Report as offensive
Mark Stevenson Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 8 Sep 11
Posts: 1736
Credit: 174,899,165
RAC: 91
United Kingdom
Message 1683589 - Posted: 24 May 2015, 5:30:27 UTC - in response to Message 1683509.  
Last modified: 24 May 2015, 5:34:32 UTC

1. this is not a pissing contest...


No it's not your right but you statement about only using Gpu's is wrong to out do a Dual Xenon system without a Gpu you need mutiple Gpu's and they will have to be very high end ones and you will need more than 1 more like about 8 , 4 X 8 = 32 . And to do 4 units on a GPU you need the very high end version which will set you back something along the lines of $400 plus ozzie per GPU , big bucks

There is a friend that has only 3 GPU's on his computers and they are Xenon with 12 threads and he is currently getting 90,000 as a rac and the GPU's are GTX 750 2 on 1 machine and 1 in another machine and he did not pay much for them second hand for the computers . His GPU's are new and 1 has just Stuffed up so he is back to 0 GPU on that machine .
So for a comparison

His dual cored Xenon with 24 cores and 2 750 GPU's is getting 38,000

That's double what i'm getting he has GTX 750 i have GTX 650 yet 18,000 more ! and that is not just because he has better GPU's they are not 100% faster than my GTX 650's more like 30% faster , 50% max yet he is getting 38,000 i'm getting aprox 18,300 , his CPU Xenon's run at 2.67 ghz my AMD runs at 4 ghz.
It just boils down to how many units at 1 time he can do i can do 8 he does 24 on the CPU's

Edit : my friend is being checky he must have brought a couple of 750 as both machines have 2 now but it doesn't change the example


If that's me your talking about Glenn 1 of the 750's in machine ID: 7403731 "let go " and had the stinky smoke from it ;-) . And the Xenon's didn't cost me that much I got 5 machines off Ebay for silly £'s and got 2 up and running to see if I could , sumthing to do 1 was sort of used for parts and aint been botherd to "play " with the other 2 yet . Other things are more important . I used to have 3 750's in that 2 in the other Xenon and the single is in my Alienware X-51 , that crunches ok +still plays Elite Dangerous
ID: 1683589 · Report as offensive
OzzFan Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Apr 02
Posts: 15691
Credit: 84,761,841
RAC: 28
United States
Message 1683591 - Posted: 24 May 2015, 5:35:58 UTC - in response to Message 1683216.  

@spitfire_mk2
When SETi@home came out, you didn't have multi-core machines...nor a 3D graphics CUDA or OpenCL to do calculations...just a beginning of OpenGL!
also, you didn't have DSL...or optics...or Solar sails...
and we didn't land on the asteroid...nor did we have NEO-WISE...
we also didn't know that stars move closer & away from our Sun...we never seen Magnetar...nor did we see a Black hole horizon...nor did we search for Pulsar on SETi@home...

your meaning is what? that SETi@home shouldn't evolve?


The project will go through the necessary evolutions it needs to in order to survive, barring any uncontrollable factors such as being de-funded.

Your suggestion to stop using CPUs for SETI@home seems to be borne out of an interest in efficiency, that on a $/watt/credit basis, GPUs are the best resource out there. But this ignores that the project isn't about using the best resources, but rather giving the public the means to partake in science by any means necessary. Otherwise, there would be no reason to expand into the mobile market with phones and tablets crunching too.
ID: 1683591 · Report as offensive
Profile tullio
Volunteer tester

Send message
Joined: 9 Apr 04
Posts: 8797
Credit: 2,930,782
RAC: 1
Italy
Message 1683629 - Posted: 24 May 2015, 8:11:45 UTC

I have a GPU card on my Windows PC and both SETI@home and Einstein@home can use it. But the CERN programs don't use any GPU and most of the time it runs them on the AMD A10-6700 CPU, which has four cores (or 4 logical processing units according to Windows 8.1). But SETI@home had started with the idea of using "unused cycles" on ordinary PCs. Noe people build home-brew supercomputers using graphic cards just to run it. Things have changed on only in hardware but in philosophy.
Tullio
ID: 1683629 · Report as offensive
KLiK
Volunteer tester

Send message
Joined: 31 Mar 14
Posts: 1304
Credit: 22,994,597
RAC: 60
Croatia
Message 1683923 - Posted: 25 May 2015, 5:42:50 UTC - in response to Message 1683509.  
Last modified: 25 May 2015, 5:48:39 UTC

1. this is not a pissing contest...


No it's not your right but you statement about only using Gpu's is wrong to out do a Dual Xenon system without a Gpu you need mutiple Gpu's and they will have to be very high end ones and you will need more than 1 more like about 8 , 4 X 8 = 32 . And to do 4 units on a GPU you need the very high end version which will set you back something along the lines of $400 plus ozzie per GPU , big bucks

There is a friend that has only 3 GPU's on his computers and they are Xenon with 12 threads and he is currently getting 90,000 as a rac and the GPU's are GTX 750 2 on 1 machine and 1 in another machine and he did not pay much for them second hand for the computers . His GPU's are new and 1 has just Stuffed up so he is back to 0 GPU on that machine .
So for a comparison

His dual cored Xenon with 24 cores and 2 750 GPU's is getting 38,000

That's double what i'm getting he has GTX 750 i have GTX 650 yet 18,000 more ! and that is not just because he has better GPU's they are not 100% faster than my GTX 650's more like 30% faster , 50% max yet he is getting 38,000 i'm getting aprox 18,300 , his CPU Xenon's run at 2.67 ghz my AMD runs at 4 ghz.
It just boils down to how many units at 1 time he can do i can do 8 he does 24 on the CPU's

Edit : my friend is being checky he must have brought a couple of 750 as both machines have 2 now but it doesn't change the example


please write in complete sentences...it's hard to read what you wanted to say...at least, you are a native speaker - I'm not!

also, Xeons cost a lot of money! ;)


NONE of my computers has a graphics card. I'm not a gamer, I don't do anything that NEEDS a GPU or other graphics card, so I'm not going to spend the money for one, no matter how inexpensive they are.

That's pretty much the reason for using the CPU- probably 90% (or more) of desktop PC systems sold don't have an addon video card. They used to have onboard graphics, these days it's on-die graphics.
And ever for those of us that do have addon video cards; If you've got it (a CPU that is), you might as well make use of it.


almost all new intel processors use intelHD...which is a GPU proc you can use! ;)

things are moving forward - not backwards...in a few years, SETi@home should be 90% on GPU power! ;)


I have a GPU card on my Windows PC and both SETI@home and Einstein@home can use it. But the CERN programs don't use any GPU and most of the time it runs them on the AMD A10-6700 CPU, which has four cores (or 4 logical processing units according to Windows 8.1). But SETI@home had started with the idea of using "unused cycles" on ordinary PCs. Noe people build home-brew supercomputers using graphic cards just to run it. Things have changed on only in hardware but in philosophy.
Tullio


thanks for understanding! future is comming...no mather do we want it or not! ;)


non-profit org. Play4Life in Zagreb, Croatia, EU
ID: 1683923 · Report as offensive
Mark Stevenson Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 8 Sep 11
Posts: 1736
Credit: 174,899,165
RAC: 91
United Kingdom
Message 1683924 - Posted: 25 May 2015, 5:47:19 UTC - in response to Message 1683923.  

1. this is not a pissing contest...


No it's not your right but you statement about only using Gpu's is wrong to out do a Dual Xenon system without a Gpu you need mutiple Gpu's and they will have to be very high end ones and you will need more than 1 more like about 8 , 4 X 8 = 32 . And to do 4 units on a GPU you need the very high end version which will set you back something along the lines of $400 plus ozzie per GPU , big bucks

There is a friend that has only 3 GPU's on his computers and they are Xenon with 12 threads and he is currently getting 90,000 as a rac and the GPU's are GTX 750 2 on 1 machine and 1 in another machine and he did not pay much for them second hand for the computers . His GPU's are new and 1 has just Stuffed up so he is back to 0 GPU on that machine .
So for a comparison

His dual cored Xenon with 24 cores and 2 750 GPU's is getting 38,000

That's double what i'm getting he has GTX 750 i have GTX 650 yet 18,000 more ! and that is not just because he has better GPU's they are not 100% faster than my GTX 650's more like 30% faster , 50% max yet he is getting 38,000 i'm getting aprox 18,300 , his CPU Xenon's run at 2.67 ghz my AMD runs at 4 ghz.
It just boils down to how many units at 1 time he can do i can do 8 he does 24 on the CPU's

Edit : my friend is being checky he must have brought a couple of 750 as both machines have 2 now but it doesn't change the example


please write in complete sentences...it's hard to read what you wanted to say...at least, you are a native speaker - I'm not!

also, Xeons cost a lot of money! ;)


Do they ? mine may not be the latest gratest Xenons going but I got 5 machines off Ebay all different spec Dell Presissions for £400 for 5 of them . I 4t they were cheep meself
ID: 1683924 · Report as offensive
Darth Beaver Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Avatar

Send message
Joined: 20 Aug 99
Posts: 6728
Credit: 21,443,075
RAC: 3
Australia
Message 1683930 - Posted: 25 May 2015, 6:16:16 UTC - in response to Message 1683924.  
Last modified: 25 May 2015, 6:18:19 UTC

Your right Mark you can get them cheap .

Klik CPU's are also developing too . You need the CPU to feed the GPU anyway so why not use it .

The projects are still voluntary and are meant to use free cycles !!!

If people become Bionic nuts as some would say there's no reason just to change the project .

I don't think there going to get rid of CPU's any time . So why not use them !!

As for the IGPU and ARM processors ...

IGPU not that fast and i would rather have more cores than lose them and have a IGPU

And ARM processors ....there a long way short of catching up with desktops
ID: 1683930 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1683933 - Posted: 25 May 2015, 6:39:35 UTC

Well, since seeing the results from my latest CPU App I was thinking of starting a thread titled something like "Why Use Low End GPUs on SETI@home?? With even a modest i7 the CPU is much faster than a lowend GPU. In numbers they quickly add up. There are a few 24 core Xeon Macs around here, just waiting for a good CPU App. Check the numbers on this i7 and be advised Those Are Not Shorties finishing in about 1.3 hours. Those tasks have an Angle Range of 0.40 meaning they are usually about 3 times as long as a shorty. 1.3 divided by 3...do the math; Computer 60533. Also compare the CPU times to the GPU.
The App is still in testing, but I believe it will be here soon. I wouldn't be surprised to see a few Windows people running Mac VMs with this App.
;-)
ID: 1683933 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1683966 - Posted: 25 May 2015, 8:08:16 UTC - in response to Message 1683923.  

almost all new intel processors use intelHD...which is a GPU proc you can use! ;)

But not necessarily Xeons!!!

http://ark.intel.com/

Claggy
ID: 1683966 · Report as offensive
Previous · 1 · 2 · 3 · 4 · 5 · Next

Message boards : Number crunching : Why use CPU on SETi@home?


 
©2024 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.