Hi all,
I'm suffering from not getting new tasks, thus I'm thinking of setting up a backup Boinc project that could make use of the idle time. Any advice on which Boinc project can be the best in terms of giving the highest RAC for GPU tasks?
Thanks a lot!

Hi all,
I'm suffering from not getting new tasks, thus I'm thinking of setting up a backup Boinc project that could make use of the idle time. Any advice on which Boinc project can be the best in terms of giving the highest RAC for GPU tasks?
Thanks a lot!

You're only interested in credit so I'm not interested in helping you.

You are aware that credit is worthless aren't you? The point is to contribute to science although sometimes I wonder whether that's what I'm doing.

I quite agree with Betting Slip on the credits. They are useful to me only for comparing GPU cards and drivers. Otherwise, you can't buy anything with them, and I don't look at them.

But to give you something to do, try POEM. They are having a bit of a server problem at the moment, and don't give out much information on what they are accomplishing, but it will be an educational experience.

Hi all,
I'm suffering from not getting new tasks, thus I'm thinking of setting up a backup Boinc project that could make use of the idle time. Any advice on which Boinc project can be the best in terms of giving the highest RAC for GPU tasks?
Thanks a lot!

You're only interested in credit so I'm not interested in helping you.

You are aware that credit is worthless aren't you? The point is to contribute to science although sometimes I wonder whether that's what I'm doing.

Thank you for giving out your options, but I think that you've misunderstood me.
I believe we're holding the same goal, and that's why we're here, contributing to GPUgrid although this might not be the project which gives the most credits. Meanwhile, it's not even able to deliver enough work units all the time which is definitely not a good news for those who only care about credits. If I'm that kind of person, then I believe I won't at all choose the GPUgrid as my main project.
The reason I love GPUgrid is the fact that it contributes to the medical science which can probably save lifes. The reson now I suffer is that the idle time makes my GPU power wasted which can definitely contribute more.
Thus, my main point is to find a project that can take use of the idle time when there's no work available for GPUgrid, but not to find a high-credit-gaining project to replace the GPUgrid.
The reason I chose the RAC as a reference for choosing the backup project is because the fact that it may be the simplest measure. Because of the limitation of my knowledge on Physics and Astronomy, I personally am not able to judge the importantce or the level of contribution of other projects such as Einstein or SETI. If I just post a question asking "what's the project that makes the most contribution to science", I don't think people are able to provide an specific answer since every people have their own opitions or judgments on the relative importantce of different fields of science and I do believe that all boinc projects have their own unique contribution to their field of science. That can be such a subjective problem that is not able to yield a simple objective answer.
Thus, to make the question specific and easy to reach an agreement, I picked the RAC as benchmarks since generally a higher RAC means more contribution of computing powers, which I believe is the set-up of the credit system for.
In short, yes, I'm interested in credits, but not for the purpose of my own joyfullness of seeing those huge amount of credits going into my accounts. I just take the RAC as a measurement of whether the use of my GPU computing power is maximized.
Correct me if I'm wrong, and thanks again for sharing your thoughts with me. Sorry for the misleading of my original post!

I quite agree with Betting Slip on the credits. They are useful to me only for comparing GPU cards and drivers. Otherwise, you can't buy anything with them, and I don't look at them.

But to give you something to do, try POEM. They are having a bit of a server problem at the moment, and don't give out much information on what they are accomplishing, but it will be an educational experience.

Thank you for your advice!
I'm also using the credits only for the purpose of comparison, but not for collecting the credits. And yes, I agree that the credits themselves not even worth a penny, and there's no point caring how much in total it's been collected.
So you've recomended POEM, but also mentioned that some problems exist. Could you provide more details if possible? Say what's the effect of that server problem? At the time I checked their website, I found that their server also does not have any available work units to deliver. Does it relate to the problem you mentioned? And if it happens quite often, then I guess it is not albe to act as a good option for the GPUgrid since they both have the problem of not having enough tasks?
Thanks again!

So you've recomended POEM, but also mentioned that some problems exist. Could you provide more details if possible? Say what's the effect of that server problem? At the time I checked their website, I found that their server also does not have any available work units to deliver. Does it relate to the problem you mentioned? And if it happens quite often, then I guess it is not albe to act as a good option for the GPUgrid since they both have the problem of not having enough tasks?

I think they have enough work, it is just that their upload server is stuck due to "insufficient disk space", or at least that is the error message. It should be easily resolved, but there is no one there on the weekend.

They usually have work though, so it should be a good backup to GPUGrid.

So you've recomended POEM, but also mentioned that some problems exist. Could you provide more details if possible? Say what's the effect of that server problem? At the time I checked their website, I found that their server also does not have any available work units to deliver. Does it relate to the problem you mentioned? And if it happens quite often, then I guess it is not albe to act as a good option for the GPUgrid since they both have the problem of not having enough tasks?

I think they have enough work, it is just that their upload server is stuck due to "insufficient disk space", or at least that is the error message. It should be easily resolved, but there is no one there on the weekend.

They usually have work though, so it should be a good backup to GPUGrid.

Go it. I'll give POEM a try. Thanks for your helpful advice!

If you only want credits and RAC, then go Collatz. It is so overpowered in credits that I changed my email there to separate my stats. Moo wrapper gpu application gives a lot credits too but it is the least useful (scientificaly) project. Then POEM@Home gives a lot of credit and does something with proteins and crystals so SCIENCE! Then PrimeGrid idk if useful, but you can actually find a prime and enter the largest prime database.
I created my own stats which project give how much credit for same work. The 3rd table is sorted by (RAC/REC*100%). Basicaly how much credit you got divided by how much you worked.

---

I created my own stats which project give how much credit for same work. The 3rd table is sorted by (RAC/REC*100%). Basicaly how much credit you got divided by how much you worked.

Fascinating table. I'd be interested to hear more about your methodology in due course.

If you only want credits and RAC, then go Collatz. It is so overpowered in credits that I changed my email there to separate my stats. Moo wrapper gpu application gives a lot credits too but it is the least useful (scientificaly) project. Then POEM@Home gives a lot of credit and does something with proteins and crystals so SCIENCE! Then PrimeGrid idk if useful, but you can actually find a prime and enter the largest prime database.
I created my own stats which project give how much credit for same work. The 3rd table is sorted by (RAC/REC*100%). Basicaly how much credit you got divided by how much you worked.

Thanks for your detailed answer that provides information on both credit and science. Then I'll definitely go for POEM@Home. Really helpful advice!

Moo wrapper gpu application gives a lot credits too but it is the least useful (scientificaly) project.

Although not scientifically significant, cyber security is pushed along by trying to crack the current encryption standards. So important compared to things like finding the next digit of Pi or something like that nonetheless. And since BOINC credit really doesn't matter much at all, I do the Moo Wrapper on the phone, but on PCs I do the dnet client directly because it gets more work done that way than wrapped in BOINC by a significant amount (usually the OpenCL client since the NVIDIAs and CPUs have BOINC tasks to do so the secondary onboard GOU needs something to do with no conflicts within BOINC itself).

---

1 Corinthians 9:16 "For though I preach the gospel, I have nothing to glory of: for necessity is laid upon me; yea, woe is unto me, if I preach not the gospel!"
Ephesians 6:18-20, please ;-)
http://tbc-pa.org

Moo wrapper gpu application gives a lot credits too but it is the least useful (scientificaly) project.

Although not scientifically significant, cyber security is pushed along by trying to crack the current encryption standards. So important compared to things like finding the next digit of Pi or something like that nonetheless. And since BOINC credit really doesn't matter much at all, I do the Moo Wrapper on the phone, but on PCs I do the dnet client directly because it gets more work done that way than wrapped in BOINC by a significant amount (usually the OpenCL client since the NVIDIAs and CPUs have BOINC tasks to do so the secondary onboard GOU needs something to do with no conflicts within BOINC itself).

My bad. I thought they trying to crack MD5 hash algo, that would be useless becouse MD5 is already deprecated and 'broken'. They try to brute-force find a key for RC5 encryption algo. A predecessor of RC6 a candidate for AES that was rejected in favor or Rijndael (used today). I worry more about someone cracking RSA asymmetric chipper, because that is the weakest link in current cybersecurity.

And the tables are generated with XSLT template from client_state.xml, I also put link on bottom of page.

---

Moo wrapper gpu application gives a lot credits too but it is the least useful (scientificaly) project.

Although not scientifically significant, cyber security is pushed along by trying to crack the current encryption standards.

No, they're not. There is this old encryption standard and we know a certain password exists for the one encrypted message they're trying to decrypt. We know that by the point they have tried all possible ones matching the given bit size (I think it's 55 - noone uses this anymore), they will have decoded this one message. Which tells us nothing, except that it might take equally long to brute-force decrpyt the next message of this type. But that's an information which we could easily deduce from the trial rate of the current algorithm. We don't have to actually do it to know it. Real attacks on encryption standards use other vulnerabilities, though.

So I second Tomas in this regard and have personally running GPU-Grid and POEM.

MrS

---

Scanning for our furry friends since Jan 2002

So I second Tomas in this regard and have personally running GPU-Grid and POEM.

Both GPUGrid and POEM are out of work now. Should we search for prime numbers now?

So I second Tomas in this regard and have personally running GPU-Grid and POEM.

Both GPUGrid and POEM are out of work now. Should we search for prime numbers now?

The long queue is filled on a regular basis (even on weekends) here at GPUGrid to give work for the active hosts in order to achieve the shortest turnaround time possible. So don't expect thousands of unsent workuntis sitting in the queue.

I know that the queues here are filled from time to time but...
POEM is officially retired since 4th October. Let's discuss what will be the next backup project.
There seem to be no other BOINC projects in the field of chemistry or biology utilizing GPUs.

*PrimeGrid
*Milkyway
*Asteroids
*Einstein
*Moo! RC5
*Search for Extra Terrestrial Intelligence

---

Though not in the field of biochemistry, the Universe project is developing a GPU app using OpenCL. It will be first for AMD cards on Linux, and later on Windows and Nvidia cards. Hopefully a Beta will be out shorty. It is a very nice project in my opinion.

Otherwise, you can do Folding.

So I second Tomas in this regard and have personally running GPU-Grid and POEM.

Both GPUGrid and POEM are out of work now. Should we search for prime numbers now?

As far as GPUGrid is concerned you've only had one task recently which, you aborted.

I know that the queues here are filled from time to time but...
POEM is officially retired since 4th October. Let's discuss what will be the next backup project.
There seem to be no other BOINC projects in the field of chemistry or biology utilizing GPUs.

*PrimeGrid
*Milkyway
*Asteroids
*Einstein
*Moo! RC5
*Search for Extra Terrestrial Intelligence

*Albert
*Asteroids
*BURP
*Collatz Conjecture
*Einstein
*Milkyway
*Moo! Wrapper
*PrimeGrid
*SETI

There may be other projects with NV apps and some Alpha/Beta/Test/Dev GPU projects too.
Previously/years ago/now depricated The Lattic Project occasionally had some CUDA work, ibercivis had some and for a few months (a couple of years ago) WCG had GPU tasks for NV cards. Bitcoin Utopia also had some GPU work as did GPUGrid's similar fund-raising venture.

---

FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

I'm focusing my "backup GPU project" efforts on both Einstein@Home and SETI@Home. And I may do research into seeing how I can get Folding@Home to play nice with BOINC, when I have no work from GPUGRID, though I don't expect it to work.