A question of manners

[Wrish]

What i would like out of PG, is a test program, similar to stresscpu. One that could download a special workunit that only ran to maybe 10 frames but absolutely hammered your system, with available workunits for Uni (all cores), SMP (all cores) and GPU (all cores), and then gave a result, so you could test for instability before it became a problem.

There are already great tools for preliminary instability testing: LinX, OCCT, memtest86+, OCCT-GPU, Furmark, MemtestG80... even that Gromacs based stressor. The only caveat is to test in actual conditions - that means not setting two fans at high because LinX or Furmark runs too hot, then settling for one medium fan for FAH to run hours on end, where heat eventually causes the error you were avoiding.

What you do after passing the relevant stresses is to simply run FAH for a trial period of like 24-48 hours. During that period you monitor the client extra-often and try to catch all the intermittent problems like unstable electrical delivery, EMI, temp spikes. After that you have an acceptable level of stability. You can set the system to auto-restart on error and auto-launch FAH, but don't rely on this too often because the auto-launch sometimes fails. You can never assume the client to run stably forever because dust will eventually build up, fans may fail, capacitors may leak, and so forth; but that's why most people will dial down from their highest OC's.

Having PG code an all-out test isn't good use of their time. All-out varies by hardware generations and even dedicated coders struggle to keep up with hardware changes. And technically all you need is FAH to run stable, not that hyper-FAH dream code, right?

[PantherX]

Personally, I have been a F@H member of almost a year and never had any problems running my Uni-processor clients(Systray) and the GPU clients (Systray). I initially wanted to yous SMP client but after I read the install Wiki, the mention of CLI made me stick with Uni-Processor. I am very happy that the SMP2 client is released as the a3 core doesn't need any CLI interface. Once the a1 core retires, the SMP2 client will be re-released without any fussy installations so I am sure that the GUI interface will allow me to setup the clients very easily.

If one were to look back when F@H started, in 2000, it would be a handful of people who started this project and within a span of almost 10 years, a handful of contributors is now in a staggering 1.36 Million and is continuously growing. The advancements Dr, Vijay and his team made are well received in the scientific community. With around 70 published papers that are peer approved, that is an achievement in itself without a doubt in my mind.

In addition, spending $700 for a new processor -as many rush to suggest- isn't so much a possibility for everyone these days. *economic crisis hinthint*

I haven't really come across anyone who tells people to change there processors(maybe I am not reading enough threads to come across this suggestion ) rather, I have seen posts where they help the individual to identify the hardware problem that is causing and the last advice is to change that defective piece of hardware.

The first is that it has to be done by hand and the people that deal with budgets are not willing to pay for that feature. The second is that the ability to transfer points from one team to another will create competition, among teams, for specific people that are very high point producers to move the rank of the team. The ability to allow these people to move their points from one team to another produces totally invalidates team rankings if you can produce points for one team and them move them to a different team, on a whim Third, I do not believe that Stanford wishes to create an marketplace where specific individuals can extort their team to stay with them, collect something to move to another team, or for a team to bribe a person away from one team to their team.

I think that this should be included in the F@H FAQs about making teams as it is very simple and highlights a lot of important points. However I would like to suggest a feature; those people who start folding with the default team, be given a single chance to move their points from the default team to a team of their choice. That way there won't be any problems as the default team doesn't represent anyone except those people didn't join a particular team.

You are welcome to add a suggestion for future improvements to the code and I'm sure that Stanford will take you suggestion into consideration in developing future code. People that do write code, do frequent this place so it may be useful

The best part of F@H is its folding community. It is extremely dynamic where people can interact with the project leaders and staff from the Pande group itself. Regarding suggestions, I have already started a thread for GPU3 features (http://foldingforum.org/viewtopic.php?f=50&t=13368) and hopefully, some of the features suggested by us will eventually make it into GPU3 if its not too late.

You can set the system to auto-restart on error and auto-launch FAH, but don't rely on this too often because the auto-launch sometimes fails.

Is this possible on Win7 32-bit? Also could the auto-launch be delayed by a fixed number of minutes so that the system warms up and is ready for folding?

In the end, I am a proud F@H member and I believe that the username chosen reflects a person's character thus I don't consider any nick to be "silly" there is always a depth to it.

[Wrish]

You can set the system to auto-restart on error and auto-launch FAH, but don't rely on this too often because the auto-launch sometimes fails.

Is this possible on Win7 32-bit? Also could the auto-launch be delayed by a fixed number of minutes so that the system warms up and is ready for folding?

Sure, the way I do it is to set the system to automatically reboot on a blue-screen, as well as to auto-login, and place a shortcut to fah6.exe in the Startup folder within the Start Menu. For kicks the system also auto-locks with a blank screen saver after a minute. If you want to delay the start by a few minutes, you'd instead place a batch file like FAH.bat inside the FAH folder, and a shortcut to FAH.bat in the startup folder. FAH.bat could read something like this:

Code: Select all

timeout 180
start /m fah6.exe -smp 8 -forceasm

This waits for 3 minutes and then opens fah6.exe in a minimized window with the specified flags.

Or you might get by having FAH as a service and thus not requiring a login at all, but I have no experience running it as a service.

By the way, auto-reboot on bsod can fail if your Vcore is too low, as the processor can hard lock and require a power cut to reset. And it's possible for FAH to refuse to start after a reboot - I've seen it give some error about already running when it isn't. You could work around this with Task scheduler, as well, replacing the whole batch file system.

Edit: For starting a systray client, the .bat file needs to be in your user folder containing the work folder, and the start path would need to be fully specified, e.g.,

Code: Select all

start /m "c:\program files\folding@home-gpu\folding@home-gpu.exe" -forcegpu ati_r700

[PantherX]

Thanks Wrish for telling me about Task Scheduler. Never knew it existed before

[Grandpa_01]

Ummm why are you blue screening woulden't it be alot easier to just back off the OC a little Yeah Right

Just Kidding actually I have tried to set SMP up as a service on Windows so it would start after BSOD or whatever but have never been able to get it to work correctley.

[bruce]

I know PG has made exceptions to their policies, but very few. They totally moved someone with half a billion points from one team to another.

As I understand it, the policy has NO EXCEPTIONS. Please provide proof that somebody's points were moved to another team. Who was it an when?

[Wrish]

Ahem, my mistake. They changed the name of the user, not the team. http://folding.typepad.com/news/2009/07/index.html
"A collaborator has donated a large amount of compute time to this project; those clients were initially running under username Anonymous/team 1. To give proper credit for the donation, we have changed the username to PDC, team 1"

[bruce]

If you started donating as Anonymous and decided to change your user name to Wrish so you would be able to find your credits, that would be a wise choice to make, but there wouldn't be able to trace the points that you had already donated as Anonymous. I don't see anything that says that any points were moved. (There were no passkeys at that time.)

Making changes to the stats database after points have already been recorded is much more complex than just adding this hour's points and if by some very remote possibility a db corruption was introduced, there would be a lot of very unhappy people. The NO EXCEPTIONS policy makes sense for a number of reasons although I understand why you'd like to see exceptions granted.

[shdbcamping]

My idea of error checking would be if a persons client scraps workunits from project x, then why keep giving out that project to them, simply mark that user id as being unstable and assign them something different, eventually it may be that the client in question cant process anything due to instability in which case there would be no work left for that client to do. But at the same time it wouldnt be tying up server resources by submitting scrapped workunits, so it would not be affecting the rest of us. At that point the only thing to be done is for the user to reconfigure their machine so it isnt unstable, and lets face it most of this comes down to badly maintained hardware, bad hardware or overclocking. For a person to start folding again on that hardware they would have to go to a web page on stanford and input there username and passkey to reset the system. But if a person repeatedly did this and it failed, then they would be locked out until a PG member reset there status, which brings me to point 2.

What i would like out of PG, is a test program, similar to stresscpu. One that could download a special workunit that only ran to maybe 10 frames but absolutely hammered your system, with available workunits for Uni (all cores), SMP (all cores) and GPU (all cores), and then gave a result, so you could test for instability before it became a problem. Something like that with ihaque's memory checker also built in, so basically a test suite for people who want to run folding. If it comes up ok at the end theres a good chance you can fold without an issue. If it doesnt well you attempt to fix the problem, or post on here asking for help.

Optimised work units for optimised cpu cores would be nice as well, but its a lot of work, and wouldnt you all rather have stats that work, work units that are available to download, clients that are stable.

In the end if you want to pay for Vijay to have a codemonkey to add new backend functionality, no one is stopping you. Make a donation, and im sure it can be sorted out. At present i think the system is fine for the most part, and only really want an SMP3 client that runs on 32bit linux clients.

Who was slammed on this post? Just asking as it's pretty vague.