Hello all,

I saw the post about in this forum about Complete app_info.xml for Windows 64 bit and understand how to write the version number for an app in the app_info.xml file with, lets say 102.10-- that would be 1021.

However, how would one write the file version number for the cuda55 app 101.12?
Would that be 10112? I just want to make sure the app_info.xml file is correct for my computer. Thank you all for your time.

Version itself in xml is not important (version_num tag). Only file names are important. Version in file name is always in form VVVRR so 102.10 is 10210 in file name. 101.12 is 10112 in filename etc.

PS: I have problem viewing some posts with tags in forum.

Thank you very much for the clarification.

Can you improve the performance of Nvidia cards by increasing number of simultaneous threads in Boinc?

Can you improve the performance of Nvidia cards by increasing number of simultaneous threads in Boinc?

Woo boy...that question is NOT a yes or no answer! At some projects yes it is better to run multiple units at once, at other no it is not helpful at all, in fact it can actually slow down the crunching. It depends on your gpu too, so a high end gpu might be able to run 3, or more, units at once and do just fine while a lower end gpu would not be able to run more then one unit at a time. The first step is to get a small program called 'gpu-z' and run it, on the first tab it will show you lots of specs about your gpu, on the 2nd tab it will show how many resources your gpu is using while crunching, and that is the key. Gpu-z can be gotten here:
http://www.guru3d.com/files_details/gpu_z_download_techpowerup.html

You are looking for both the 'gpu load' and the 'memory usage' lines. If your gpu load is above 70% then you will probably not be able to run a 2nd, or more, units as the card is almost out of resources. The lower it is the better and it is worth trying. Each gpu has built in memory, 1gb, 2gb etc how much of this is used is displayed on the 'memory usage' line. If more then 50% of the total is used when running a single unit then running multiple units will not be helpful and will actually slow things down.

What you are asking is to use your gpu closer to its maximum possibilities and possibly burn it up. LOTS of people do it, but like overclocking it is a choice that should not be taken lightly. The idea is to push it CLOSE to the maximum without actually getting to the maximum, finding the edges of that still safe zone is all about trial and error. What works at one project may not work at all at another project as all units are different. Some projects even have different kinds of units, it may work on one kind but not on another kind.

Thanks mikey!

I was mostly trying to get someone involved with the admin of this project to respond here on the forums.

Worried that this BOINC project is just another graduate level workshop class that no one really cares to let us know what is actually going on.

The project already lost a lot of support when they failed to respond to EmSti.

Here is more details: http://asteroidsathome.net/boinc/forum_thread.php?id=275

Are we allowed to point out failures in this project?

Or should we be quiet and trust that our money spent on electricity is being used wisely?

I have already explained what was going on. There was just bad one batch of workunits for an asteroid so it had to be canceled. What was computed and validated was also credited. WUs had more users and I can't search everyone and give them credit for it. It was only second problem of that kind.
And I am sorry for not responding there.