Windows password for SMP2 Client
clifford_cooley
Arkansas, USA Member
Requirements for installation
- A computer with at least a recent dual core processor or better.
- Microsoft .NET framework v2.0 (or newer)
- Windows XP Service Pack 3 is strongly recommended.
- The Windows user account that will be running the SMP client MUST have a password. The Windows user account that will be installing the SMP client MUST have and have enough administrator rights. Blank passwords will not work and you must have a password only with those valid characters : A-Z, a-z and 0-9.
Does this mean I must insert a password for Windows login?
I have never used a password and never intend on using one if I can get the SMP2 Client working it. First thing I don't understand the requirement.
This is what I have tried.
- I've used the standard F@H Clients and completed 4 WU.
- I've used the GPU2 and completed 1 WU. However I think I have something configured wrong because I was using CPU cycles.
- I've not used any flags at this stage.
- I've currently got two CPU2 Clients running because I couldn't get the first client running with the -smp flag.
The 1st SMP2 Client is running - core_82(Amber)
The 2nd SMP2 Client is running - core_78(Gromacs)
Why is it MPICH2 needs to login to a domain to run SMP2? I don't understand this login requirement.
I may need a little help/convincing here.
0
Comments
You should also then make sure your screensaver settings don't require a password upon resume.
I may be a little behind on F@H shtuff, but if you're running the standard SMP client under Windows, you should not have to specify the number of CPU threads unless you have more than 4. I don't think SMP2 requires you to specify. At least, I never did.
I never used a password to avoid the login screen. It's a good thing the checkbox was an option. I would have been disappointed an give up if I had to login just to be able to fold.
I now have the SMP2 Client running with the "-smp -advmethods" flags. Fahcore_a3.exe was downloaded and using 100% CPU.
I must have been confused about the numbers behind the -smp flag. Are the numbers for the number of threads and not the core count?
You may have to include numbers in GPU clients to force it to use a specific GPU core, but that's not required for SMP2.
Also, what GPU are you running, and what drivers? If we know that combination, we can help minimize the CPU impact that the GPU cores have.
Outdated Catalyst 9.12
I was curious to know if this graphics card would fold anything at all. I am thinking about an upgrade for CPU and GPU.
Also, I just noticed... are you trying to run two SMP2 clients? You shouldn't. You should only run 1 SMP2 client for any number of cores greater than one in most desktops; you should run 1 GPU2 client for every graphics core.
That said, with a 4350, you're probably killing your SMP2 production by running it at all. I would abandon GPU folding and run ONE SMP2 client using the -smp and -advmethods flags. You're loading your system up way too much - you'll never return useful work units if you're running an ATI GPU client on that card and two SMP2 clients on that machine.
Now that the -smp flag is working correctly, I'm only running one SMP2 client.
I'm trying to get a grip on what each client will do. Running one type of client at a time is the only way I know to do this. Getting the GPU2 to run without using CPU cycles is next on my list. First I wait for the SMP2 client to finish.
Not that I'd actually recommend people do that, I'm a huge advocate of having strong passwords on anything you don't want hacked, cracked, broken or borked.
As far as the actual PPD scores, I'm not used to calculating them yet.
I will need to do another WU with the GPU2 client after this WU has completed. You maybe correct about the GPU2. As I stated earlier, I don't think I had it configured correctly because one of my CPU cores was maxed out through the whole WU.
To get GPU2 CPU usage low it takes adding windows system variables so it really isn't anything concerning the client.
Possibly, but I password my account anyway, so it had no bearing on me.