Frame Query Please?

Mancunian

First of all apologies if I'm posting this in the wrong place but I've been wondering about this before so decided to post and ask the question rather than 'suffering' (ie wondering) in silence.

If you check out http://vspx27.stanford.edu/psummary.html for currently running projects, it gives number of frames for that protein and the score and what have you right? I'm currently doing p1140_RIBO_FSpeptide_. etc and that page/URL says it should have 100 frames, if I understand it correctly, and I think I do?

However my folding client (GUI) says it's processing 250 frames! Just curious why there's a difference in what it says on Stanford's protein info page and what it says in my client - could that be why my PC seems to take a long time (?) for each WU? I know most of them are fairly large and I've got it in my preferences that I'm happy to take them (as indeed I am) but would I be better off just taking smaller ones which take less time to process but which would, of course, have smaller points awarded for them or wouldn't it make any difference?

I still sigh when reading about overclocking and stuff. I simply don't know enough about this and other techie stuff to know whether or not there's anything I could do to speed things up a bit more or simply leave things alone and leave things to their own devices? I've said it before, I do use the computer at the same time for emails, surfing, reading and writing here in the forums and even sometimes playing online games (Club Pogo mainly) so maybe that's why.

Anyway my main point for this topic was to enquire about the frames 'descrepency' (?) so I'll leave it there. Time for bed said Zebedee.

profdlp

Hmmm...

The only thing similar that I'm currently crunching is an 1141 - another 600-Pointer. It shows 100 frames both at Stanford's site and on my local Folding monitor (EMIII).

Maybe csimon will see this and have an idea. With all the computers he has going there's a decent chance he's working on one of the 1140's. (Or someone else, maybe?)

Volvo

Hey Mancunian... I relate to some of these WU's being quite time consumeing...It can be hard work to quickly compleate a WU. I have both an Athlon 64 and some P4's crunching them...
As to the Overclocking if you don't have a techie friend to cardfully guide you and you push it a bit to hard it's easy to smoke a part/parts... then somehow the PC won't fold or much else until you replace the toasted part/parts. The majority of OCers are: young, quite knowledgeable and quickly learning from their experiences. They don't always start with their best gear either. You'll seldom find an old fart like myself willing to risk his best machine unless he has a lot of faith in his knowledge or extra parts/extra funds to recover from an unplanned disaster.
Thanks for being so active in our forums. We appreciate your presence and folding. Fold on Manc
-=Volvo=-
Just folding along 24/7

profdlp

The best way to get into overclocking is by reading through some of the threads in that forum. Pretty much all of the basic questions have been asked before and the info is right there for the reading.

One thing to remember is this: Parts like CPU's and Memory are designed to operate within a certain range. Once the parts are actually constructed they are tested to see if they fall into the high or low end of that range and marked accordingly. An Athlon 64 3200+ and a 3500+ might well have come out of the same batch on the production line. One does better in the tests and gets the higher rating. The actual design of the chips is identical. If you've ever come across an article where a chipmaker makes a comment like "Yields have been very good", all he is saying is that most of the batch tested in the high end of the range and that there weren't too many flat-out duds.

There have even been instances where a a run of CPU's (I'm using them as an example; this applies to memory as well) all tested so high that the chipmaker marked some of them at the lower speed just so they would have a sufficient quantity available for sale. What overclockers do is look for a hot batch of an item, then push it to its full potential.

Where this applies to Folding is in getting the most out of your CPU. The people with massive Folding farms typically use dinky little HD's, ancient video cards, less memory than you'd want on your personal machine, and an old slow cdrom drive, (if they use one at all once the OS is installed). What they do have is the fastest CPU they can afford and a no-frills MB which can handle it. Raw CPU power is still where it's at as far as point production is concerned. Some newer WU's have greater memory demands than in the past, but if all the machine does is Fold it's not like the program is competing with anything else for memory usage.

Mancunian

Volvo wrote:

Hey Mancunian... I relate to some of these WU's being quite time consumeing...It can be hard work to quickly compleate a WU. I have both an Athlon 64 and some P4's crunching them...
As to the Overclocking if you don't have a techie friend to cardfully guide you and you push it a bit to hard it's easy to smoke a part/parts... then somehow the PC won't fold or much else until you replace the toasted part/parts. The majority of OCers are: young, quite knowledgeable and quickly learning from their experiences. They don't always start with their best gear either. You'll seldom find an old fart like myself willing to risk his best machine unless he has a lot of faith in his knowledge or extra parts/extra funds to recover from an unplanned disaster.
Thanks for being so active in our forums. We appreciate your presence and folding. Fold on Manc
-=Volvo=-
Just folding along 24/7

Hi Volvo. Many thanks for your kind message mon ami. I don't understand it very well yet either, although in the last couple of days for some strange reason things to be picking up a little and the WU completion coming down, although it does fluctuate a little depending on what I'm doing, which seems to make sense if nothing else does.

As for the overclocking, and thanks Prof for your invaluable insight too ( ) I have read the overclocking pages and have realised it's definitely NOT for me. I don't have the knowledge OR the funds either and don't really think in my case it would make much difference. Just hoping I might be able to get more out of my P4 2.4 GHz PC than currently doing but from what people have advised me, think it best just to let it do what it's doing - if it ain't broke etc?!

And finally as for the forums, it's nice to be appreciated so thank you for that too, Volvo. Very nice of you to say so and glad someone's noticed ... not that I'm the only one of course and I certainly don't do it for the glory, just to support and encourage others as they have kindly done to/for me, that's all. Anyway you can be sure I'll carry on folding (hey, just thought of a good title for a film, or do you think it's been done already? ) so take care and catch you around the forums again very soon I hope my friend.

take care and thank you once again. :ukflag:

mmonnin

Its 100 frames. Look at your log file, it has 100 frames. The GUI displays something else, maybe checkpoints.

Mancunian

Hi mmonnin. Thank you for your post. I didn't check the log file just what the GUI displays compared to what the Stanford site said, as I explained. Maybe it is checkpoints, I don't know, but having looked again now, it says Frames Completed. At the moment it's showing 178 out of 400 for p638 and this time the info on the other site agrees with it. Maybe they put 100 instead of 250? Anyway it doesn't really matter but when I get time I'll check the log file too - although not sure how much of it will make sense until I look at it in more detail.