• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Does memory make any differance in Folding?

C

CyberSlave

[H]ard|Gawd
Joined
Jul 31, 2001
Messages
1,025
I just got my old 400mhz comp back (build momma a new system) and was wondering of adding more memory makes any differnce in folding?

Right now I am unsure of how much memory is in ti, either 64 or 128, but I do know the boad will hold a max of 384megs.

Right now the old computer is sitting on the floor waiting for a new harddrive since it has none at the moment, so I figured I might s well hrow in a bit more memory if it makes a differance.

So ,would more memory help in folding?
 
If it's using alot of swap space then yes it would help, but unless you get the memory super cheap I'd save it to go towards a new system.
 
The memory is pc100, I can get a stick of 128 off pricewatch for 15, so I might boost it a bit.
 
It depends which OS you're intending to use. Win XP has a minimum requirement of 128 MB, while Win 2000 has a minimum of 64 MB. If it's already equipped with 128MB of RAM, it won't make a very big difference in Folding. F@H is not a memory-intensive app. Get the extra RAM anyway, it won't hurt F@H and you will notice an improvement with multitasking and some apps will benefit, especially in Win XP.
 
For example, I have a 2Ghz AMD folding now which is running with 64MB RAM and Windows 2000. Production is about 60PPD when it should be 72PPD. A signifigant drop when just a few MB below what it needs, which is about 90MB. 128MB is on the way but it's folding. :)
 
Dont mean to hijack the thread but is there a place to compare PPD for different CPU's?

Turmelle - you said you 'should' be getting 72 PPD. Right now my XP 2600 (2088MHz) is getting 88 PPD with SSE turned on and 40 PPD without it. My 2.4B P4 laptop is getting 89 PPD. These totals are taken from averaging the last 5 WU's. Both are now using the -ForceSSE and -advmethods flags.

As for your question CyberSlave : I just pulled my P3 550 out of the closet and it had 64 Megs ram. I was only getting 60 PPW with it so I jumpped the ram to 256, now I am getting a blistering 60 PPW with it. :D The system is running Win 98SE with nothing else installed. The total seems a little low but what can you expect from a P3 550... I cant believe I just said that, I remember getting that CPU New and it was the greatest at the time...512K cache, Slot 1...That thing was a tank in its day. :)

Sorry for getting of track there;

Kroz
 
There are lot's of threads floating around about frame times, PPD, etc. The difficulty is getting people with dedicated boxen - 24/7 no other work - to run the EXACT SAME WU. Other than that comparison's are just approximate.
For example I've a dually PIII 933 that got the same protein for both instances. It was churning frame times of about 31 minutes, I started running some compression routines and frame times increased to 45 minutes.
With the many variables involved you just have to keep on folding and hope for the best.
 
CIWS did a lot of testing with identical WUs on different CPUs. I don't know if he kept up with the newest CPUs, but I know he tested TBred XPs and P4s.
 
Hello Folks

I'm new at this and also have an old machine with a P3 500 mhz with 192 MB ram running windows98 SE and havn't got though a full WU after almost 2 weeks of folding.

What do you mean by PPW ?
And how do I check this?

BTW I have a 1ghz AMD Athlon 384MB ram running Fedora Core 4 and I have already folded 4 WU and only had it up a couple days before the old Pentium3 PC.

:confused:
 
So, can we make a fake protien? I mean can we make a data file that would be usable on a regular basis to test system configs? Something that would change the core if needed, always run the same test quickly, give us a baseline for these systems? Kinda like running Super PI or something allong that line, but giving output info on how we fold.
 
The speeds vary from computer to computer and chip to chip. AMDs are better at some, p4s at others. You'd need ~3 WUs to test (1 for each core), and would probably have to finish all 3 to get a good benchmark.

 
One thing to keep in mind, is folding is the same as any other computer program. It needs a certian amout of memory to run efficently, at some point the program run entirely in RAM so more would not increase anything. Same as any other program, swapping a high end game, the game won't run too good, ie. too little RAM.

Knowing the OS overhead, what WU's you want to run should lead you to a solution.
 
You must log in or register to reply here.
Back
Top