How much difference does it make?
Dragstk
Syracuse, N.Y.
Right now, I'm running an XP2100, on an Asus A7V333 with 512 of generic memory. All this is a dedicated 24/7 Folding box. I have been averaging 2WU's a day.
I see some of you are running big processors, on your Folding rigs. While I realise there is no hard and fast rule here, how much of an upgrade is required to see a difference is Folding output?
Is it better to have 2 slower rigs, vs. one big fast one?
Thanks;
Dragstk
I see some of you are running big processors, on your Folding rigs. While I realise there is no hard and fast rule here, how much of an upgrade is required to see a difference is Folding output?
Is it better to have 2 slower rigs, vs. one big fast one?
Thanks;
Dragstk
0
Comments
For P4s something that has HT. Say a 2.4 or 2.6 that OCs really well. Nice overclocking CPUs and boards are good for FAH since we can get the best for our dollar.
The more raw cpu ghz/mhz you have folding the faster it will fold.
Having 2 slower rigs or one fast one makes no difference as long as they are equal in mhz collectively. There may be advantages to 2 slower rigs but that's a whole nother issue.
So, if I were to build a 2nd box, with (example)-an XP 1800/XP1900 w/ 512 PC2700 DDR, running SSE, am I looking at the same output as my Folding set up?
What I'm looking at is (with my finances, as I'm now unemployed), I could build a box, in my example, now, or wait a couple of months and build some big thing.
It just seems to me(in my limited knowledge), that 2 similar rigs, in the long haul, is the better set up.
Let's look at three options:
1) Put together a rig running at 1.8GHz for an outlay of $300 by using some part laying around.
2) with a few selective upgrades to the stuff in box #1 you could run at 2.2GHz and the total cost would be $500.
3) Shoot for the moon and build a new system, but use very little of your existing stuff. Have a box running 2.8GHz and spend $800.
#1 $167/GHz
#2 $91/GHz
#3 $286/GHz
Go with option #2. Do you see the picture? Don't just compare CPU costs but look at hte system prices.
This gets messy if you want to compare different chip families and such.
I still have an old TBird folding (no SSE), but it is running 1.6GHz and the mobo will not support a TBred. I can buy a pally, but it will not run any faster and I will only gain SSE for my $60. I am not doing it.
Please don't take this as I am trying to dispute yours or any of the other answers. I just don't know enough about this, and am trying to learn.
Thank you;
Dragstk
What I was trying to say was that since you already have most of the parts for a system you need to look at the cost maxing it out. If you have a mobo what is the fastest cpu that it will handle? Since all of the other costs are the same, it might be worth going with the fastest CPU possible.
The diff between #1 and #2 that I had in mind was CPU, cooling and PSU. Changing/adding memory doesn't have enough impact to worry about.
I am a firm believer in putting together whatever you can and running it. I still have 2 old 133A mobo folding (my wife and Dad are running them too) and my new box is built on a 266A mobo. I don't even have any DDR memory yet. I buy used stuff and referbs. Now I just need a way to hid machines from my wife.
seriously ed ...I think you've got the right idea to just max out the meggahurtz but I'd definitely go with the SSE proc.