Search results

I'm seeing fewer lockups this week but I'm mostly getting the same 32pt WUs. Could be coinsidence. Same core as before. Version 1.55 (December 22, 2003) As for how many boxen is too many, that's up to the pocket book. My grammy used to give 10% of her wages before taxes to charity...

That's correct that the power is about the same, which is why its an advantage to run fewer fast machines than more slower machines. It slipped my attention for a while until I did the math calculating costs over a year and 2 years. Now looking "forward" to summer there is an additional...

It is summer over there so they're experiencing the same thing we do every summer. Heat and rising AC costs. I'll have to cut back quite a bit this summer as there's just no way my Apt can handle all the machines I've added. It's 10 deg out and all my windows are open to compensate. Actually...

400K :D

Yeah they'll all be multiplier locked, but that hasn't stopped people from overclocking. P4s have been multiplier locked since day one.

All new systems of mine will be P4s until they fix the SSE problem. I believe AMD has a legit bug in their SSE implementation but hopefully Stanford/Gromacs team can find a workaround.

Nice work Lockheed! Hopefully this can get wrapped into LTSP 4 so it'll work "out of the box".

Did you see my results when factoring power costs into a year of folding? They surprised me. Unless you get free power take a look. http://www.hardforum.com/showthread.php?threadid=712857

Wow! Please ask him what he averages for PPD. I'm very curious. I have 3 P4s non-HT that average 40 PPDG with the latest proteins and cores. At 3.9Ghz that'd be around 156 PPD. My Barton has been getting 45 PPDG. About 100 PPD @ 2.2Ghz.

Really? My Dell system with 2.53Ghz always seems to be a few points ahead of my 3200. It's close though. The reason I've been leaning towards Intel lately is because of the SSE problem with AMD. I'm rebooting probably 1 machine a day because of it.

Oh. I've never actually seen the graphical client in action. Guess we're in the same boat. Lol Anyhow that page lists all the proteins currently being worked on. Click on the descriptions to see what the goal of each one is. Like project 610: "This project studies the behavior of one part...

Not sure what you're referring to. There's steps and frames with F@H. Generally the more proteins the more steps. The largest I've seen is 2.5 million steps. http://folding.stanford.edu/psummary.html

My vote goes to Intel 2.4C. See this thread for details. http://www.hardforum.com/showthread.php?threadid=712857

LOL! That's quite the pr0n locker.

I'm well aware of the costs of these systems. I've done server contract work for both Sun and IBM (4-way to 64-way). Both develop high quality high uptime products and you get what you pay for. Not every project needs that level of service. Thus the rising popularity of RAIC (Redundant Array of...

That's really ambitious. Perhaps with Power5 (IBM, 8 dual core CPUs in an MCM) or Cell (joint IBM/Sony/Toshiba CPU) you could get that dense. Powering it and cooling it in the same 4U case would be prohibitive I think. Even with conservative 50Watt/Ghz you're talking about a 4KW power supply.

Actually I got it down to around $75/Ghz but who was talking about making profits on something that folds for charity? That was out of left field. "pathetic blade system" - That's a new one. LoL Someone woke up on the wrong side of the bed today. :p

There's tricks for getting high bandwidth and low latency out of "normal" switches. I can't find the link right now but there was a formula for how many switches, how many ports 64/96/128 and how many NICs each PC would need. For 1024 nodes it was something like 4 gigabit network cards in each...

Cray is getting back into the game as well with an Opteron supercomputer. Exciting times for high end computing.

The design for my "grid" is in the forums. 70 Ghz strong. Hehe

According to http://stats.zerothelement.com we broke 800K this week. WOW! That has to be a new record for the [H]orde. Congrats!!! Fold On!

Probably no less than 160 Watt per dual system. Add overhead for the networking equipment and surge protectors. 200 Watt x 1100 = 220,000 KWatt $12/mo/node = $13200/mo = $158,400/yr in Electricity I wonder how that fares against a comparable supercomputer.

WB Midnight's production. 12,200 today according to Arachnid :) http://stats.zerothelement.com/cgi-bin/folding3/render-user-detailed.pl?Id=2 Mine went up a little higher due to Stanford releasing more varied WU for the linux console.

CONGRATULATIONS [H]orde FOLD ON! (this thread has gotten wayyyyy off topic!)

Nope. Just saw that one in the OC forum as a cheap Intel Overclocker. $69 is even better. That would bring the prices down to; 4932 1 Yr 5940 2 Yr Hmmm. so how much extra for a peltier solution that will let you hit 4ghz? :D

From extrapolating 2.8 scores to 3.2ghz scores that's 115PPD. $79 Intel IS7-E Motherboard $10 video card Raise electric to $7/mo. Intel 2.4C @ 3.2Ghz 115 PPD 805 PPW 41860 PPY $431 Base cost + Electricity for a Year 500000 Desired Production for the Year 12 Number of Systems to...

I'm still thinking it was IT 2. One that doesn't make you look like as much of a geek when riding it. :p

The term addicted/addiction is thrown around loosely here. If someone really was addicted as in detrimental to their health, happiness, career, then by all means NO. They should not be setting up a farm. I don't think so, because if you took away all the home farmers the upper ranks would...

Definitely was looking on the cheap side, sub $50 range with some overclocking capability. Single channel DDR or SDRAM even. I picked up an ASRock M266 all-in-one board for $37 that can bring a 2.4c to 2.8 with little trouble. I figured that was reasonable without getting into more advanced...

WOOOT!!! :)

I was running through some calculations today, basically looking for the sweet spot (again) in folding computers. Comparing a P4 2.4C vs a AMD Barton 2500 vs an XP1700, based on the average results I've been getting with my own systems. Intel 2.4C @ 2.8Ghz 100 PPD 700 PPW 36400 PPY...

Congrats Kroz! How's that farm comming along? ;) (There is a thread for achievements though. :p )

Their pics are usually accurate. Looks like "AIRGA" which is A core.

1000 P4 @ 2.4Ghz = 72,000 PPD or 500K PPW = raising the bar for becoming #1 F@H producer /me must get back to work as I do plan on taking that title

I built a pair of Duallies before switching to lower cost blades. 2 x XP1700 @ 2Ghz. 1.65v. 15 x 133 2 x XP1700 @ 2Ghz. 1.65v. 15 x 133 Still running 24/7

WOWOW!!!

WOOOOOT and Welcome! :)

http://www.mwave.com/ They do barebones with assembly/test. Good prices.

Did you try raising the TRas? I adjusted my Barton to TRas 11 and it stopped crashing. 400Mhz DDR.

Yeah US!!!!! :D Rock those WU!