If geeks love it, we’re on it

Howdy, Stranger!

You found the friendliest gaming & tech geeks around. Say hello!

Big folders who have disappeared?

TimTim Southwest PA Icrontian
edited Aug 2010 in Folding@Home
What has happened with some of the previously large scale folders we used to have here, like Technocrat and Jonshandbrake? They used to do a lot of points, and now it's virtually nothing.

Layoffs/downsizings and no more access to the computers that used to fold? Sold the folding farm and went on to other things?

Comments

  • Cliff_ForsterCliff_Forster Baltimore, MD Icrontian
    edited Mar 2010
    Okay, I'm not trollin, honestly I'm not, but I am going to ask a question in all seriousness that I think allot of folders don't bother with.

    Folding at home, if you look at it critically what it is, its the worlds least efficient supercomputer per watt. I know how non enthusiast this makes me seem, but I actually choose to power down my computers and other household appliances when I am not using them because I see that as a more effective contribution to my community not to mention my checkbook. Part of me really wanted to be in the cool kids folding club, but when I did more research on it, in ten years they have 72 papers, some of them written to simply justify the project. I'm not saying I would not love to contribute to a cure for Alzheimer's, I have seen the disease first hand and it is horrible, but I'm wondering if folding at home is an inefficient way to do it.

    Folding in my mind has also become a sort of marketing ploy from the GPU manufacturers. Its like added E-Peen for people that need that to validate how geek they are. Its something I have trouble fully understanding.

    So here is the question, E-Peen aside, folding's potential aside. On the level of moral contribution to society is it better to contribute to the least efficient computing project the world has ever known, or to reduce my carbon footprint? (yes, its a catch-22)

    No right or wrong answer, I just wonder what people are thinking.
  • _k__k_ P-Town, Texas Icrontian
    edited Mar 2010
    That is 72 peer reviewed papers, 158 papers completed dating back to 1994. The papers are from the Panda Lab which I assume if you use the program to do your work you have to publish it or write through their lab so it gets credited for the work.

    I stopped, no idea where the remainder of points are coming from. I leave 2 computers running 24/7, main rig and file server, though my wattage is way down.
  • mas0nmas0n dallas Icrontian
    edited Mar 2010
    I stopped altogether for a while and am back to just a single SMP2 client. The power usage was getting out of control.
  • ardichokeardichoke Buttes Master B Lansing, MI Icrontian
    edited Mar 2010
    Some of my coworkers crunched the number and found that to keep your average home PC running 24/7, it only costs about $0.60 in electricity per day. That comes out to about $18/mo. per computer. Of course that would fluctuate based on your hardware but given that I've been running 1 desktop and 1 laptop and my highest power bill so far has been approx. $70, I'd say that number is probably about right (keeping in mind, of course, that the computers aren't the only thing in my house eating up juice). I look at the money I spend as a contribution to charity Cliff, since I don't really give to charity otherwise with the exception of donating old clothes and whatnot to Salvation Army.
  • BobbyDigiBobbyDigi ? R U #Hats ! SoCal Icrontian
    edited Mar 2010
    Reply to Cliff: I Fold unused cycles on computers that are going to be on anyways. 2 Computers at work are on 9-5 CPU folding (slowly but surely). My rig at home is on 5:30-9+ for work, and any other time I boot the computer it pretty much stays on 'til I go to bed. It folds GPU and CPU anytime its on, unless its Wednesday night (League).

    It doesn't get me a lot of points but it helps contribute cycles that would have just gone to waste.

    -Bobby
  • Cliff_ForsterCliff_Forster Baltimore, MD Icrontian
    edited Mar 2010
    ardichoke said:
    Some of my coworkers crunched the number and found that to keep your average home PC running 24/7, it only costs about $0.60 in electricity per day. That comes out to about $18/mo. per computer. Of course that would fluctuate based on your hardware but given that I've been running 1 desktop and 1 laptop and my highest power bill so far has been approx. $70, I'd say that number is probably about right (keeping in mind, of course, that the computers aren't the only thing in my house eating up juice). I look at the money I spend as a contribution to charity Cliff, since I don't really give to charity otherwise with the exception of donating old clothes and whatnot to Salvation Army.
    I guess thats the challenge. Would Stanford be better off with a monetary contribution from every concerned folder? Would it enable them to build a better more efficient supercomputer? Less wasted energy not to mention less required bandwidth to process the calculations over the internet?
  • ardichokeardichoke Buttes Master B Lansing, MI Icrontian
    edited Mar 2010
    Doubtful, as many people who fold wouldn't actually donate the cash. They have this resource sitting there, they want to use it, so they do. My computer would likely be turned on all day long regardless of folding or not. The bandwidth is pretty negligible for an institution like Stanford, especially since it only uses bandwidth to send the work unit out and to return the completed work unit and neither the work unit or the results are very large from what I've seen.
  • TimTim Southwest PA Icrontian
    edited Mar 2010
    We donate cash in the form of paying the power bill for the extra amount that the PC is using. And for buying better GPUs that can fold more. Stanford doesn't need cash, they need the work packets to be processed correctly to get the data.

    Why is my folding sig not working right? I got it this far, what needs changed? I replaced the opening "[URL" with 4 X's so it'd show the code here.

    Here's the code in my signature control panel:

    XXXX="http://folding.extremeoverclocking.com/sigs/sigimage.php?un=Tim225&t=93"]http://folding.extremeoverclocking.com/sigs/sigimage.php?un=Tim225&t=93[/URL]
  • edited Mar 2010
    Cliff_Forster said:
    I guess thats the challenge. Would Stanford be better off with a monetary contribution from every concerned folder? Would it enable them to build a better more efficient supercomputer? Less wasted energy not to mention less required bandwidth to process the calculations over the internet?
    Cliff, you are asking valid questions that need to be investigated and answered. After being a folder myself for some time, now I am also wondering if Stanford is just wasting too much electricity worldwide for what can be done more efficiently at a central facility (and with accountability). I am wondering what percentage of their jobs submitted to Folding@Home are analyzed vs. wasted? What cure has been found during the life time of this project? If anything was found what was the cost of it vs. using a central facility. I checked their website and did not see clear answers.

    By the way, it is not correct that Folding@home is using only the wasted cycles. Modern PCs reduce power usage drastically when the computer is not in use.
  • ardichokeardichoke Buttes Master B Lansing, MI Icrontian
    edited Mar 2010
    The cost of power is not the only thing one would have to take into consideration when determining if it is better to just do it all at Stanford. There's also the cost of buying the servers, the cost of constructing a data center to house them. The cost of cooling the servers. Replacement parts when servers fail. Hiring more staff to man the data center and maintain the servers. Simple electricity costs don't even begin to cover the expenses that go into running a supercomputing data center. In fact, compared to the other costs, the electricity is pretty much negligible.
  • DonutDonut Maine
    edited Mar 2010
    Tim, I can only speculate. I know that Jon and Sally were running a large farm of P4's. The cost of upgrading that large an operation, I could only guess. Personally, I just moved on. At my peak, I think I was running 10 rigs, I just couldn't justify the upgrade cost.
    Counting my wifes' machine, I've got 3 in the house, and none of them are the latest and greatest. I still do DC projects, but not like I used to. It's not worth it to me.
  • SPIKE09SPIKE09 Scatland
    edited Mar 2010
    running 2 GPU clients, one smp, 3 uniprocessor for my home team 3 uniprocessor clients for team 93 wish i could run more for both got over 5 million points so far and yes wish i had more. wish we had a cure for alzheimers or some of the cancers being studied by F@H can't believe stanford is classed as a junior university ffs:confused:
  • Sledgehammer70Sledgehammer70 California Icrontian
    edited Mar 2010
    I use to fold at a point with 48 CPU and got myself in the top 17 people on team 93. I was gunho about folding and honestly still am. Why have I stopped or why did I stop? well about 4 years ago I left my old job which allowed me to utilize 12 Beastly work stations for their time. After I left those system still folded, but a year later that company went out of business directly and those system disappeared. Honestly I lost focus on keeping the points rolling...

    In all it is funny that I see this thread today as I did start to fold once again on 3 machines 1 is a AMD tri core system, and a Core 2 Duo Laptop and my i7 system with my GTX 295 GPU's. In all a good 11 cores spitting out numbers but this time not for team 93 :( why not team 93? well because I have found another team that will reward my systems folding in bonus money to buy products that will allow my PC to be better when I use that bonus cash to buy products. No I do not fold 24/7 like I had done before. But when my PC is on it is folding & applying points to something I will use later on.

    While my point score that stands under team 93 was good for its time, it is utter crap compared to those who pushed on & posted massive amounts of points. I look at some of the GPU clusters running now and see them breaking the 100 million point mark while folding for only 1/2 the time I did. I am amazed that every 55 minutes my GPU's will post 768 points per core... It is much better than it use to be & in general I recommend anyone who is folding to use their GPU's over their CPU's... but it never hurts to use both :)
  • QCHQCH Chicago Area - USA Icrontian
    edited Mar 2010
    Mat, it's been 4 years since you left that place. Wow, how time flies.

    For me... When single fast CPU's we the only way, I had like 50 CPU's going between work and home. When Dual (and now quad) and then GPU folding came around, my 50 was worth 6 or 7 of them newer ones. Add in the power issue and the change in point weighting going towards the GPU and SMP... It really wasn't worth it to keep fighting to keep F@H on work systems. I am still VERY interested in F@H just not a good time.
  • Sledgehammer70Sledgehammer70 California Icrontian
    edited Mar 2010
    In general these days 1 of my GPU cores produces almost 30 times more points per day. Efficiency wise I think it would be better to buy a quad SLI setup & support the program that way.

    I mean really 2 GTX 295's vs 1 i7 GPU so 4 cores vs 4 cores breaks down like this.

    GPU x 4 cores PPD: 30,496

    CPU x 4 cores PPD: 1,296

    I am sure the SMP2 clients yield a bit more points for the CPU side with the bonus point deal but those usually take longer to break down.
  • LeonardoLeonardo Eagle River, Alaska Icrontian
    edited Mar 2010
    Cliff, there is just no way Pande Lab and Stanford could come up with the necessary supercomputer time to replace the donated processing. They would literally need to design, build, power, and operate their own supercomputer. It's just untenable. As it is, they have the freedom to experiment, revise, and move around that scheduled supercomputer time would not allow. What Pande Lab has now is the world's largest supercomputer, running 24/7, no scheduling contracts.

    Yes, it certainly is inefficient running hundreds of thousands processors distributed across the world, but it gives us a chance to directly contribute, to do more than mailing a check or pulling out the credit card. We are real players. We control machines that contribute directly to basic research. For me, it is doubly satisfying, as I get to participate in a charity by means of a hobby I love.
  • Sledgehammer70Sledgehammer70 California Icrontian
    edited Mar 2010
    Leonardo said:
    Yes, it certainly is inefficient running hundreds of thousands processors distributed across the world, but it gives us a chance to directly contribute, to do more than mailing a check or pulling out the credit card. We are real players. We control machines that contribute directly to basic research. For me, it is doubly satisfying, as I get to participate in a charity by means of a hobby I love.
    I support this post...
  • DanGDanG I AM CANADIAN
    edited Jun 2010
    I know I've dropped off in production, I was pushing ~20k points a day and am down to about 5k a day. It was just getting too hot in my office at home and costing me roughly another $50 a month running a pair of OC'd i7's.
    I still have a handful of E7500's at work running dual service clients, so I'm still alive, just not as active at home anymore when it comes to PPD.
  • clifford_cooleyclifford_cooley Arkansas, USA
    edited Jun 2010
    worlds least efficient supercomputer per watt

    Lets not put down a worthy cause when there is so much wasted energy in watching movies and sports broadcasting. Is there really any need in me listing more inefficient unworthy projects. What is the power consumption in creating and viewing one movie? I do love watching a good movie but seriously which would we likely benefit from most? Honestly I can't answer whether we would benefit from Folding. I can only hope that my contribution will someday help. As for the movies, seriously don't make me laugh by saying we really need them.
  • Cliff_ForsterCliff_Forster Baltimore, MD Icrontian
    edited Jun 2010
    clifford_cooley said:
    worlds least efficient supercomputer per watt

    Lets not put down a worthy cause when there is so much wasted energy in watching movies and sports broadcasting. Is there really any need in me listing more inefficient unworthy projects. What is the power consumption in creating and viewing one movie? I do love watching a good movie but seriously which would we likely benefit from most? Honestly I can't answer whether we would benefit from Folding. I can only hope that my contribution will someday help. As for the movies, seriously don't make me laugh by saying we really need them.
    First off, lets not misunderstand that I'm saying folding at home is "wasted energy". The argument is more like this. Would the world be better off not running all those extra processing cycles? In other words, does the folding at home attempt to solve a really big problem, by causing another really big problem?

    Its a fair question. I'm not saying I have the answer, but in short, as someone that cares about the environment, and someone who did directly see someone die at the hands of Alzheimer's (its horrible), I still lean towards thinking I'd be just as well off to save that cash on the energy bill and just donate it to someone that might want to build their own centralized super computer to solve the problem. I know all the reasons that this is fundamentally impractical, I get it, I'm just saying, why not challenge ourselves a little bit. Lets call it what it is, its not un pragmatic for me to say its the worlds least efficient super computer per watt, because, that is exactly what folding at home represents.

    I admire the sentiment, I really do, and I hope something comes of it at some point, but honestly, I think environmentalism being a fair argument against my participating in the project is at very least a fair thought that most enthusiasts would not consider.

    Stanford might do the world more good by killing the folding at home project and encouraging everyone to turn their PC's off when they don't use them, getting a cut of the savings from the utilities, and put the funds into building a vastly more efficient supercomputer for solving this problem? Sure, sounds impossible, but, so does finding a cure for Alzheimers..... I'm just dreaming out loud.
  • MyrmidonMyrmidon Baron von Puttenham California Icrontian
    edited Jun 2010
    It's kind of unfair to suggest that saving energy to save the environment and spending money to help Alzheimers are comparable. It depends which one is more important to you as a person, right? It's impossible to compare - every dollar in energy you save has a measurable impact on the environment. Every dollar you spend toward Alzheimer's has an impact that you can't measure for the SOLE REASON that University research is not a guaranteed thing.

    I agree WHOLEHEARTEDLY with Cliff that "what do I spend my money on" is a fair question - only because it's so unfair to compare them. Personally, the environment is a more pressing issue to me so I don't fold, but then I've never seen what Alzheimer's can do.
  • clifford_cooleyclifford_cooley Arkansas, USA
    edited Jun 2010
    If you don't want to fold then don't but stop pointing a finger at a worthy cause when there are so many other things you can point at first that are not worthy. If you can prove that folding is not a worthy cause then I will listen. Until then I will be pointing my finger at everything else that is simply Entertainment and not a worthy cause of power consumption.

    Sure lets live without folding. Lets live with the prospect of never finding a cure. Lets point a finger at Stanford for using prebuilt PC's that does not require their full attention simply because they have chosen to use PC's that are less efficient. Thats right say no to finding cures, say no to watching movies, say no to living under AC. Just how much are you willing to live without before you decide you are living Environmentally Friendly? Where exactly is this line drawn? Cliff I admire your wish to be environmentally friendly but I have only known you for a short time and have only seen you question Folding and not any other projects or ideals that we have grown accustomed too. Perhaps there is allot I have missed.
  • BuddyJBuddyJ Dept. of Propaganda OKC Icrontian
    edited Jun 2010
    Look at the Top 10 supercomputers. You'd need four Cray XT5 systems built to Jaguar's specs to equal F@H. That's $800 million to build (just the computers) and 28 megawatts/hour to operate. So that's the power for 28,000 homes every hour.

    Folding's 327k active cores seem pretty darn practical compared to the 896k Opterons it'd take to get the same processing power.
  • Cliff_ForsterCliff_Forster Baltimore, MD Icrontian
    edited Jun 2010
    clifford_cooley said:
    If you don't want to fold then don't but stop pointing a finger at a worthy cause when there are so many other things you can point at first that are not worthy. If you can prove that folding is not a worthy cause then I will listen. Until then I will be pointing my finger at everything else that is simply Entertainment and not a worthy cause of power consumption.

    Sure lets live without folding. Lets live with the prospect of never finding a cure. Lets point a finger at Stanford for using prebuilt PC's that does not require their full attention simply because they have chosen to use PC's that are less efficient. Thats right say no to finding cures, say no to watching movies, say no to living under AC. Just how much are you willing to live without before you decide you are living Environmentally Friendly? Where exactly is this line drawn? Cliff I admire your wish to be environmentally friendly but I have only known you for a short time and have only seen you question Folding and not any other projects or ideals that we have grown accustomed too. Perhaps there is allot I have missed.
    Your taking my question completely out of context. I'm not criticizing anyone that folds, I just gave my personal reason for not doing it, and wondered if that resonated with anyone else. I mean, this thread is about people that stooped folding?
  • _k__k_ P-Town, Texas Icrontian
    edited Jun 2010
    But you never started, plus you are always trolling except not as well as Thrax.

    Heads up, I have almost completed my homage to Scott M so my production is going to drop low, sub 5k. Summer is here in Texas and I am trying to keep the electric bill down and my room cooler so I decided once I passed Scott's point count I was going to take a step down for a while.
  • Cliff_ForsterCliff_Forster Baltimore, MD Icrontian
    edited Jun 2010
    I did preface my comment by saying that I was not trollin, in other words I meant it real real..... I do that sometimes...... anyhow.....

    God forbid anyone try to have a productive conversation on the topic. I mean, why in the world would anyone ever stop folding, OMG!! Well, I come in and say, well maybe because its inefficient, and it costs too much for too little realized return to date, and people go, OHHHH your so mean, you don't want to cure disease.....

    Come on guys, give me a tiny bit of credit here. Now, Buddy J's response is thoughtful, it addresses the issue. I'm not 100% sure about the numbers, but its an attempt to validate that folding is more valuable then I gave it credit for. I dig that. I'm just saying, don't get testy here just because I'm the only one in the whole damn room that would do something silly like question popular convention. Trust me, my instinct says, folding at home, I gotta do that, its my duty as a nerd, then I thought about it, enough to do actual research, and I made a value judgement against it. My cold black heart :grumble:
  • _k__k_ P-Town, Texas Icrontian
    edited Jun 2010
    You mad bro?

    You complain about the number scientific papers that have been published from folding but I look at it and see a lot of work. Providing proof to the scientific community is hard when it is experimental because it involves a huge number iterations. Also you need to realize that some groups/people will have data from several years worth of study and then take several more years to write, refine, and validate their papers.

    There is also a lot of time spent rechecking work from previous series WUs when changes are made to folding clients or new cores are introduced. When GPU2 launched they spent about 6 months just doing work that SMP clients had already calculated to make sure the data returned was valid and would stand up to scrutiny.

    To really be fair you can't say a super computer or workstation cluster would be better suited until someone builds one and does the same work load to compare. Remember there was a AMD/ATI rig built with quad GX2s to do the work of an op farm, just think of the eco saving effect there.
  • Cliff_ForsterCliff_Forster Baltimore, MD Icrontian
    edited Jun 2010
    _k_

    Thats all I'm asking, a real discussion.

    I'm not some meanie that wants people to die. I'd almost like to see someone talk me into joining the team.
  • clifford_cooleyclifford_cooley Arkansas, USA
    edited Jun 2010
    Forgive me Cliff for taking you the wrong way.

    And for the record, I didn't quit folding. I wanted to continue folding for you guys. However I wanted to attract others if that is even possible. I wasn't sure if I could pull members to this team from the forum that I help moderate. I figured the best solution would be to create a new team for the forum (team w7forums.com).

    Here are my statistics for both teams - http://fah-web.stanford.edu/cgi-bin/main.py?qtype=userpage&username=clifford%5Fcooley. I am working real hard to reach the top 6000 ranking teams so that the team can monitor their statistic through ExtremeOverclocking.

    I want to thank you guys for helping me out and getting me started with this cause.
  • TushonTushon I'm scared, Coach Dallas Icrontian
    edited Jun 2010
    Cliff_Forster said:
    _k_

    Thats all I'm asking, a real discussion.

    I'm not some meanie that wants people to die. I'd almost like to see someone talk me into joining the team.
    If the research papers that have been published as a result of folding time, this may be of help.
  • _k__k_ P-Town, Texas Icrontian
    edited Jun 2010
    Cliff_Forster said:
    but I'm wondering if folding at home is an inefficient way to do it.
    Heck even the US army is using PS3 in cluster instead of a server farm. It will process faster and consume less energy, though some might argue the precision from GPUs is less than that of certain CPUs.

    This is slightly off topic, the seti project is something like f@h but it illustrates how money that was not available to a group was compensated for by creating the distributive software to process massive amounts of data that is continually being back logged because lack of computer time.

    Even if you feel there is misuse or waste in the process it is not only yielding data on proteins but a framework around GPU computing and pretty darn decent distributive computing practices.
  • the_technocratthe_technocrat IC-MotY1 Indyerner Icrontian
    edited Aug 2010
    Tim said:
    What has happened with some of the previously large scale folders we used to have here, like Technocrat and Jonshandbrake? They used to do a lot of points, and now it's virtually nothing.

    Layoffs/downsizings and no more access to the computers that used to fold? Sold the folding farm and went on to other things?
    Yep. I used to administrate an entire school district and had it on the image. (with their permission) Over time, the machines get replaced, and I don't work there any more to put it back on.
Sign In or Register to comment.