I just don't get why people use RC5 and SETI@Home

[Pyro]

No, I know how they work, but I really have to question why anybody in their right mind would run these as screensavers when they are not at their computer. Now, I know RC5 can be used while you are working at your computer, but why would you run these apps when you are away from your computer?

RC5 IMHO is pretty damn useless - from D.net site it turns out you are just trying to crack 64bit encryption with brute force. Everyone knows this can be done and overall RC5 IS pretty damn useless.
SETI on the otherhand is a bit more useful, but not that much. The chances of finding a alien radio signal are very insignificant and not very important.

So what should you run? I say run Stanford U's Folding@Home. It simulates the folding of proteins. This IMHO is MUCH, MUCHmore useful than crackin encryption or searching for aliens. Learning how to fold proteins will lead to numerous discoveries in medicine and can even to applied to self assemblying structures.

So take a look at Folding@Home and consider switching to it. Right now we got 14 new 733s at work and soon they'll all be Folding Proteins

 

[Viztech]

If Folding turns you on, go for it.

Just remember that SETI, Prime 95, RC5 and DCypher paved the way for these other distributed projects. Many of us just want to see them through, so don't knock the folks who are trying to finish what they started.

viz

 

[Russ]

There is one very simple reason why I'm not doing any folding. There is no Team AnandTech folding.

Russ, NCNE

 

[Mong0]

Actually Russ there is but it is comprised of just several people me being one. Arstechnica has quite a few people over there so they have a big lead right now.

 

[MGallik]

Isn't Folding@Home the project that will have all the home users results
rendered mute when that specialy built super computer comes on line in 2001?

I wouldn't waste all my time testing the code for a custom built machine
that will finish the entire project in a few months.

 

[Assimilator1]

Pyro

>>>>SETI on the otherhand is a bit more useful, but not that much. The chances of finding a alien radio signal are very insignificant and not very important.<<<<

OK the chances are small of finding ET soon ,but if you never look then you'll never find it ,also as time goes on 'The search' will get more sophisticated &amp; improve the odds.

Are you seriously telling us that when ET is found that you &amp; no one else will bat an eyelid?!:Q ,I don't think so ,it could open or expand whole new areas in science &amp; philosphy.

Ref folding @home ,nice idea &amp; project ,but poor stats &amp; poor client.eg no WU caching ability (last I heard) ,this really sucks for dial up connections.
BTW if you do a search for folding@home in this forum you'll find a few threads about it &amp; the names of people who have already tried it/doing it.

 

[lane42]

ME THINKS WE TRYED TO BE RECUITED:Q
Dam, Need Spell Check

 

[Pyro]

nah, I didnt try to 'recruit' you, but try and get you to think about switching.

Yes, RC5 and Seti were very important pioneers you cant deny that, and no, the client is not very advanced, but its a new project so you need to give it time.

Assimilator, the search fot ET SHOULD go on, all I am saying is that some poeple should consider switching because Folding@home will give you much more immidiate and useful results.

and finally, MGallik, I dont think anyone will bother to waste their time developing a distributed system only to scrap it in few months. And why would they say no to all that computer power anyway? the sooner we learn how proteins fold the better.


and really finally, the more the think about it, the less useful RC5 is. I mean, you know that they key WILL be found and you can calculate how much time it will take you so its really just a waste of cycles...

 

[Ken g6]

If you want a project that's really interesting, try GOLEM@Home. It's virtually a sure way to discover &quot;alien&quot; lifeforms.

On the other hand, if you want a useful project, I don't know of anything more useful than fighting AIDS. Their system occasionally uses your computer for other projects, but about 85% of the time or more, you are fighting AIDS.

 

[Assimilator1]

Ken g6

Nice 1 ,I hadnt heard of either of those 2 until now.

 

[Poof]

Pyro - I am running SETI and Folding and RC5. I also do so for a couple of teams (mainly for the sites I post to). I think it would be better for you to build up some team-building threads for TA Folding rather than knocking the other distributed projects. The Team here probably would like to see some threads going on it...

And right now, if you monitor the Folding discussion list, you'll see an unfortunate lacksidaisacal attitude from the project leaders themselves regarding what is really required to handle 10s of thousands of client accounts trying to post 100s of thousands of WUs to a server... It is unfortunate that Vijay aggressively promoted the availability of the project during a time when the clients are still really beta quality. I am still waiting for the updated x86 linux client (over a month now and I recently found out that the current one seg faults with the 2.4 kernel), or even a gnulibc 1 static one, let alone ANY alpha linux client. The stats unfortunately got fux0red when they were trying out some new stats software (and then they discovered that they had a bad backup tape), so now I'm now missing at least 300 WUs from their current totals and no new ones are showing up anymore because the old stats stuff is apparently turned off. The website has yet to be updated, despite the many pleas from list posters. You have folks from Ars writing WU progress apps for them right now (I'm beta testing 2 of them). There's no caching for folks on intermittant dialups, etc.

None of the above is unusal for such a project when first starting out, but my argument with them is that the distributed model is NOT new and they had plenty of &quot;been there done thats&quot; available to them to use as a template... if anything to at least size their hardware for the expected deluge of volunteers.

I will stay with the project and see it through its growing pains but I think this is at least an opportunity to grab those among the TA readership who aren't interested in seti or dnet's projects for the reasons you cite, to at least contribute to the TA cause... (even though I know the Borg will try to ass laminate them anyway... ) - BUT to do so in the name of &quot;TA's Distributed Computing&quot;, not an us vs them kind of thing.

 

[amok]

The reason its wasted cpu time is that Blue Gene will be able to do as much work as the entire S@H and RC5 projects combined could do in a year, in a matter of a few days. The people that are involved in the F@H project should instead work on something like the AIDS project. It has tangible real-world results as well. And noone is building an unbelievably massive supercomputer to do work on that project. Not to mention the fact that the F@H project isn't very precise. Blue Gene (whose power is measured in PFLOPS) will take approximately 9 months to do a full detail simulation for 1 protein.

Just my 0.02

 

[Poof]

amok - interesting that you bring up Blue Gene... Vijay Pande who is project leader of Folding@home posted the following today about the stats problem and Blue Gene:



<< UPDATE:

STATS: We're still looking into it. The problem is that our most recent backup tape flaked out on us, so we're scrambling to see what we can do to fix things. I'm done w/teaching for the quarter (and most of my students are finished w/classes) so we can hopefully get this straightend out next week. Also, there is a bunch of press about to come out (Economist, Science Magazine, etc) and we should be getting lots more users. I want to fix all of this mess *before* they come on line.

BLUE GENE: I've visited IBM to talk about how we do our simulations and compare to theirs. They have a pretty tough/ambitious goal (i.e. to use a million processors for a single fine grained calcualtion). Our parallelization is much more coarse grained, and so I was pushing our methods as a good alternative. We'll see what they choose. In the end, I think our approaches are pretty complementary.

Vijay >>

 

[dennilfloss]

I do rc5 for fun. Just like when I used to play hockey. It had no purpose but I just loved to score.

 

[Poof]

<< I just loved to score. >>



And assimilate...

 

[ViRGE]

Pyro, let me just say that you've managed to sink any interest I've had in F@H.[img]i/expressions/face-icon-small-mad.gif[/img] Now, as for your attacks on RC5, we're plenty aware of the fact that it won't seriously affect how RC5 is considered as an encryption scheme, or if cracking a 64bit code is considered impossible. However, unlike you, we don't do DC projects just for their usefulness. Dnet has set up a rather good system for RC5, where we get good clients, plenty of stats, and a lot of interaction with Dnet themselves. While I can't say the same for F&amp;H, since I've never done it; I'm willing to wager that you don't have all of that. On top of that, we'd like to finish what we start. RC5 may not accomplish a whole lot in the future, but it's something we've cared about for almost 2 years now, and it's something we'll continue to care about. So, I think I'll be sticking with RC5.:|

 

[networkman]

Yep, what ViRGE just said. Add to that the lack of meaningful stats as has already been posted, and most folks who would've considered the FOLDING project will look elsewhere; believe me, there are some real STATS FREAKS on all of the Distributed Computing teams, not just on TA!

BTW, I'm primarily looking for ET(SETI) and getting back into the RC5 project too.

 

[Phiton]

Hat's off to the members of the TeAm who have responded to Pyro's comments in such a gentelmanly fashion. Since I wish to be thought of as a gentleman also, I will not put into print what is on my mind.

 

[IsOs]

I tried folding but the client wouldn't connect in a firewall &amp; proxy environment. Nice screensaver, can even run together with RC5 but the inability to connect via server/nat/firewall.

 

[Dantoo]

Hey look......Every so often somebody comes in with a &quot;brilliant&quot; suggestion to change to this or that &quot;useful&quot; project.

The simple fact is this. Distributed computing can readily be analogised to football. Folks are particularly interested in their own different &quot;codes&quot; be they Gaelic football, American football, Association football, Rugby (union and league), Australian football and the other dozen or so codes beloved of their fans. This is where their interest and their focus lies. They support their team, wave their colours and when they get home they don't particularly care that the world gained no obvious benefit from their Saturday afternoon's cheering down at the park.

Now you could stand outside the stadium and tell all those people that they should be spending their time and money saving the world from hunger, poverty and ignorance. You could be right. Try it sometime.

Right now I'm going to sit back, watch the RC5 game, keep up with the score, support my team and try and figure out how we can score enough to win the pennant now we have fought hard and won through the playoffs!

 

[RaySun2Be]

Pyro,
You can't have it both ways:
<nah, I didnt try to 'recruit' you, but try and get you to think about switching>

If you come in here trying to get us to think about switching, and we switch, then you have successfully recruited. Your post was intentional in it's attempt to disparage RC5 and SETI, and get folks to think about folding as a better project and switch to your pet project.

I don't care what you call it, but that is recruiting!

We've had these discussions before about which project is better, and it's a personal decision. Like some of the others, I started with RC5 and want to see it through. Plus, I have a lot of PCs cracking at work, and the work PCs require a very stable client, some way to buffer (perproxy) work, etc.

Plus I like the available stats and helping out the team. When RC5/OGR are done, I will look at other possibilities.

 

[divide by zero]

Ditto, Dantoo!!!

 

[SSP]

Mlah.. I'll pass.

I run RC5 for fun. The competetion between us and the DPC's are too good to pass up. If you want to recruit anyone to folding@home, I suggest you don't knock other DC projects. It just piss us off, and we'll never try it.

Besides, I have Seti and OGR on my list. After rc5 that is.

 

[Tetsuo316]

i understand that f@h is a good cause with some immediate results, but who say immediate results were necessary? i think you have to be somewhat of a romantic at heart to run s@h. sure, there are the same great stats and team spirit with rc5 and prime95, but man just think...alien life...


he he he...i'm a nerd.

 

[lane42]

WOW, What a bunch of class guys here.
I wonder if the F@H guys would be as nice if one of us went there and did the same as pyro did here, INSULT US.
Oh well, i guess you cant please everyone.