Message boards :
Number crunching :
The future of Boinc
Message board moderation
Author | Message |
---|---|
sir_LOIN Send message Joined: 20 May 99 Posts: 5 Credit: 1,545,040 RAC: 0 |
I posted this on the Overclockers.com forum but thought I should in here too... I think Boinc is simply kinda slower, in that it doesn't show instantanious results like the old seti. It is however alot more accurate and it's going to make life harder for cheaters. Over a certain amount of time the results will average out and we will be able to monitor the stats and have a certain consistency. Maybe not as fast as the old one to spit out credits, but more accurate in the end because it will really show the processing that has been done. I like it better this way because since not all workunits are equal, then this is a more acurate representation of the actual crunching that was done. It also allows consistency across different platforms and projects since they base the credits by the same standards. We might have a stat in the future that we can't have now, total processing power. Say for example that folding eventually changes on boinc and you are processing for seti at the same time, we should be able to establish a grand crunching total for both and wichever project you're crunching for. Wich is impossible now because of the different credit systems. We could have teams and leaders across projects so we could compare ourselves with millions of people more for total crunching. Overclockers.com seti team vs Overclocker.com folding team.... All the Overclockers.com teams combined against the world! This is way more interesting! Boinc has alot more possibilities, it's just starting so of course it's buggy. Let's just give it some time. Happy crunching! |
John McLeod VII Send message Joined: 15 Jul 99 Posts: 24806 Credit: 790,712 RAC: 0 |
|
D.J. Schweitz Send message Joined: 29 Oct 02 Posts: 157 Credit: 871,078 RAC: 0 |
|
mattman6 Send message Joined: 4 Jan 03 Posts: 7 Credit: 162,193 RAC: 0 |
Thank you sir Loin. Now, if only more people could be patient and level-headed. |
Byron Leigh Hatch @ team Carl Sagan Send message Joined: 5 Jul 99 Posts: 4548 Credit: 35,667,570 RAC: 4 |
sir LOIN , I would also like to add my thanks , Well said. |
Guido_A_Waldenmeier_ Send message Joined: 3 Apr 99 Posts: 482 Credit: 4,774 RAC: 0 |
[/url] [/url] Bei der Eroberung des Weltraums sind zwei Probleme zu lösen:Die Schwerkraft und der Papierkrieg. Mit der Schwerkraft wären wir fertig geworden.Wernher von Braun |
Mark Chester Send message Joined: 3 Apr 99 Posts: 6 Credit: 19,504 RAC: 0 |
Well said sL, I enjoy boincing even with the bugs. |
AMDave Send message Joined: 4 May 01 Posts: 2 Credit: 100,208 RAC: 0 |
> Well said sL, I enjoy boincing even with the bugs. > > Bugs ? They're simply protein pills in different wrappers. But seriously, Sir LOIN says poigniant words. Someone knight that man...oh they already did ? Bravo. If you always expect too much, you are always likely to be dissappointed. Be thankful when you don't have the BSoD and when you do...marvel at its truth! - Dave - |
KWSN - Toolman / LadyUnicorn Send message Joined: 8 Aug 99 Posts: 1 Credit: 586,938 RAC: 0 |
Sir Loin, I agree. Everyone needs to calm down and give Seti/Boinc some time. It seems everyone forgets the begining of the Seti@Home client and all we went through back in those days.... Boinc is far superior than the old Seti Client and time will prove this out. You are wise beyond your years, are you sure you are not a Knight Who Says Ni!? Ni! :) KWSN - Toolman Member of The Knights Who Say Ni! http://kwsn.timtaylor.net |
sir_LOIN Send message Joined: 20 May 99 Posts: 5 Credit: 1,545,040 RAC: 0 |
Thanx everybody for all the positive comments. I don't know why some people are bashing Boinc. Yes there's a few bugs here and there but so what, big deal. Just give them time to sort it all out. Boinc has so much more possibilities, this is just the tip of the iceberg. |
Guido_A_Waldenmeier_ Send message Joined: 3 Apr 99 Posts: 482 Credit: 4,774 RAC: 0 |
[/url] [/url] Bei der Eroberung des Weltraums sind zwei Probleme zu lösen:Die Schwerkraft und der Papierkrieg. Mit der Schwerkraft wären wir fertig geworden.Wernher von Braun |
Luis Goes Send message Joined: 10 Jun 04 Posts: 8 Credit: 15,190 RAC: 0 |
> Well said sL, I enjoy boincing even with the bugs. > > It's not the bugs that are annoying. You're totally missing the point here. It's the lack of work that's annoying (or the extremely bad load balancing, that makes some guys receive tons of work, while others receive nothing for weeks). I don't understand why BOINC isn't using the same input channel(WU uploader) as classic SETI. Or is it? |
KWSN - Tim Send message Joined: 22 May 00 Posts: 13 Credit: 1,886,308 RAC: 7 |
Gee, I said something to that effect yestarday on overclockers.com and got put in my place about how it isnt useful as a gauge of system performance any more by somebody afraid of change :) *cough cough skyhook cough* |
SkyHook Send message Joined: 20 Feb 02 Posts: 9 Credit: 194,866 RAC: 0 |
> Gee, I said something to that effect yestarday on overclockers.com and got put > in my place about how it isnt useful as a gauge of system performance any more > by somebody afraid of change :) > *cough cough skyhook cough* > > You apparently are coming down with a cold there aren't you? To say I put you in your place is a bit of a reach, I simply stated then as I will state now, "How can I accurately gauge my system performance on a day to day basis as I did until recently with Classic SETI, when the new credit system scores my work against other unrelated systems"? The example I used in my reply over at Overclockers.com went something like, I do a WU and my system says I should get 40.09 Credits, another individual does the same WU and his system claims 19.55 Credits, the third follows suit and claims 25.48. So all three of us are awarded 25.48 Credits and this tells me what about the way my system did it's work? I am not bad mouthing BOINC, hell I'm into it up to my armpits, I'm not even bad mouthing the Credit system. I simply pointed out that these changes were going to bring about some radical differences in the way our Team had in the past operated as far as Team Races, Gauntlet Race, and the such. Not Good or Bad just preparing my Team members for the fact that things are going to be different. You need more proof, at the moment of posting this I have the #1 Ranked Computer according to the BOINC Stats, however I can assure you it is not my best performing computer. I have 4 other machines that have run just as many or more WUs through them, but because of either luck of the draw or ET's intervention this particular machine got more of it's WUs Granted Credits while the other set around with "Pendings". Is that "Good" or "Bad", who knows, but it sure is a difference that deserves pointing out. Respectfully, SkyHook |
Heffed Send message Joined: 19 Mar 02 Posts: 1856 Credit: 40,736 RAC: 0 |
> I do a WU and my system says I should > get 40.09 Credits, another individual does the same WU and his system claims > 19.55 Credits, the third follows suit and claims 25.48. So all three of us are > awarded 25.48 Credits and this tells me what about the way my system did it's > work? I don't know. Your individual results give you a very good indication of system performance. <a> [/url] |
Toby Send message Joined: 26 Oct 00 Posts: 1005 Credit: 6,366,949 RAC: 0 |
SkyHook: If you were using seti-classic to benchmark your system accurately against other architectures/operating systems, then you were using a very poor indicator. They optimized the client the most for x86/windows since that is their biggest single target audience. Comparing the results of that to the results of a SPARC processor or even an x86 running linux is not accurate. It may be a rough indicator but then again, so is the new system. I assume you know that classic work units varied greatly in their processing time. This, once again, detracts from accuracy unless you are able to compare the results from the same work unit to each other. I do agree with you about the changes the 'pending credit' will bring to the rivalry between users and teams but then again I was never concerned much about my stats so I don't really care :) I just crunch what I can and leave it at that. BOINC is definitely different as you pointed out and I think overall a good change considering the added security and anti-cheating features this system has. keep crunchin'! Toby |
sir_LOIN Send message Joined: 20 May 99 Posts: 5 Credit: 1,545,040 RAC: 0 |
I have a similar thing going Skyhook, my fastest computer is also my lowest ranking! But this is only part of the changes. Since it just started and they had their share of problems, I believe that over a certain amount of time my results will average out and my fastest system will be on top as it should be. I agree that with Boinc you don't have some of the races like we have now. But like I posted earlier, I believe that in the future Boinc will offer new races like we've never seen before. The best is yet to come. |
KWSN - Tim Send message Joined: 22 May 00 Posts: 13 Credit: 1,886,308 RAC: 7 |
> You apparently are coming down with a cold there aren't you? To say I put you > in your place is a bit of a reach, .... Hey, you knew who I was. You must have felt somewhat like you put me in my place. Maybe "put me in my place" was a bit of an exaggeration, sorry about that. FWIW, my faster cruncher has a proportionately larger number of credits attached to itafter accounting for its downtime after running out of credits where the slower one had work to do. Last night when Seti came back online and results were uploaded from around the world the stats worked themselves out for my 2 systems. |
Pascal, K G Send message Joined: 3 Apr 99 Posts: 2343 Credit: 150,491 RAC: 0 |
BOINCs future is so bright it will need to wear shades... |
Mark Chester Send message Joined: 3 Apr 99 Posts: 6 Credit: 19,504 RAC: 0 |
BOINC, oh, I mean BUMP ;) |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.