Message boards :
Number crunching :
Average Credit Decreasing?
Message board moderation
Previous · 1 . . . 12 · 13 · 14 · 15 · 16 · 17 · 18 . . . 32 · Next
Author | Message |
---|---|
William Send message Joined: 14 Feb 13 Posts: 2037 Credit: 17,689,662 RAC: 0 |
Got a spreadsheet somewhere on Googledocs for analysing Albert data [Edit: may have been seti-beta CPU, will see], Will dig it out on the weekend. Data from Eve, now sadly demised. Beta iirc, only place that keeps results almost indefinitely and is best if you need to compare stuff. I'll rummage. edit: on second thought, you may have a spreadsheet, I've got the original data... somewhere. dang. edit 2: I did find the spreadsheet though. I can't find the original graph on the cloud. Must be somewhere though... A person who won't read has no advantage over one who can't read. (Mark Twain) |
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
Oh I definitely have the spreadsheet, just going to have to wait until the weekend. Yeah, Eve's Seti-Beta credits for shorties-only, versus Correct credit driven with controllers, versus /3 to compare spikiness/instabilities. Will have to find it, but IIRC the cool observation was some really obvious self similar (non-identical repeating) patterns, which are evidence of undamped chaotic behaviour (Seems so much easier to explain after time) "Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions. |
Ulrich Metzner Send message Joined: 3 Jul 02 Posts: 1256 Credit: 13,565,513 RAC: 13 |
Ok, i'm done with this futile discussions... I'm off to milkyway, see you again, if this mess is sorted out... Aloha, Uli |
ML1 Send message Joined: 25 Nov 01 Posts: 21099 Credit: 7,508,002 RAC: 20 |
Got a spreadsheet somewhere... The source problem is that we are not measuring reality, or even a consistent consistently measurable abstraction of anything tangibly real. The "Cobblestone" was a useful but imperfect measure for compute-intensive projects. However, since then, we have other considerations and also even vastly differing measures of compute rates between old x86-FPU, SIMD, GPU, and others... ... Try following some of the ideas of NIST calibration to get real?... (Note, this is a very old discussion...) Myself, I favor NIST-style calibration to award for transistor-transitions. That should work well for for both compute intensive and network intensive tasks until we move into quantum computing... Happy crunchin', Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) |
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
Let's be clear, for estimates that drive the whole thing, Don't need perfect precision here. +/-10% prediction from measured runtime would be excellent compared to the wackiness now in play. There's a difference between having a pet hamster and petting it, and petting it so much you strangle it to death. "Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions. |
Brent Norman Send message Joined: 1 Dec 99 Posts: 2786 Credit: 685,657,289 RAC: 835 |
I'm still taking bets as to the date we get our RAC down to half of what it should be. |
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
I'm still taking bets as to the date we get our RAC down to half of what it should be. It's been around 1/3.3 +/- since creditNew introduction, in two main downward jumps. (Initial, then stock CPU AVX optimisation) "Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions. |
uglybiker Send message Joined: 6 Dec 02 Posts: 32 Credit: 11,417,951 RAC: 42 |
I just look at my credit and start singin' like the Boss. I'm goin' down, down, down, down I'm goin' down, down, down, down..... |
Brent Norman Send message Joined: 1 Dec 99 Posts: 2786 Credit: 685,657,289 RAC: 835 |
Follow that song with Tom Petty's song Free Fallin' |
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
Freefall - Skydive "Ã Feel Wondeful..." "Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions. |
OTS Send message Joined: 6 Jan 08 Posts: 371 Credit: 20,533,537 RAC: 0 |
And follow that with Roy Orbison's "It's Over" - or would "Crying" be more appropriate :). |
Al Send message Joined: 3 Apr 99 Posts: 1682 Credit: 477,343,364 RAC: 482 |
And, sadly, you can probably add to the playlist "You Ain't Seen Nuthin' Yet" by BTO... |
betreger Send message Joined: 29 Jun 99 Posts: 11412 Credit: 29,581,041 RAC: 66 |
Why care, it is commonly known CN is gibberish. I shall crunch on. |
Jeanette Send message Joined: 25 Apr 15 Posts: 55 Credit: 7,827,469 RAC: 0 |
If it's impossible to calculate the correct amount of Cobblestones for any wu, why not just change the credit to be a WU-credit, giving 100 credit for a WU and if some WU are more calculation-intensive then give them some more - like 125. (don't focus on the numbers here, only the idea - which seems more like Einstein@home's approach) This would follow the KISS priciple and I have a feeling that everyone would be happy as the credit would rise to something similar to before the last CreditDropdown started. |
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
If it's impossible to calculate the correct amount of Cobblestones for any wu, why not just change the credit to be a WU-credit, giving 100 credit for a WU and if some WU are more calculation-intensive then give them some more - like 125. (don't focus on the numbers here, only the idea - which seems more like Einstein@home's approach) No-one ever said (to my knowledge) that it's impossible to calculate the 'correct credit', only that the current mechanism is broken. "Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions. |
Jeanette Send message Joined: 25 Apr 15 Posts: 55 Credit: 7,827,469 RAC: 0 |
No-one ever said (to my knowledge) that it's impossible to calculate the 'correct credit', only that the current mechanism is broken. OK - I rest my case.... |
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
No-one ever said (to my knowledge) that it's impossible to calculate the 'correct credit', only that the current mechanism is broken. I'd stretch it to cobblestone scale award (~3.3x now), and call it good. "Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13841 Credit: 208,696,464 RAC: 304 |
I notice some AP work is being split. That should slow the rate of fall for those that get some. Grant Darwin NT |
William Send message Joined: 14 Feb 13 Posts: 2037 Credit: 17,689,662 RAC: 0 |
how was that about a wise (wo)man accepting the things she/he cannot change? *ducks* A person who won't read has no advantage over one who can't read. (Mark Twain) |
Wiggo Send message Joined: 24 Jan 00 Posts: 36577 Credit: 261,360,520 RAC: 489 |
Well my 3570K rig (that was out of action for several weeks at the end of V7 MB and came back online just after the V8 MB intro) has gone from a nice gentle rise to a 28,800 RAC under stock (expected to reach around 29,500 at the time), to an extreme drop (currently down to 22,000) which doesn't look like stopping any time soon. My 2500K has just topped out at just over a pitiful 14,000 RAC (which has only been using it's GPU's here since the 19th of April) and I'm expecting that to drop now as well. What was wrong with the system we were using back in the MB V5-6 days? At least a person could make direct comparisons back then, but then we got slammed (thanks to you know who) coming up to V7 and now after the weeks of CPU's only and the combo CPU/GPU V8 stock testing I've been doing has been made totally useless and for nothing now as the try line is being moved again. It's just good that I like this project so much or I'd be off to other projects that pay just as much as SETI@home use to or even better. Cheers. |
©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.