Message boards :
Number crunching :
Faulty WUs
Message board moderation
Previous · 1 · 2 · 3 · 4 · Next
Author | Message |
---|---|
Send message Joined: 27 Jul 11 Posts: 342 Credit: 252,653,488 RAC: 0 |
I noticed a credit drop of around 500 per WU. No big deal per se, but just wondering if anyone has noticed a slight drop. Like 5% or so. Yes! Crunch times now give 6.5K which recently gave 7K, a drop of 500 per WU (as you point out) |
Send message Joined: 27 May 11 Posts: 10 Credit: 150,418,811 RAC: 0 |
Teemu, please give us our run time credit back for these old Dnet resends. You have the power with a few keystrokes to make it happen. 500 per WU is a lot to us. You can always change it back once the regular WUs are back (or not). We will love you for it. |
Send message Joined: 17 May 11 Posts: 3 Credit: 300,061,742 RAC: 0 |
I also have noticed the drop in credit for WU's, now the % of errors is growing. This project has been rock solid for months, so what has changed? and when will we return to normal? |
Send message Joined: 2 May 11 Posts: 57 Credit: 250,035,598 RAC: 0 |
I also have noticed the drop in credit for WU's, now the % of errors is growing. This project has been rock solid for months, so what has changed? and when will we return to normal? The Problem was fixed, I've only had 1 error yesterday & none today. Did you abort the Wu's you had and download new ones ??? |
Send message Joined: 17 May 11 Posts: 3 Credit: 300,061,742 RAC: 0 |
Will abort the WU's now and let you know Thx STE\/E |
Send message Joined: 2 May 11 Posts: 21 Credit: 173,527,396 RAC: 0 |
Hello, The problem is not fixed no where close actually, for around 4 days now I have had crashing WU and weird timings for the same length of WU If i look at my MSI Afterburner trace the twin cores on my 5970 are way out of sync which instead of working together at 100% there working at an average of 75% and thats a very shaky average as there all over the place every few seconds to back this up GPU-Z currant supply traces show massive ups and downs ever few seconds from 8 amps to 48 amps instead of the constant 48 amps with one card this of course affects my RAC but more important does not look very pretty and even more important all this fluctuating load is putting unnessisary strain on my power supply as it churns out square waves of currect all over the place every few seconds i have rebooted everything and reset the project and downloaded new work but with no change hopefully a fix for everybody is on its way best regards Ian ----> Please Join team Scotland HERE |
Send message Joined: 2 May 11 Posts: 57 Credit: 250,035,598 RAC: 0 |
That's strange, other than the Wu's still taking longer to run I've only had 3 Error's the last 3 days, will have yo keep an eye on the Box's I guess ... |
Send message Joined: 22 Jun 11 Posts: 2080 Credit: 1,844,066,440 RAC: 94,284 |
Hello, It's a conspiracy against you Ian because you have so much darned credits and RAC!!! It is giving us littler guys a chance to move up in the rankings a little bit while holding you back just a little!! The conspiracy against Steve obviously failed so will need to be rethought! I dumped all my Moo units and switched over to Collatz while Teemu fixed things and now that I am coming back everything seems to be doing fine. With the exception of the 500 points loss per unit as opposed to the pre-problems time. |
Send message Joined: 27 May 11 Posts: 10 Credit: 150,418,811 RAC: 0 |
Teemu, Teemu, are you trying to avoid this credit issue? Or are you on vacation (holiday). Reread this thread and you'll see everybody's talking about the lower credit to run time problem with these different WUs. And please don't just say "well these will hopefully pass soon". We are losing credit NOW. These worthless outside of Boinc cobblestone credits actually do mean a lot to some of us in the 'Boinc world'. It's so easy to change with a few keystrokes. We will all be very grateful despite no one else having the balls to ask you directly like this so far (most therapist say to attack a problem head on and not be passive aggressive and dismissive about it). Maybe even some compensation for all the bad WUs wear & tear on our computer hardware, not to mention the wasted electricity (money). Thank you for listening |
Send message Joined: 2 May 11 Posts: 21 Credit: 173,527,396 RAC: 0 |
Hi Mikey Your a very funny man, I have less total credits than you and half your RAC and you have 11 times the computers I do so I hardly think I'm a giant killer lol Still a very funny synopsis of the reason for my problem best regards Ian ----> Please Join team Scotland HERE |
Send message Joined: 11 May 11 Posts: 44 Credit: 291,412,341 RAC: 0 |
Hi Teemu well these will hopefully pass soon It's now almost 10 days & no changes Like I said before this affects different cards in different ways My slower HD4870 runs almost the same times as before with the shorter WUs On the other side the HD5850 still runs 30% longer. I can do better @ Collatz You have to do something to fix this because we are losing a lot of credits Surely this is not a nice Xmas present |
Send message Joined: 2 Oct 11 Posts: 238 Credit: 386,060,374 RAC: 11,423 |
I thought it was too good to last, I am now seeing :-
2. My HD5850 card now does 2200 secs for 6500 credit, whereas before it got 7000, a 7% drop.
I iz also got icons! |
Send message Joined: 22 Jun 11 Posts: 2080 Credit: 1,844,066,440 RAC: 94,284 |
Hi Mikey I know it wasn't helpful but you do have to admit it was a idea good though, don't you? As for the actual units being completed most of yours are now doing fine, we both have a few "can't parse file" errors, although they are different codes. But other than that we both seem to be crunching just fine, except for one of my machines that has not returned a single unit correctly yet!! It is a pc thing and will probably be shut down shortly, it is working on Collatz right now but will NOT work here at Moo. ps where in Scotland are you from, I was there for a short time while I was in the US Navy. LOVED the place and wish I could go back except for you guys driving on the wrong side of the road. I went to St. Croix and was honking and honking at some guy on the 'wrong side of the road'! I finally figured out IT WAS ME on the wrong side, I felt stupid and he just laughed! |
Send message Joined: 2 May 11 Posts: 21 Credit: 173,527,396 RAC: 0 |
Hello, yes I shifted the client core from three to two which seems to have it working better although not perfect instead of 14 mins each the WU of the same size are now taking anywhere from 20 to 25 mins each I had the client core at 3 for months using it with one 5970 and with twin 5970's quite happily so no idea what the differance is now plus the STDR output page which used to be about a page long and end with the time of 14 mins ( or 7 mins on the twin 5970's ) now that page you have to scroll down as its around 3 or 4 pages long and makes no sense and ends at 20 to 25 mins for the same work unit so whatever has been changed its not something small On your p.s. I live in a small town called Tillicoultry which is 10 miles north of Stirling I's also known as the gateway to the highlands as it sits at the foot of the Ochil hills, one of which my house is at the base of lol I nearly got ran over on my first trip to America looking the wrong way when crossing the road after coming off the plane lol so I know how embarresed you feel working out you were in the wrong lol best regards Ian ----> Please Join team Scotland HERE |
Send message Joined: 5 May 11 Posts: 233 Credit: 351,414,150 RAC: 0 |
Geeez ..... I go away for a week or so rebuilding both boxes and come back to find you broke it .... tusk tusk :) Looks like "fun" ..... I'm going full tilt on a PrimeGrid challenge at present for the Team, then I'll be back throwing my hat back in the Moo Ring ... looks like I chose the right time to strip down and rebuild. Back in the ring Thursday .... meanwhile try not to break anything else rofl :) Regards Zy |
Send message Joined: 22 Jun 11 Posts: 2080 Credit: 1,844,066,440 RAC: 94,284 |
Hello, I don't know anymore about how to fix your Boinc, I am out of ideas right now, sorry! We are using the same version of the AMD software too! Oh...have you thought about taking out one gpu and seeing if the other runs okay, then reversing the process just to make sure it isn't a gpu gone bad? I was in Dunoon for a year and traveled to Edinburgh to see a play and get a tattoo, the semi usual Sailor stuff! I was able to rent a room in Town so I didn't have to stay on the ship the whole time, it was VERY nice and fun. It was kind of like going to work all day then coming home at night, with the pub being on the way home! 'Warm' beer is okay after you get used to it, almost all of the US beer is cold, some of it near freezing! I liked the pints of 'heavy', man we could drink those like water sometimes!!! It has been a VERY long time since I was there, more than 25 years now!! I do not miss the almost daily rain showers though!! Nor do I miss the 24 hour days when we had to work painting submarines!!! Painting a submarine with ROLLERS is not my idea of fun, they are frickin' HUGE!!!!!!!!!!!! Each sub has 7 layers of paint from the bare metal to the top coat, each applied AFTER the last one is dry! AND in some places we had to blast the old paint of before we could put the new paint on, that was no fun either!!! Each sub was in dry dock for 10 days and then it went back to sea, so it was non stop work from the minute it came in until the minute it left, then it was time to get setup for the next one, which took about 7 days! Not one single sub has the exact same hull design so all the blocks, to keep them upright, had to be custom made to exacting specs so it didn't roll over while in dry dock! |
Send message Joined: 23 Jun 11 Posts: 87 Credit: 798,452,366 RAC: 0 |
I also have been having problems. for like 3 days now 2 of my pc's that are boincing moowrapper are having wu's build up that are pending. one has like 9 pending, the other has like 40 pending. The numbers keep rising as i check day to day. Any ideas on what is causing this? |
Send message Joined: 2 Oct 11 Posts: 238 Credit: 386,060,374 RAC: 11,423 |
Mikey, a pint of heavy hasn't changed in 25 years!! Edinburgh is brilliant, I was there for a week about 3 years ago, loved it! I iz also got icons! |
Send message Joined: 5 May 11 Posts: 233 Credit: 351,414,150 RAC: 0 |
Grateful 5970 owners post GPU-Z readouts for me to compare. I just rebuilt my two boxes and seem to be going fine, currently running in a PrimeGrid challenge for the team up to Thursday. That app is a low utilisation due to it being OpenCL, not native Stream. It therefore runs cool, allbeit relatively slowly. A good start test for the new boxes. Sooo ran the Prime Grid app 790/300, ran great combination of the app itself and the high pressure fans I put in worked well .... GPU was at 58 degrees, VRMs hottest was 68 degrees. Tonight I wacked up the juice to give the boxes a first serious test o/c to 850/300 with volts at 1.1 ..... real good, solid at GPU 62 degrees, VRMs around 88-92 degrees, current draw VRM was 27 amps ... well happy. Then ..... Seeing the recent hassles on Moo, I setup GPU-Z on sceen to monitor VRMs so I could switchoff at first hassle .... and changed to a Moo WU, they broadly are also light on memory, strong on GPU. I left the settings at 850/300 1.1v, then switched. ... wow .. Current amps draw for VRM shot up to 64.5amps, and VRMs hit 134 degrees in 3 seconds flat, and still climbing, as I lept for the suspend button !! GPU however was still OK at around 68 degrees, wierd. Not a clue whats happening at present, but the immediate current draw to 64ampos broke out sweat on the brow :) The VRM temps going to 138+ in 3 seconds flat was terrifying, never seen that before. Hence the request for comparitor readings for GPU/VRN/Volts and current draw. I am certain the card is fine, reacting well to other Project apps, but - for the first time - not at moo. I am starting to wonder if the base app has changed in its method of running and is now hitting GPU memory hard. Too early to jump to conclusions, but as the card is definitely OK, something strange is happening.... So would be grateful for the comparitor numbers. EDIT: All that was on 11.12 the latest drivers. They seem to be changing a lot in the driver makeup to prevent users overheating cards .... wild shot at present but could be the driver, so I am changing to 11.9, see what happens ... that was ok for me just before I shut down to rebuild the boxes. Regards Zy |
Send message Joined: 5 May 11 Posts: 233 Credit: 351,414,150 RAC: 0 |
It was the drivers .... had to roll back to 11.9 Both 11.11 & 11.12 went crazy wacking up the VRMs to horrible numbers (138 degrees !!) Now back to normal. Moral of the tale, stay away from 11.11 or 11.12 - certainly at the leat switch them with great caution. Regards Zy |