First one crunched (CPU) and a credit drop of about 4 or 5 per cent. That about what you were aiming for Travis?

Haven't touched the awarded credit yet. Making sure the server is running smoothly first :)

---

Yes just letting you know how things are working as per your request on the home page.

Thanks :) It's appreciated.

---

Well, given that the SETI project has some reliability and loading issues, it should not be a surprise to see it dropping off a bit.

I mean, they have a weekly 4 to 6 hour outage every Tuesday, and, since they are running very close to the edge with their internet pipe, that means that after the maintenance outage they have an 8 to 12 hour traffic jam with uploads and downloads. They also have periodic upload issues (periodic as in multiple times during the week), and general load balancing issues -- since they 'burst mode' release their Astropulse work units, when they are released they have another traffic jam. It can be frustrating for those users who are 'SETI-centric' unless they are among the SETI advocate folks who simply see the reliability performance there as being impressive given their 'sparse' resources.

As a result of all that, there are more than a few SETI users who have walked away (in some cases I fear, their experience with SETI may have resulted in them walking away from BOINC entirely).

For me, performance issues drove me several years ago to multiple BOINC projects and the more recent performance issues over there have resulted in me simply reducing SETI way down my current active project list.

Then again, between DA doing the behind the scenes pressure on other projects, along with some of his influence regarding how the BOINC client development has gone, I suppose these days, I don't find myself particularly put out by what is going on over there.

May be related to the fact that MW is now producing ~49.8 million credits/day(76% of) where Seti is 65.6 mill. Certainly wouldn't want another project to beat Seti at its own game.

---

Haven't touched the awarded credit yet. Making sure the server is running smoothly first :)

Oh, OK. Must be natural variation between different searches. Thanks.

I REALLY think that before you mess with the credit awards, you should get CUDA off the ground -- if only to have something to use as a comparison -- compare MilkyWay *released* not optimized application credit to other *released* not optimized credit schemes (say GPUGrid for example). After that, then consider the MilkyWay *released* not optimized CPU application to general CPU client performance within BOINC (say compared to Spinhenge or POEM or one of the other small but reliable BOINC projects).

As to the optimized ATI GPU application, you can compare that to the optimized ATI application at SETI -- oh wait, it isn't supported at all since the BOINC client has no support for ATI GPU's to begin with. Oops -- guess you can't revise credit schemes at all then <grin>.

Haven't touched the awarded credit yet. Making sure the server is running smoothly first :)

---

I REALLY think that before you mess with the credit awards, you should get CUDA off the ground -- if only to have something to use as a comparison -- compare MilkyWay *released* not optimized application credit to other *released* not optimized credit schemes (say GPUGrid for example). After that, then consider the MilkyWay *released* not optimized CPU application to general CPU client performance within BOINC (say compared to Spinhenge or POEM or one of the other small but reliable BOINC projects).

We actually have a working CUDA double precision app running right now, our new researcher Anthony is working on it. I think he's going to be releasing it as a beta application in the next week or so -- the only issue is that we're having an issue with the likelihood calculation that we want to resolve, when that happens we'll have to update all the applications.

---

With a fixed credit system like at MW, the amount of credit you receive is entirely based on the initial vanilla MW app, and the optimisations therein of future apps.

Ultimately when the vanilla app is optimised further, the credit/WU will drop but the credit/time should stay the same for that app. For other apps left untouched, their credit/time will decrease.

If the vanilla app is left untouched and other apps are further optimised, then credit/time on those apps increases.

It's all a balancing act, deciding where to distribute resources, and so on and so forth. At the end of the day it's all up to the developers, but imo no credit change should happen if the vanilla CPU app stays the same unless the whole system is shafted.

---

You may have it -- but *we* don't. ;)

We actually have a working CUDA double precision app running right now, our new researcher Anthony is working on it. I think he's going to be releasing it as a beta application in the next week or so -- the only issue is that we're having an issue with the likelihood calculation that we want to resolve, when that happens we'll have to update all the applications.

---

Posted on main page:

July 21, 2009
I'll be updating the server daemons tonight with better result verification and updated particle swarm and genetic searches. There will probably be some downtime while I get everything up and running correctly. We've also been examining some issues we're having with our fitness calculation, so we'll be releasing a new application and its code to deal with that as well. It also seems we're having some excessive credit issues again so things are going to have to be tweaked (have fun flaming our forums :P).

I would say how can credits be too high when DA himself told you how to calculate them and that is how they are set. Or is seti lowering the bar again?

Oh well, it's maybe time to move my Ati cards to FAH.

---

I'm having trouble with my nVidia crashing with computation errors on GPUgrid.
Travis, can you please get MW CUDA up and running so I can crash it on MW instead. Thanks ;)

You may have it -- but *we* don't. ;)

We actually have a working CUDA double precision app running right now, our new researcher Anthony is working on it. I think he's going to be releasing it as a beta application in the next week or so -- the only issue is that we're having an issue with the likelihood calculation that we want to resolve, when that happens we'll have to update all the applications.

---

I'm having trouble with my nVidia crashing with computation errors on GPUgrid.

You ain't the only one !!

I'm having trouble with my nVidia crashing with computation errors on GPUgrid.

You ain't the only one !!

That's at least two of us ready to crash-test MW CUDA.

---

I'm having trouble with my nVidia crashing with computation errors on GPUgrid.

You ain't the only one !!

That's at least two of us ready to crash-test MW CUDA.

Make that 3 of us, I need to run a stable project like MW on my 9500 GT.
;-p

---

I'm having trouble with my nVidia crashing with computation errors on GPUgrid.

You ain't the only one !!

That's at least two of us ready to crash-test MW CUDA.

Make that 3 of us, I need to run a stable project like MW on my 9500 GT.
;-p

I started crunching GPUgrid last week with a GTX 260 and ended up with computation errors on every single WU. Then i switched to an older nVidia driver (181.22) and - bingo : everything is running smoothly now. So i think it's a driver issue with all 3 of you.

Of course i will crunch for MW as soon as the CUDA app is out and running.
I'm really looking forward to it ...

Yep , me too- 181.20 with BOINC 6.5.0 = zero problems with a dual GTX 260 rig @ GPUgrid 24/7 for many months... XP Home, Q6600 @ 3.0 ... a second identical rig runs great but shut down for summer to give the A/C a break ;)

Is it just me or is the server/site slow after Travis played with things today??

---

I started crunching GPUgrid last week with a GTX 260 and ended up with computation errors on every single WU. Then i switched to an older nVidia driver (181.22) and - bingo : everything is running smoothly now. So i think it's a driver issue with all 3 of you.

I've already re-installed and expanded nvlddmkm.sys a few times now without success. I put it down to Vista being rubbish and me being rubbish for buying a Vista laptop that I can't downgrade to XP.

Of course i will crunch for MW as soon as the CUDA app is out and running.
I'm really looking forward to it ...

You could be on your own if the rest of us can't get these drivers sorted out ;)

---

Is it just me or is the server/site slow after Travis played with things today??

Same trouble. Errors out some of the time.

---

Doesn't expecting the unexpected make the unexpected the expected?
If it makes sense, DON'T do it.

Gday Bandit..So I'm not going crazy then...:))

---

Is it just me or is the server/site slow after Travis played with things today??

Yes it has been very slow for the last 4 or 5 hrs at least. Travis made some changes. I think at least one of them isnt working right?
;-p

---