yeah "we don't worry we just crunch!!"
and why not get both ati and nvidia cards?
I lost more than anyone else and it's nothing
I can get more,lol
Spring cleaning and a new ati app??
ought to be interesting

Its amazing what a tiny Baileys can do with your thoughts. LoL
Dont worry. Just Crunch. ;-)

what ? no nothing was done with my thoughts. I had alrdy written a big flamepost and then i deleted it, cause i came to my senses. also they are just humans. but it's hard to see that 1 million results got lost like a *puff* and was blown away.
Sure i ll go on and crunch =)

why not get both ati and nvidia cards?

i guess there are more important projects in other science fields, but they don't support ATI ;) so I ll think about what happened again, when I get a nvidia card ...

More important projects?
It doesn't get any better than this.
Fundamental nature of the universe trumps medical discoveries or math problems in my book.
But it's all more knowledge to add to the human understanding and as such is ALL important.

And what about this distributed computing itself?
Seems to me we can eventually have a huge supercomputer at all of our fingertips!!
But by then voice recognition will have our hands free.

I just really hope when it wakes up it decides to like us, LOL!

Using both nvidia and ati just gives you more options.

Just idle musings and remember to keep it fun!

Its amazing what a tiny Baileys can do with your thoughts. LoL
Dont worry. Just Crunch. ;-)

what ? no nothing was done with my thoughts.

I was talking about my thoughts. ;-D
Guessing its the Baileys from last night. lol

Well.. yes.. They are humans. And MilkyWay is a great
project. What ever needs to be done should be done, to
get it running again. The best would of course be if it
could happen with nothing being deleted.. But theres not
much to do with the matter now. If using the weekend to
make this work, its great. Crossing fingers. ;-)

---

But something big IS coming once these issues are sorted out.

The Credit Junkie might just be VERY Happy...

---

..... Seems to me we can eventually have a huge supercomputer at all of our fingertips!! .....

For the credit junkies out there, some idle musings along those lines from Quantum Computing realities.

The Aqua BOINC Project - which is working on developing an existing Quantum Computer over at D-Wave (production versions circa +/- 40 years away) - not long ago had a WU that ran for 18hrs on a top range i7. Such lucky owners gained around 28,000 credits per WU for their efforts - crudely 38,000 credits a day for the i7.

The WU was running the same problem as the quantum computer, as - in crude terms - a kind of double check on the slavering beast.

That works out at the Quantum Computer being 16 Trillion times faster, as the problem being crunched simulated 4 nanoseconds work on the experimental Quantum computer.

The i7 daily rate of 38,000 for that WU, works out at 608 Quadrillion credits a day, or 168 Trillion credits a second for the Quantum....... Dont know whether or not to laugh or cry comparing the credits I toss out there these days rofl

The current all up total for MW is a comparitively puny 80 Billion since the Project started a few years back .....

I know, I should get out more, but whats a BOINC addict to do whilst they put humpty back together again :)

Regards
Zy

I'm sorry to guys. I screwed up big time.
I use a Asus MB with AMD 6 core processor and Geoforce 460. processor.
Sharing the processors between Seti and MilkyWay. Been with Seti since
1993 approx.
Normally I would get 36 max. Milkyway (cuda_opencl)wu's work units for the 460.
Until last week noticing approx 1700+ work units. No problem I thought, until
this morning. I went to suspend the MilkyWay Wu's but to my horror, I must not
hit the box correctly in Boinc Task, and aborted them instead of suspending.
I'm here in complete bewilderment seeing these work units now showing as "aborted". Is there any way to retreive these work units, though I think not. Any restoration methods would be greatly appreciated. I'm sorry for the inconvience, and can only wait to obtain new work units. Sorry all.

David.

I would kill the guy who is responsible for this :P
about 1 million results was simply burned in the oven?!
such a waste =(

That would mostly be my fault

Matt, I don´t blame you! It is science we are making. And science is experimental any time. And experiments can go wrong.

I think that you do not need to apologize to someone.

Take your head up!

---

Don't kick yourself too hard, Matt. There have been far worse headlines in the news lately.

---

At least we have decent communication and effort going on for the fixes. Really, that's about all we can fairly ask.
Keep after it!

---

I would kill the guy who is responsible for this :P
about 1 million results was simply burned in the oven?!
such a waste =(

That would mostly be my fault

plz take a look at the smiley ... that means it was more a joke ... no so seriously

I wouldnt kill u =) I WOULD ******** *** ** * * ** * *** U :D just kidding

I guess the most of the users havent noticed it yet and also will never notice it.

More important projects?
It doesn't get any better than this.
Fundamental nature of the universe trumps medical discoveries or math problems in my book.
But it's all more knowledge to add to the human understanding and as such is ALL important.

yes, but I think we have to solve some "smaller" problems in biology, chemistry or physic first, before we look at more complex problems like the stars and our universe =)
Or think about what zydor has said, the computation for the milkyway would be no problem if we had quantum computers. so why shouldnt we solve this problem first ? :o

I would kill the guy who is responsible for this :P
about 1 million results was simply burned in the oven?!
such a waste =(
That would mostly be my fault

It is a moment of learning Matt. Do you think all the credit hunters here have never made a mistake to learn?

No worries keep up the good work and peek at Travis once in a while...

---

Greetings from,
TJ

I would kill the guy who is responsible for this :P
about 1 million results was simply burned in the oven?!
such a waste =(

That would mostly be my fault

I see this more as a "demonstration experiment". All the posts I've seen with people asking why we can't have large WU caches can now clearly see the reason. :-)

Just wondering.

Why don't you make a limited test run when you make a change in the software or
did you make a test run but missed a checkpoint regarding the limitation of the wu \ core?

yes, but I think we have to solve some "smaller" problems in biology, chemistry or physic first, before we look at more complex problems like the stars and our universe =)
Or think about what zydor has said, the computation for the milkyway would be no problem if we had quantum computers. so why shouldnt we solve this problem first ? :o

Luckily we have all kinds of smart people running around doing their parts on the things that they each think merits attention.
Me being a space head, naturally I want the universe, LOL

I am happy crunching here! For me it’s not a problem losing some wu’s.
But I hope we get here a “new”server here, with some ssd disk, and a good backup.

Just wondering.

Why don't you make a limited test run when you make a change in the software or
did you make a test run but missed a checkpoint regarding the limitation of the wu \ core?

I updated all of the BOINC server components (when really all I wanted to do was to fix some scheduler problems) and they at some point added new workunit / GPU limitations to add to the WU/CPU limits, and this wasn't set by default or something like that, so GPUs got unlimited workunits.

I would kill the guy who is responsible for this :P
about 1 million results was simply burned in the oven?!
such a waste =(

That would mostly be my fault

The red queen has decided that you can keep your head just this once :)

I would kill the guy who is responsible for this :P
about 1 million results was simply burned in the oven?!
such a waste =(

That would mostly be my fault

The red queen has decided that you can keep your head just this once :)

Actually it wasn't entirely his fault either. For awhile we had both max_wus_in_progress for CPUs and GPUs set, and the scheduler was just ignoring the GPU limit.

We had to add a config_aux.xml and specify a total limit in that (which seems to work), but it's still not paying attention to the GPU limit.

---