So why you do not give credits in relation to the time needed?

As I have explained few times in different topics, currently we have very short tasks series (2k WU each) where every batch have different computation time. Also, different versions of BOINC Managers are not consistent in showing estimated time and I have serious problem with FPOPS setting for those WU's.

This will slowly stabilise through next series but is difficult to say when exactly this happen...

Ps. I really like beer ;)

---

Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home team
My Patreon profile
Universe@Home on YT

I know this is an old thread but today I noticed some tasks taking like 3 times as longer than normal. On my 2P 2670 units were usually taking 3-4 hours. I come home today and most of the running tasks are going to take 10-11 hours. I checked a wingman for one of the tasks already verified and they also took 3x as long for the same 333.33 points. Reading back the previous instance they were adjusted. 3 hours or 10 hours makes no difference to me as long the results are relative to each other.

Read: http://universeathome.pl/universe/forum_thread.php?id=130

Read: http://universeathome.pl/universe/forum_thread.php?id=130

Just more reports. No solution.

No, credits also will be changed :)

Well, apparently, credits were never modified and are fixed regardless of task duration...

You're replying a thread almost 6 years old!

Task credit has varied several times since then.

Haha ha hah. Yes we all weeped when they halved the credits last year after the whinging from the ULX task crunchers complained about task credit inequity compared to BHSpin2 tasks for approximate equal runtimes.

---

A proud member of the OFA (Old Farts Association)

Haha ha hah. Yes we all weeped when they halved the credits last year after the whinging from the ULX task crunchers complained about task credit inequity compared to BHSpin2 tasks for approximate equal runtimes.

I wish they would bring back the ULX tasks again, I still need some hours to get my 50k badge on wuprop

Haha ha hah. Yes we all weeped when they halved the credits last year after the whinging from the ULX task crunchers complained about task credit inequity compared to BHSpin2 tasks for approximate equal runtimes.

I wish they would bring back the ULX tasks again, I still need some hours to get my 50k badge on wuprop

It is a chance that the ULX's come back about end of July.

---

Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home team
My Patreon profile
Universe@Home on YT

I don't understand the point of your "answers" after my post, I'm just mentioning the fact that the credit for BHspin is fixed regardless of the task duration (double, triple...). Period.

If there is an explanation / justification / information for this, it is interesting.

If there is none, who cares.

Hi, Krzysztof,

Congratulations on all the work you've been doing!

I have several Ubuntu VMs, which have been successfully processing Universe (approx 23M cobblestones) for a while. I put them onto other work for a while to up the score of some of my other projects. But now, when I switch back, I can't pick up any Universe work units, even though there are >3M available.

My original config was Ubuntu 21.xx with BOINC 7.16.x, and that ran fine for many months. That machine now doesn't pick up your work units.

As a test, I have built a new machine on Ubuntu 22.04 with BOINC 7.18.1 (the version in the repository) and, though it picks up other projects, it also tells me that there are 0 Universe work units available.

Any ideas?

Many thanks,

Nick

---

There is currently a multi-team competition ongoing, including project Universe (among others), maybe the server is under quite a lot of stress at the moment ? I can't even reach the status page of the project, not responding...

I have a small debian host crunching on it for several days and I can see it cannot get any task neither.

edith : after a long time server status finally responded with 3kk tasks also, I think the machines fail to connect to the project due to overload.

OK. Thanks!

I'm usually oblivious to competitions, being in a team of one.

---

I've also stopped receiving new work units in the past couple of days. Should the competition be having this much effect?

Yes. Any time you 10X more requests than normal hitting the task buffer every second and emptying it immediately is going to cause problems getting work consistently.

Once the contest ends and all those participants leave and go back to their normal projects, the Universe project will return to normal.

---

A proud member of the OFA (Old Farts Association)

I've also stopped receiving new work units in the past couple of days. Should the competition be having this much effect?

As of this post, due to the competition, there are 1,096,446 tasks in progress (down from over 1,300,000 a few hours ago).

And we are crunching at a rate of 1.02443173 EXAFLOPS

So yes, the effect of this annual competition is quite significant

Obviously i meant PFLOPS. I derped. Sorry lol

Hi

Is there a problem with the servers?

It appears that at present NO Completed tasks are able to upload and NO new tasks are being sent out.

Its been like this for several days now.

The server status page does NOT load at all.

Regards

Michael Blucher

What planet have you been on the past two weeks? Do you not read the forums?

Click to read ANSWER: Message boards : Number crunching : Server Thread

---

I have been offline for a couple of weeks.

I rarely read the forums.

My processing clusters generally run without any need to intervene.
I noticed the problem yesterday when I found a significant number of completed tasks not uploading.
They all appear to be stalled.
This appears to be a problem going back several days.