I just got 150 new tasks for my primary machine (I9-9900K, NVIDIA Geforce 1660 TI); they're all GPU tasks.

04/15/20 19:18:53 | SETI@home | Scheduler request completed: got 150 new tasks

Lucky people here. I only got a total of 10 or so resends. Granted, CPU work only, but still...

---

Lucky people here. I only got a total of 10 or so resends. Granted, CPU work only, but still...

I've picked up a few more than that. But it is only a few.
Down to getting 1 Task every couple of days or so.

---

Grant
Darwin NT

Geez, I've been getting nothing here. I see a few others have had better luck.
The kitties would luv to sniff out a batch of 100 WUs to nibble on!
They'll keep on trying.

Meow meow meow!

---

"Time is simply the mechanism that keeps everything from happening all at once."

I wound up with 67 for the 15th UTC.

Cheers.

I wound up with 67 for the 15th UTC.

I got 1.
The first in about 3 days.

---

Grant
Darwin NT

I wound up with 67 for the 15th UTC.

I got 1.
The first in about 3 days.

ditto... it has been dry for a few days... I just got one... it was garbage. I did it in under a minute...sigh. I'll keep trying.

15-Apr-2020 12:44:48 [SETI@home] Scheduler request completed: got 77 new tasks
15-Apr-2020 19:29:41 [SETI@home] Scheduler request completed: got 1 new tasks
16-Apr-2020 06:47:14 [SETI@home] Scheduler request completed: got 25 new tasks

some times, needs to uptade manually ...

---

How is the decision made regarding which app version of a requesting client a resended wu is assigned to? Is it random choice by the boinc scheduler? Is it GPU first (read this somewhere)? Are my chances to get some work the same if requesting work for CPU only vs. CPU and GPU.

Strange question? If I catch one of those rare unicorns, at least I want to admire it for an hour or longer. ;-)

---

why does boinc need a manual kick to continue to ask for seti tasks? I would like it to continue asking as much as seti allows (every 30:18) but it gets lazy and stops asking and I have to manually ask when I notice.

I no longer have any seti WUs running, so this fact may play a part ( but I don't want it to).

BOINC is supposed to contact projects every 60 minutes at minimum. Sometime it does and most times it does not. But you can run a script to manually invoke the boinccmd application for a project update. Easy to do in Linux with the watch command. You certainly should be able to put something together with PowerShell or something similar in Windows and then invoking the script with Windows Task Scheduler.

https://boinc.berkeley.edu/wiki/Boinccmd_tool

boinccmd --project http://setiathome.berkeley.edu/ update

and put a 65 minute timer on it.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

No, I don't think that's quite right. The 1 hour timer applies in various places:

1) At the GPUGrid project - that's a project setting.
2) If you have completed a task for a project, and it hasn't reported yet, the project will update one hour after the task became 'ready to report' (after uploading).
3) At this project, most Tuesdays...

But there's no regular contact, unless there's something to do.

My 2500K just hit the jackpot. :-O

264 tasks just turned up. :-D

Cheers.

FYI only

There is a bug on the client code who make it sometimes "forget" about that 1 hr timer and stop to asking for work when the host is out of work, something common on this last days of S@H.
Besides travel on the code and find why that happening, correct, and compile your own client from the user side, we have very little to do unless use the Keith suggestion who forces the client to update.

@Wiggo yes you hit the jackpot! Congrats.

---

My 2500K just hit the jackpot. :-O

264 tasks just turned up. :-D

Cheers.

I just got 96 someone opened the flood gates

---

There is a bug on the client code who make it sometimes "forget" about that 1 hr timer and stop to asking for work when the host is out of work.

It's not a bug. It's irritating for compulsive crunchers, but an exponential backoff applies after repeated failures to supply requested work. It's designed to reduce communications traffic to dead servers, but unfortunately it was designed to apply universally, with no consideration of the reason why no work was supplied.

In my book, 'feeder was emptied 30 ms before you asked' is different from 'project has gone bankrupt', but not to BOINC.

And 82 here, nice! If it keeps up, 500,000 credits seems tantalisingly within reach.

---

Gazing at the skies, hoping for contact... Unlikely, but it would be such a fantastic opportunity to learn.

My alternative profile

looking through a few of the new tasks I just received, the original wingperson will timeout within 2 days and have not contacted the server for several weeks. Someone must have done some tinkering on the servers to send out tasks from hosts that, for all intents and purposed, have quit the project.

---

My 2500K just hit the jackpot. :-O

264 tasks just turned up. :-D

And then my 3570K also picked up 264 tasks. :-O

Is there something about the number 264 or was it just a coincidence?

Cheers.

Is there something about the number 264 or was it just a coincidence?

Double Jackpot? You need to try do that in Vegas.

---