Applikationer unsent in progress avg runtime of last 100 results
Sierpinski Base - short 39,342 19,734 0.30 (0.06 - 5.36)

Why can't I get these unsent WU ?

I have stopped sending them for a while because of S79.

Applikationer unsent in progress avg runtime of last 100 results
Sierpinski Base - short 39,342 19,734 0.30 (0.06 - 5.36)

Why can't I get these unsent WU ?

I have stopped sending them for a while because of S79.

Are you able to say "Do not send these work units until these other work units are finished"?

Not finished but nearly all WUs sent out, its only a backup. S79 has priority.

I noticed that the Hall of fame list was updated earlier today, but the first post at https://srbase.my-firewall.org/sr5/forum_thread.php?id=157 shows that I haven't found any primes yet.

This seems a bit strange since my profile at https://srbase.my-firewall.org/sr5/show_user.php?userid=3161 says that I've done 54,646 tasks. Those tasks did not include TF credit.

Am I just unlucky, or is it common to test 54000+ candidates and not find any primes?

I am not sure how we are going to complete the goal for S79 with 42K of the n range uncompleted.

I have discussed this yesterday with the CRUS admin. I dont think that we can finish this. Its nearly 3 times longer in the 50-100k range. The deadline is 1d and we have over 7M WUs left. We could complete it up to 80k. This is more realistic.

Is this thread dead? Because it seems to be either ALOT of people crunching here lately or there aren't alot of tasks available right now. I mean it seems every day or so a project with lots of tasks all of a sudden has none for several days to a week or more.

S66 batches are small. We can finish them in 1-2d per batch. If a new batch is loaded the weight will be set to 0 for 1-2 apps to get all hosts to S66. Its possible to process S79 this year too. Sometimes the feeder cant load enough work so it takes some time.

weight 0 = dont send out work for a specific app

The project was choosen.

Good luck everyone!

I can speak for all of us when I say, "we want to run the same amount of TF projects that we have of GPUs". Exp: 2 GPUs, we should run 2 TF projects, One for each GPU. Or one TF project on ANY GPU and run whatever project BOINC wants to run on the other GPU". Oh, wait BOINC does that on every project out there like, Prime Grid, GPU Grid, Amicable Numbers, Einstein, ext. Even all of SRBase. Well, for some reason NOT for TF projects.

Mine are also running for about 10 hours on my fastest pc, AMD 5800x, but they are taking over 30 hours on my slowest pc's, Intel Xeon running at 2.3ghz, but my main problem is the estimates for the 'time to completion' I'm getting for the tasks, 3 hours and 35 minutes!! It means I often end up aborting tasks I never even get to start before the deadline comes around. Yes I abort them and then get new tasks with a deadline of the day after those that I aborted. I have my cache settings set at 1 day plus 1/2 day additional and because they come in as going to only take about 3+ hours I get WAAAAY too many of them. I'm mostly talking about the Average3 tasks but the other time estimates while better aren't recognizing my pc's capabilities and reducing the amount of tasks I get accordingly. I thought this was a Project side setting to increase or reduce the amount of tasks sent to me based on my crunching X amount of tasks, does this Project not support that feature?

mikey

That will always happening for new apps where the server has no datas to calculate for estimated runtime. BOINC has always 0 estimated runtime after you have some results sent back to the server. The time to completion will be corrected if you getting new tasks. If you cant do the tasks until deadline the WUs will be aborted by the server.

https://srbase.my-firewall.org/sr5/forum_thread.php?id=6&postid=5765

The GPU apps can only run on device 0, it was not designed for multi-GPU.

But they run on both if I allow both. There must be a way to make it only run on the second one (device 1)? All the other projects do this.

I had 3950x and with air cooling temperatures were high as-well.
Easiest thing is to limit max temperature i thing or simply let it run in ECO mode where max consumption is limited to 87W.

Iam expecting out of work tomorrow. I need to check the average results first before I can load new work, the database hits nearly the limit.

Finally got an "M" badge :)

so you are talking about things like:
Intel x86-compatible CPU

will be stopped but things like:
AMD x86_64 or Intel EM64T CPU
and
Intel 64bit Mac OS 10.5 or later

will continue to be sent out correct? If so it looks to me like you already have 64bit apps for everything.

Thats correct, I have also planned an update of all 64bit apps with llr2 from PG, still in test...

Thanks.

On a side note...you have to be one of the best Project Admins in responding to posts (and quickly) on their forums. A lot of Projects have Admins that are MIA and just don't give a shite. So, thank you for that!