Posts by mikey
log in
1) Message boards : News : Award Badges Going Live Soon! (Message 66832)
Posted 5 days ago by mikey
I made a 9 and 10 year badge today. They should be released tomorrow, just waiting for a chance to see Sidd and Heidi to make sure they think they look good.

I also checked to see if we need a new credit badge, but we are still a couple billion credits short of needing that.

Jake


Thanks!!
2) Message boards : News : Award Badges Going Live Soon! (Message 66830)
Posted 6 days ago by mikey
Yes people have to be getting close by now.
3) Message boards : Number crunching : Nvidia GTX 1080 Ti vs AMD RX Vega 64 (Message 66765)
Posted 4 Nov 2017 by mikey
I can't decide what GPU to buy.

Gigabyte Aorus GTX 1080 Ti Waterforce Extreme Edition
3584 CUDA cores / 1746 MHz
12.5 TFlops

AMD RX Vega 64
4096 Stream Processors / 1546 MHz
12.6 TFlops

Which one is faster for BOINC crunching?


If it were me I'd get the 1080Ti as it's been around longer and is better supported, it also a couple hundred seconds faster at Einstein, but that doesn't mean it's faster at every project.
4) Message boards : Number crunching : How to configure client to use GPU-only for MilkyWay@Home while still using CPU for N-Body Simulation (Message 66764)
Posted 4 Nov 2017 by mikey
Thanks for the response. However, even if that was a proper solution to my problem I wouldn't be able to implement it. I'm currently in a gridcoin pool so I have no control over the project settings.

I was hoping for a client-side solution.

~KT


I thought of another way....do two different installs of the Boinc software on your pc, making each is in it's own set of directories, and don't use the gpu on one of them and only use the gpu on the other one.
5) Message boards : Number crunching : How to configure client to use GPU-only for MilkyWay@Home while still using CPU for N-Body Simulation (Message 66721)
Posted 21 Oct 2017 by mikey
I've created an app_config.xml file to run multiple GPU work units:

<app_config>
<app>
<name>milkyway</name>
<gpu_versions>
<gpu_usage>.2</gpu_usage>
<cpu_usage>.2</cpu_usage>
</gpu_versions>
</app>
</app_config>


I want my client to get CPU work units for the N-Body Simulation application, but I'd like to disable CPU work units for MilkyWay@Home 1.46.

I tried disabling the CPU by adding the config below to my app_config but it doesn't seem to have an effect. I thought maybe the empty plan_class tag would tell the CPU app to use 0 CPUs, but still allow GPU tasks.

<app_version>
<app_name>milkyway</app_name>
<plan_class></plan_class>
<avg_ncpus>0.0</avg_ncpus>
</app_version>


Is it possible to use GPU-only for 1.46 and still use the CPU for N-Body?

Thanks,

~KT


On the same computer I've never seen that before, on different pc's you can thru the 'venues' home work and school, just turn off the cpu tasks on one and the gpu tasks off on the other one.
6) Message boards : News : Clearing the Queue (Message 66720)
Posted 21 Oct 2017 by mikey
You would at least score any points for those tasks that were cheated, but because of your fault they were not recognized as correct.

Assignments are rejected by hundreds, and the CPU time is spent on them.

Because now all the dissatisfied will be cut off from the project and there will be no one to cheat the tasks.


I think you meant to use the word 'crunched' instead of "cheated", but yes I knew that.

But as I said earlier this same scenario played out a few months ago where hundreds of tasks were lost by me alone, not counting all the tasks lost by others that were all in various stages of being crunched. I only keep a 1.5 day cache of workunits on my pc's, that's still alot considering I can do a gpu unit in about 70 seconds and to lose them twice in a few months just isn't worth my time and effort anymore. I even contributed cash to this project at one time.
7) Message boards : News : Clearing the Queue (Message 66714)
Posted 20 Oct 2017 by mikey
I stopped too, this is the 2nd time this has happened this year, 500 workunits of mine each time, just gone poof all the work for nothing.
8) Message boards : Number crunching : Suddenly lot of validate errors on Nvidia? (Message 66711)
Posted 20 Oct 2017 by mikey
hsdecalc -

You have upwards of 200 that aren't validated yet because they have to be compared to other users results. They are not "can't be validated".

Of the few tasks that you had that were "invalid" (approx. 10), you will notice that every other user shows them "invalid" too. This suggests there was something wrong in the data not your computer.

You also have about 4 that were "error". Sometimes that happens.


They had database problems the other day and had to cancel hundreds of workunits!! It in the News section.
9) Message boards : News : GPU Issues Mega Thread (Message 66710)
Posted 20 Oct 2017 by mikey
Sorry wrong thread!
10) Message boards : News : Clearing the Queue (Message 66709)
Posted 20 Oct 2017 by mikey
All the work I did today is in the state of 'invalid'(Completed, can't validate)
or 'Validation inconclusive'(Completed, validation inconclusive)
Not a single wu is validated today!
State: All (2124) · In progress (159) · Validation pending (1) · Validation inconclusive (528) · Valid (393) · Invalid (554) · Error (489)


I stop doin' work for the wastebasket....


I sent this to a teammate the other day:
State: All (3660) · In progress (400) · Validation pending (0) · Validation inconclusive (481) · Valid (2779) · Invalid (0) · Error (0)
Application: All (3660) · MilkyWay@Home (3660) · MilkyWay@Home N-Body Simulation (0)

Right now I see this:
State: All (1148) · In progress (80) · Validation pending (0) · Validation inconclusive (80) · Valid (388) · Invalid (216) · Error (384)
Application: All (1148) · MilkyWay@Home (1148) · MilkyWay@Home N-Body Simulation (0)

Today every single on of the 400 in progress and 481 inconclusive units were cancelled!! On top of that the same thing happened a few months ago!!! I will NEVER put all of my pc's here at MilkyWay again, this is getting ridiculous!!! I put my pc's here by choice and the project can't seem to figure out how to respect the help I/we give them. I have crunched for MANY projects and NOT ONE of them has had the problems that MilkyWay is having recently!!
11) Message boards : News : Clearing the Queue (Message 66702)
Posted 19 Oct 2017 by mikey
I have noticed that recently a number of completed work-units (MilkyWay@Homev1.46) have been marked as "Didn't need", only some of which have been allocated credit. Please explain!

Also I obtained a number of "Completed, can't validate" results. Again I suppose this is due to the recent Transitioner problems etc.

Alan Barnes


If you look under your "didn't need" workunits you will see that they each say "errors WU cancelled", so they too were part of the clearing out process.
12) Message boards : Number crunching : Errors (Message 66644)
Posted 18 Sep 2017 by mikey
all broke:
de_modfit_fast_18_3s_146_bundle5_ModfitConstraints WithDisk_Bouncy_Random_1

runs fine:
de_modfit_fast_Sim19_3s_146_bundle5_ModfitConstraintsWithFixedDisk_Bouncy_2_Random_1

Win 7
AMD 208x
Driver 17.7.2

Boinc 7.6.33 and 7.8.2

Please Admins stop generating junk und fix this.


It was the weekend, no one is there over the weekend, today is Monday and they should be back to work around 8 to 10 am CST, which is about -6 GMT.
13) Message boards : Cafe MilkyWay : Fortschritt (Message 66606)
Posted 13 Sep 2017 by mikey
Hi Felix,
das würde ich auch gerne wissen: Wieviel ist bereits gemacht, wieviel fehlt noch? Da es auch keine Grafik zu den einzelnen Tasks gibt (und ich diese bei SETI@home einfach nicht zum Laufen bekomme), und es mir zu blöd ist den Fortschrittsbalken zuzusehen, steig ich aus.

Hi community,
I am interested in overall progress of the project as well as Felix does. There is no grahics available while a task is calculated (I should tell you that I dont see any grahics with BOINC, even not with SETI@home), and I am fed up watching those progress bars developing towards 100 per cent, so, I am out.

greetings, Hannes [guestAT]


There are PLENTY of gpu workunits here, that's all I'm running here!! Maybe if you stayed we could help you figure out why you don't get them, but since you are gone good luck!!
14) Message boards : Number crunching : Varying GPU time per task (Message 66599)
Posted 9 Sep 2017 by mikey
Are you saying all hosts should be taking longer to process tasks? Is there somewhere where this info is posted?


I have no clue if they announced it Officially or not, they did say new units are here, don't know where I saw that now though, and they just take longer than the previous ones.
15) Message boards : Number crunching : Varying GPU time per task (Message 66596)
Posted 8 Sep 2017 by mikey
I've noticed for a while now that the time projects take to complete seems to vary over time somewhat unpredictably. When it's good, I'm crunching a task every ~65 seconds but this past week, for example, I've been closer to 180-350 seconds. I've been seeing my RAC dive from about 110k to about 60k now. Just want to know if this is normal or if it's something I might want to look into. I've performed plenty of BOINC resets, computer shutdowns, and graphics card updates to feel confident none of those is an issue.


One thing is you are doing different tasks right now and the new tasks are taking twice as long as the old ones.
16) Message boards : Number crunching : No gpu work units? (Message 66588)
Posted 5 Sep 2017 by mikey
Problem solved

It was the last official driver problem.. just uninstalled it and let windows update to get the driver only and everything is fine


Glad you got it fixed!!
17) Message boards : Number crunching : No gpu work units? (Message 66582)
Posted 4 Sep 2017 by mikey
Still no work units!!

Anyone having the same problem?


I'm still getting workunits though, so there must be some!!
18) Message boards : Number crunching : No gpu work units? (Message 66579)
Posted 2 Sep 2017 by mikey
Greetings:

I just returned to milkyway project after formatting my pc and noticed no gpu work units. Any problem?

Thanks


As of Server software version: exported / 2 Sep 2017, 19:13:29 UTC

they had Tasks ready to send 3,923

Maybe Jake Weiss can tell us why that number is so low.
19) Message boards : News : Scheduled Server Maintenance Today (Message 66542)
Posted 16 Jul 2017 by mikey
This is from my main box;
State: All (4793) · In progress (160) · Validation pending (0) · Validation inconclusive (516) · Valid (4113) · Invalid (4) · Error (0)


Rut Row!!
20) Message boards : Number crunching : GPU missing (Message 66541)
Posted 16 Jul 2017 by mikey
My Sapphire AMD HD 5850 1Gb card has been faithfully crunching away problem free for more than a year but now my BOINC app is reporting "GPU missing"? and is not crunching anything. Wonder if a fresh re-install of BOINC would solve that problem or perhaps its' something more minor.


Whatever it was it seems to be working as you have a bunch of new tasks and even some valid ones now.


Next 20

Main page · Your account · Message boards


Copyright © 2017 AstroInformatics Group