Posts by HAL9000


log in
1) Message boards : Number crunching : Show us your cruncher(s) (Message 1550307)
Posted 1 hour ago by Profile HAL9000
So much for show and tell . HELP !!!!!!

do the pic's have to be a certain size ?

i drag pic to post after i put IMG but something weird happened it show'd up in a seperate browers page .......

So i know this is annoying but could you please tell me how to do it i'm using Chrome or is that the problem "Chrome"

This site does not host images.
The image tag, as noted in the Use BBCode tags to format your text help, is just a pointer to where the image is elsewhere on the internet.

If the images are huge, for the sake of those with slower internet connections, it is generally nicer to just provide a link to them instead of using the img tag.
2) Message boards : Number crunching : FEED ME MORE - FEED ME MORE! (Message 1550282)
Posted 3 hours ago by Profile HAL9000
I still think if we cut the number of AP splitters down to one or two instead of 6, AP splitting would slow down closer to the speed of MB splitting and then we wouldn't end up with these huge backlogs of waiting for MB to catch up...but I guess the staff doesn't see a problem with the way things are and it seems to work fine..it just kind of annoys some of us crunchers, but it makes no difference to the science. *shrug*


At the moment there are only 5 of the 7 MB splitters working with the other 2 being disabled and they have been for some time now. It would improve things some what if/when they were put back online.

The 5 MB splitter are generally more than fast enough to keep up with demand. They actually cycle off every so often as the amount of work RTS is large enough. If there is an error it takes a bit of time to catch up.
Some time ago, shortly after moving to the CoLo IIRC, it was mentioned that they limit the speed/amount of the splitters because they were hitting a disk i/o limit with the storage array.
3) Message boards : Number crunching : Show us your cruncher(s) (Message 1550280)
Posted 3 hours ago by Profile HAL9000
Haven't done one of these in a bit.


Gaming system
-i5-4670K w/ Noctura D-14 cooler
-Gigabyte GA-Z87X-D3H
-16GB Crucial Ballistix 1.35v kit
-Sapphire HD6870
-Rosewill 650W 80Plus Platinum PSU
-Stuffed in a NZXT SOURCE 220 case


Low powered system for running various network things.
-Celeron J1900 w/ random 80mm fan plopped on heatsink & ram
-ASRock Q1900-ITX
-4GB from 2 mismatched 1.5v 2GB DDR3 SODIMMS I had sitting around
-Intel HD Graphics
-Old Enermax 425W 80Plus Bronze PSU
The 3.5" HD is just extra load so the system will turn on.
-Elevated on a fancy bit of the box the MB was shipped in
4) Message boards : Number crunching : non-abandoned, abandoned tasks (Message 1550073)
Posted 10 hours ago by Profile HAL9000
My machine: ID: 7023772

Looking at the task list shows "abandoned" and there are no work units in progress.

Looking at the machine itself shows that all of the tasks are still there and running as they normally do. They have 8/21 deadlines. The data and time are correct on that computer, also.

As far as I know, absolutely nothing happened to the computer.

Does anyone have a theory?

Is there anything in your BOINC logs at 17:34:07 UTC when all of the tasks got stuffed?
5) Message boards : Number crunching : Puzzled, once more (Message 1550071)
Posted 10 hours ago by Profile HAL9000
It appears, from the little I've observed, that my ATI Radeon 6950 does less crunching when there is no monitor attached to it, does that make sense? Must admit, it doesn't to me ;-)

P.

If the monitor goes to sleep, from the Windows power save settings, the GPU slows down to its lower clock rates. With a "no monitor" condition I would expect the same behavior.
If a monitor is connect but powered off this does not happen.
6) Message boards : Number crunching : "Waiting to run" / "Scheduler wait" (Message 1549931)
Posted 13 hours ago by Profile HAL9000
Glenn,

I think it's a driver issue. Not sure how to go about correcting it but for now it runs and the results validate so I'll keep muddling through it. I think Joe is correct about the setting should be 1-2 minutes rather than 120. Thanks Glenn and Joe for your help.


Zalster

If you select Activity > Use GPU always does the GPU run normally?
7) Message boards : Number crunching : CUDA Versions (Message 1549894)
Posted 14 hours ago by Profile HAL9000
All the wus/tasks are the same. There is no such things as a "CUDA task" of any version. The "version" is determined by the app that was assigned when the work was requested.
When using the stock applications the server sends each app. Then once it determines which runs best on your host it will then run it exclusively or most of the time. It may check that the other versions at a given interval. I don't recall.

With the optimized apps you can specify the CUDA version at the time of the installation. However, when using the optimized apps you have to update the apps manually when they change.

IIRC app_config.xml takes precedence over app_info.xml. So you may not have to modify anything if you choose to run the optimized applications.

If you are going to get the optimized apps. Once you get to the Lunatics downloads. Looks for the line They are available from arkayn's Crunchers Anonymous Windows downloads and Mike's download page.. Then on either site get the Lunatics v0.41 setup for the platform you need.
8) Message boards : Number crunching : Astropulse 601 - when to give up? (Message 1549580)
Posted 1 day ago by Profile HAL9000
Have you guys used optimized clients ?

They're almost mandatory in order to get decent performance, especially from older CPUs.

I just ran an Astropulse on an old notebook (Pentium M 1.86GHz) and it took it some whopping 61 hours to complete, using the stock linux client.

I'll dump some old Vista32 onto it and see what it can do with the latest optmized clients.

My old 1.5GHz Pentium M notebook does an AP in about 20 hours with Windows 7 x86. Using ap6_win_x86_sse2_cpu_r2137 optimized app.
9) Message boards : Number crunching : The current top cruncher (Message 1549285)
Posted 2 days ago by Profile HAL9000
Yeah their output has been tapering off recently to only about 6,000,000 a day.
10) Message boards : Number crunching : Astropulse 601 - when to give up? (Message 1549282)
Posted 2 days ago by Profile HAL9000
Your computer shows 0 astropulse tasks. It looks like it was already assigned to another host, turned in, and deleted. I have a system similar to yours, but a bit slower. I have not run AP on it since AP v5, but I don't recall it taking 2-3 days to complete 1 AP task.
11) Message boards : Number crunching : FEED ME MORE - FEED ME MORE! (Message 1549276)
Posted 2 days ago by Profile HAL9000
You can have more AP when you finish with all of your MB!
12) Message boards : Number crunching : Some questions about BOINC for Android... (Message 1549193)
Posted 2 days ago by Profile HAL9000
Folks, I could still need some help here.

Even with Claggys update-trick I can not get more then 4 WUs on my phone. And it's just 4 if they are shorties, Usually I can just get a max of 3.

The problem is that I need the charger at work for my other phone and I don't have Wifi at work. So with just 3 tasks I have to carry the phone from work to home every 2 days or so.
So I need more tasks on the phone, enough for 1 week. I would take the phone to work on monday, let it run there the whole week and take it back home on friday to upload & report the tasks done and fetch new work.

I have root access on this phone so I thought I just have to edit some files, but it looks like I can't find the correct file to get it done.

First I went to data/data/edu.berkely.boinc/client There is a file called global_prefs.xml that includes a line "<work_buf_min_days>". I changed the value there from 0 to 7 but nothing happens.
Then I found a file called sched_request_setiathome.berkeley.edu.xml which includes the same line. Again I switched the value to 7 but again no luck.


So can maybe anybody tell me which line in which file I have to edit to get more work on my phone?

I am pretty sure global_prefs.xml is populated from server data. So changing it may not have any effect.
In your BOINC settings you would want to set Maintain enough tasks to keep busy for at least n days to your desired value for the venue where this device is set.
13) Message boards : Number crunching : Puzzled... (Message 1549030)
Posted 2 days ago by Profile HAL9000
It's 100 tasks per GPU now, rather 100 tasks for the GPU(s).

Claggy



Ok, but my Intel chip has never been used by Boinc, so effectively
I have only one gpu, the ATI card. C'est la vie I guess ;-)

Thanks for the reply.

P.

Yeah it is kind of a bonus for some in that way. It has happened because of some kind of error in a fix they were trying to implement for BOINC. Which was intended to limit each type of GPU to the designated task limit.


Ok, now I'm getting confused. Phil states that his iGPU is not being used, but BOINC recognizes it. I have a 4770K w/ a HD4600 iGPU, which BOINC does not recognize. BTW, it is explicitly set 'Use Intel GPU No' in my project preferences in my account.

So long as BOINC detects the iGPU it will increase the GPU task limit by 100. Changing "Use xxx GPU" has no effect on the limit. It is something else that we noted as broken when this change was implemented.
The GPU not being used would have to be disabled in BIOS or BOINC would have to be told to ignore it, via cc_config.xml, to not be included in the limit calculation.

On my i5-4670K systems I had to use <ignore_ati_dev>0</ignore_ati_dev> if I wanted BOINC to only consider my iGPU for task limit. However this presented issues for me. As I am using a version of BOINC that does not officially support iGPUs.
14) Message boards : Number crunching : Puzzled... (Message 1548950)
Posted 2 days ago by Profile HAL9000
It's 100 tasks per GPU now, rather 100 tasks for the GPU(s).

Claggy



Ok, but my Intel chip has never been used by Boinc, so effectively
I have only one gpu, the ATI card. C'est la vie I guess ;-)

Thanks for the reply.

P.

Yeah it is kind of a bonus for some in that way. It has happened because of some kind of error in a fix they were trying to implement for BOINC. Which was intended to limit each type of GPU to the designated task limit.
15) Message boards : Number crunching : Points count? (Message 1548947)
Posted 2 days ago by Profile HAL9000
Also you can use sort of use SET@home as a baseline to see how much other projects inflate their credit.
16) Message boards : Number crunching : FEED ME MORE - FEED ME MORE! (Message 1548834)
Posted 2 days ago by Profile HAL9000
In the past I seem to recall AP work being much more common. With the few AP only crunchers only mentioning running out of work every few months.

- With the introduction of v7 of MB a large difference in credit between MB & AP lead more people to run AP only.
- We now have a Nvidia AP app. So the crunchers that have beefy GPU systems can tear though the AP work faster than ever.
- With people favoring AP work over MB the MB tasks are not getting burned through as quickly as they otherwise would.

- Moving to the colo may also play a part in things. As they move the data for splitting to the servers and back over the network. I'm not sure if that is how it was done before or if they were just plugging drives into the system to dump the data.
17) Message boards : Number crunching : Some questions about a laptop (Message 1548652)
Posted 3 days ago by Profile HAL9000
Another small update here: The machine is running without any problems so far, at least on CPU.

Last week we had a rainy day here where temperatures were abut 10 degrees (celsius) lower then the days before so I thought I give the GPU another try. But it took again less then 15 minutes until core temps hit 90 degrees and I had to suspend the GPU again. I wonder how such a small GPU can produce that much heat?!?

THE CPU itself, although running in burst all the time, seems to be a bit slower then suspected. When I compare times and GFLOP-ratings with Hals Bay Trail machine, his seems to run a bit faster. But thats ok, for a machine that sucks just ~16 watts the performance is still nice.

The Bay Trail chips have a maximum core temp of 105°C, but that doesn't mean they will be happy running at 104°C. I prefer to only run up to about 10ºC shy of the maximum temperature that Intel states. I imagine the built in throttling mechanisms should kick in before temps get to near the maximum values. It is better to be safe than sorry with something you want to last a long time though.

You really think that running the CPU at 95 degrees 24/7 would be ok?

I think I should better qualify what I said.
If that high of core temperature is not effecting any other components of the system. In such a way as to make them exceed their thermal design. I would let my system run to a core temperature within 10ºC of what Intel specifies.
If the heat caused by the CPU is causing issues then it is obviously not OK for the system as a whole.
Having said that. I have an old Pentium M notebook I have been trying to kill. It might be better to say I am trying to burn it out rather than maliciously kill it. I have it's little 1.5GHz chip overclocked to ~2.4GHz and running AP non-stop. During the day its core has only been getting up to 85ºC. As it has not yet been that hot this summer.

On paper I would think that your Bay Trail system would come out ahead.
- Maybe it is not constantly staying at the Burst frequency?
- Maybe it is that I have 2 dimms in my system. In order to take advantage of the dual channel abilities of the CPU. Memory bandwidth of Bay Trail is 10.6GB/s for single channel & 21.3GB/s for dual channel. When I was selecting a MB for Bay Trail I made sure to get one that had 2 dimm slots. There were many I found that only had 1 slot.

At the moment I only have 1 CPU MB tasks to look at in my results. It is a VLAR that ran 4.5 hours. Where your normal AR tasks look like they are running about 6 hours. Most of your recent CPU MB tasks have been VHAR, or shorties. I think that might lower the "Average processing rate" GFLOP rating of an app, but I'm not 100% sure on that one. However, AP processing rates are much closer than MB.
18) Message boards : Number crunching : A journey: iGPU slowing CPU processing (Message 1548628)
Posted 3 days ago by Profile HAL9000
Wow, so the iGPU is really slowing down the CPU to 50% ?? That's interesting!

For my Haswell, i5-4670K, systems.
Yes, For SETI@home project running any combination of MB or AP tasks with CPU & iGPU has the effect of about double CPU run times. I had hoped it would be something simple like "you are doing this wrong". However, I think what Joe mentioned about the cache is accurate.
Probably good to mix non cache heavy project on either CPU or iGPU. Then SETI@home on other processing device.

For my Silvermont/Bay Trail-D system.
CPU tasks are slowed about 8% when running iGPU. At least for AP. I must still do more test to see what the % is for MB tasks, but it may be the same.
It appears that iGPU produces enough output to more than make up for slower CPU times. Much like when running Hyper-Threading on a CPU. The processing times are slowed 15-30%, but output increases. Because of greater number of tasks completed.
19) Message boards : Number crunching : Considering new Graphics card (Message 1547432)
Posted 5 days ago by Profile HAL9000
I don't wanna get a card that uses more power because I'm not sure if the power supply could handle it. The 630 would for sure be a improvement, my G210 has 16 cuda cores IIRC , LOL.
The only thing that makes me a bit unsure about this GT630 is the fanless design.

I attached a fan to my fanless GT 8500. The fanless design they implemented could not really handle running SETI@home 24/7. I just use some small 60mm fan I had lying around and connected it to run at 5v, instead of 12. So it just pushed the heat away from the heatsink.
20) Message boards : Number crunching : A journey: iGPU slowing CPU processing (Message 1547427)
Posted 5 days ago by Profile HAL9000
Ah, okay...

Running a similar setup (albeit with a rather slow mobile Core i3) myself since just a few days, I also found out the performance ratio on SETI comparing CPU and iGPUs is unexpectedly close.

So in this case, you're right as the iGPU doesn't contribute so much to the overall score when bundled with fast CPU cores.

My Bay Trail system the iGPU I think adds to overall output. I am still early into test, but so far the data looks like this.

4 CPU AP + 1 iGPU MB
~23.3hr CPU time + ~5.5hr iGPU time

4 CPU AP + 0 iGPU
~21.5hr CPU time

Using 650 average AP credit & 90 average MB credit
4 CPU AP + 1 iGPU MB
~3070 estimated RAC
4 CPU AP + 0 iGPU
~2900 estimated RAC

When running SETI@home for both CPU & iGPU it looks like overall output is a gain. Most notably CPU is not slowed to nearly double run time when using iGPU. Some CPU slowdown must be expected as iGPU does share resources with CPU. So far CPU is slowed only 8% it would seem.


Next 20

Copyright © 2014 University of California