Posts by Raistmer


log in
1) Message boards : Number crunching : I've Built a Couple OSX CUDA Apps... (Message 1791865)
Posted 4 hours ago by Profile Raistmer
Also, any similarity with AMD issue on OS X that Chris follows very good? Gaussians missing there - what pattern here (NV) if any?
2) Message boards : Number crunching : I've Built a Couple OSX CUDA Apps... (Message 1791861)
Posted 4 hours ago by Profile Raistmer
I believe you are aware of a post on main about this? I think you were sent some files....

Better to stop referring to my awareness. First of all I don't follow Mac issues too closely at all. I have lot of other things to remember and recall everyday, not to remember Mac issues. So if I ask smth - better just to try to cooperate (if ultimate goal is problem solving, not futile arguing ).
Also, all those rockets, sailors and fruits(?) tell nothing to me at all. OS X version/driver version. I never saw "Yosemite" in stderr output.
As you could know non-Windows binaries provider is Urs. If I need to intervene - then I need solid info what problem is, what affected range and so on. So far I see that we need plan_class correction to exclude OS X 15.x - right? What else?
3) Message boards : Number crunching : I've Built a Couple OSX CUDA Apps... (Message 1791843)
Posted 5 hours ago by Profile Raistmer
You can start here, http://setiathome.berkeley.edu/forum_thread.php?id=78569&postid=1762177#1762177
Like I said, any Mac nVidia LapTop running Darwin 15.x.

Why you highlighting laptop nature of host. Non-laptops not affected with same driver/OS version?

Also, what driver versions affected?
I see 10.10.10 310.42.25f01 is bad one. What another?

EDIT: regarding why not to replace with smth better - you know procedure cause AFAIK some of your builds were at beta at least (released to main already btw? ).
You need to send binaries to Eric with descriptions and suggestions about corresponding plan_class they can work under.
And then to "ping" him until binaries will be deployed to beta :D
4) Message boards : Number crunching : I've Built a Couple OSX CUDA Apps... (Message 1791835)
Posted 5 hours ago by Profile Raistmer
It's real easy, just look for any Mac nVidia LapTop that's running El Capitan.
This is the First LapTop at Beta, http://setiweb.ssl.berkeley.edu/beta/results.php?hostid=62183&offset=40
The others are similar.
I've posted links before, surely you saw them.


App on main behaves the same? I would like to see links for main (production app, not beta one).

this particular host aborted work on main.
5) Message boards : Number crunching : I've Built a Couple OSX CUDA Apps... (Message 1791833)
Posted 5 hours ago by Profile Raistmer
It's real easy, just look for any Mac nVidia LapTop that's running El Capitan.
This is the First LapTop at Beta, http://setiweb.ssl.berkeley.edu/beta/results.php?hostid=62183&offset=40
The others are similar.
I've posted links before, surely you saw them.


App on main behaves the same? I would like to see links for main (production app, not beta one).
6) Message boards : Number crunching : OpenCL MB v8.12 issues thread attempt 2 (Message 1791831)
Posted 5 hours ago by Profile Raistmer
Just using -period_iterations_num 300 does not help screen lag.

Maybe it means lag not from kernel I expected it to be. But to check that I need either hardware at own disposal or someone who can use nVidia profiler tools like NSight and experience lags.
7) Message boards : Number crunching : I've Built a Couple OSX CUDA Apps... (Message 1791826)
Posted 5 hours ago by Profile Raistmer
The current stock App produces nearly 100% Inconclusive results on the LapTops with the current OS, it's been that way for quite a while. How much longer should we wait to replace something that doesn't work, 6 months, a Year? You can always update the app later, all those bad results are unrecoverable.

Few links to such hosts please.
8) Message boards : Number crunching : OpenCL MB v8.12 issues thread attempt 2 (Message 1791822)
Posted 6 hours ago by Profile Raistmer
please post links to results that used -period_iterations_num 300 and those that used -use_sleep.

Also try -period_iterations_num 500 -sbs 512

P.S. as debugging measure please try to run -use_sleep -v 6 on non-GUPPI task and send me stderr.txt from slot directory.

Default value for -use_leep's Sleep is 1. sleep_ex values can be from 0 to any reasonable positive integer number (sleep time in ms )
9) Message boards : Number crunching : Credits seem very low with GTX 980? (Message 1791680)
Posted 14 hours ago by Profile Raistmer
Hi all, had a question, previously I was crunching with a GTX 980 and was getting around 10k-15k credits/day from what I remember. I started crunching again and now I'm only getting 6k-8k per day or so and wasn't sure why. My hardware and OS (Windows 7 x64) is identical to what I had before, and I'm using around 80%-99% of the GPU typically, as I was before. Three concurrent WUs on the GPU, same config etc. files as before also.

Any thoughts? Thanks

Изменился набор рассылаемых заданий. Текущие, при бОльшем времени счета, дают меньше кредитов.
Проблема связана с несовершенством системы расчета начисляемых кредитов. И усугубляется относительно более низкой производительностью ГПУ для нового типа заданий (по сравнению с ЦПУ).
10) Message boards : Number crunching : OpenCL MB v8.12 issues thread attempt 2 (Message 1791640)
Posted 20 hours ago by Profile Raistmer
no difference wether 1 or 4 monitors...

Then it seems only driver difference with working w/o lags system remains important. Worth to try downgrade from 36x.xx to 35x.xx driver.

Did it help with lags?
11) Message boards : Number crunching : Why need 5 different stock AMD OpenCL GPU applications? (Message 1791364)
Posted 1 day ago by Profile Raistmer
Though I must say that doesn't instill me with much confidence in the development teams.

Instead of religious questions here no confidence needed for things to work. They work because of testing and coding and debugging, not because someone confident or not in them :D
12) Message boards : Number crunching : Why need 5 different stock AMD OpenCL GPU applications? (Message 1791361)
Posted 1 day ago by Profile Raistmer

Well yes, that is the topic at hand. Glad you can point out I'm on topic. Now, what to do about it? I'm sure I already know your answer though.

I know answer of course - to continue optimization efforts. And to ignore such threads to save time, perhaps? But I don't like to tolerate wrong rumors.
13) Message boards : Number crunching : Why need 5 different stock AMD OpenCL GPU applications? (Message 1791359)
Posted 1 day ago by Profile Raistmer
Why wasn't there more focus on simplification and unification in the first place?

Because I want to maximize project throughput. And unification+simplification though reduce my time spent on the project will result just in opposite.
Taking to the limit, most easy and simply way to do work... just doesn't do it at all. Use stock CPU app only, for example.
And unification w/o simplification actually will require much more efforts.
14) Message boards : Number crunching : Why need 5 different stock AMD OpenCL GPU applications? (Message 1791349)
Posted 1 day ago by Profile Raistmer
Excellent. Since you're already part of development and testing, I take this as tacit acceptance of the challenge laid out before you.

And you had any doubts that to make optimized software instead of "just working" is challenging? Hm :) Yes, it's a challenge. And very few times when I asked some of "advisers" to implement their "advise" I got any code in return.
Yep, it's challenging, it's the art. And almost always it can be done better then it done already. So what? Do better, be part of the team - all just benefit from that.
15) Message boards : Number crunching : Why need 5 different stock AMD OpenCL GPU applications? (Message 1791343)
Posted 1 day ago by Profile Raistmer

How many times have I seen you assisting ATi/AMD users with a specific problem to modify the command line .txt file or to use a specific driver version or both? Still think users don't have to do anything but let apps run?

And your point here? To illustrate how far we from one size fits all? Yes indeed we are very far from that. Part of answer on thread title? no?
16) Message boards : Number crunching : Why need 5 different stock AMD OpenCL GPU applications? (Message 1791339)
Posted 1 day ago by Profile Raistmer
So what`s the real problem in this case ?

As I already said, the real problem that some want to transfer tech-oriented forum in cafe filial.
Self-imposed issue that actually only in mind of asking.
If one cares about how it will scale to iPhone - iPhone will get just baseline app (though I doubt it ever happened). ATi users will get apps more suitable for particular hardware. Also this will allow to select what path most suitable and allow further improvements.
17) Message boards : Number crunching : Why need 5 different stock AMD OpenCL GPU applications? (Message 1791335)
Posted 1 day ago by Profile Raistmer


I don't think the intent is to make this personal for the ATi/AMD developers and testers, but rather to make development, testing, and supporting easier. Jord mentioned that the OpenCL standard was supposed to simplify our lives with a single unified application, yet it seems everyone here at SETI (Green Team and Red Team) have chosen the opposite path - all with the mindset of getting the most optimized speed available.

I think the question really comes down to: Can we reconsider our path and can we simply this process?

No. If one wants to simplify testing and supporting one can take part in both activities.
18) Message boards : Number crunching : Why need 5 different stock AMD OpenCL GPU applications? (Message 1791331)
Posted 1 day ago by Profile Raistmer
It would run on one application but seti members want more.
Especially speed.


I think Jord's point is that the stock app should be universal without speed optimizations with the focus on simply just working stably. Then the Lunatics pack should contain all the speed improvements and various applications required to achieve the most speed possible.

I'm not so sure that all SETI members want more, though I'm sure many do. There are some people out there that simply want to contribute and are happier with stability and simplicity over speed. Those that want speed can follow the same advise we've given for the past 10 years: install optimized applications.

Users who want simplicity just attach to project and go to own tasks and deeds. No matters 1 app, 10 app or few GB ow initial download as some projects have.
It's too curious ones who wonder. And then, when they got full descriptive answer they wonder why so complex??? Because!
19) Message boards : Number crunching : Why need 5 different stock AMD OpenCL GPU applications? (Message 1791296)
Posted 1 day ago by Profile Raistmer

Mere users don't really need to dive into the highly technical BOINC documentation to get an answer to their question.

Depends on questions and details they would like to have.Of course if answer is "because" no any details needed :)
20) Message boards : Number crunching : Decided to Bail Out on SETI for a Short While (Message 1791293)
Posted 1 day ago by Profile Raistmer
So, can you explain better what a 6-month GBT task would look like (download data size, perhaps), what mathematics would be performed, and why it can't be broken down into smaller sub-sections? That's asked out of genuine curiosity.

Would I be project scientist - perhaps. I will know design decisions when they will be formed. From common sense to get improvement over Arecibo data longer time periods can be used for signal accumulation so longer initial arrays for FFA. Smth like large FFA or bigger. Targeted search in general allows longer time domain for signal accumulation that results in better signal/noise ratio => increased sensitivity on the same hardware.


Next 20

Copyright © 2016 University of California