SETI orphans
Message boards :
Number crunching :
SETI orphans
Message board moderation
Previous · 1 . . . 21 · 22 · 23 · 24 · 25 · 26 · 27 . . . 43 · Next
| Author | Message |
|---|---|
 Raistmer  Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 
 |
Hm, I checked new BOINC options - very nice addition I really appreciate - now can set device name just as for Windows version. It's much more convenient to use device given name than just some auto-assigned ID... but also number of usable cores slider still in place. I set mine to 4 cores but to use all 8 cores still configurable too. What do you mean then? Not BOINC but Android OS restriction? If 8 tasks would be started simultaneously their CPU time will be much different than elapsed so only 4 CPU cores of 8 are allowed for them - that's ?
Also listed as OpenCL 2.0 supporting so same app should go if it would be written...
I avoid to root mine too still so seems can't use cc_config too (I have non-rooted access to Android/data on internal storage but there is no BOINC package so it uses some another inaccessible on non-rooted devices place). To cope with this issue on NativeBOINC there was special import cc_config menu item so user could import cc_config from SD card (and export too). That's good thing to use in "official" BOINC for Android (pity that NativeBOINC sources seems to be ignored by BOINC main devs, that was truly good app but compatibility with modern Android OS is lost unfortunately). SETI apps news We're not gonna fight them. We're gonna transcend them. |
Richard Haselgrove  Send message Joined: 4 Jul 99 Posts: 14654 Credit: 200,643,578 RAC: 874 
 |
Pity that E@h hasn't smth like our APR to gauge app performance it seems...The APR system was added at the same time as CreditNew (and integrated with it), as a multi app replacement for DCF. Einstein made a conscious decision to reject CreditNew, and thereby rejected APR as well. They have effectively been maintaining their own private fork of the BOINC server code for the last 10 years - but they still haven't added their own private replacement for the still-running single-valued DCF - whether APR or anything else. We live with it. |
|
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14654 Credit: 200,643,578 RAC: 874
|
Oh - I'd forgotten about that. (I submitted a detailed bug report, but got the response 'we know about all those already'). It's been a long time since there's been a newer test release, so I just let it sit there and mind its own business - just giving it a tap as I walk past to check it hasn't lost WiFi.At Einstein, it runs four tasks at once on the CPU cores - that's automatic, not under user control.Hm, I checked new BOINC options - very nice addition I really appreciate - now can set device name just as for Windows version. It's much more convenient to use device given name than just some auto-assigned ID... but also number of usable cores slider still in place. I set mine to 4 cores but to use all 8 cores still configurable too. What do you mean then? Not BOINC but Android OS restriction? If 8 tasks would be started simultaneously their CPU time will be much different than elapsed so only 4 CPU cores of 8 are allowed for them - that's ? |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
Pity that E@h hasn't smth like our APR to gauge app performance it seems...The APR system was added at the same time as CreditNew (and integrated with it), as a multi app replacement for DCF. Einstein made a conscious decision to reject CreditNew, and thereby rejected APR as well. They have effectively been maintaining their own private fork of the BOINC server code for the last 10 years - but they still haven't added their own private replacement for the still-running single-valued DCF - whether APR or anything else. We live with it. I see, thanks! So, where few apps possible old testing methods by relative time comparison between platforms could be used. SETI apps news We're not gonna fight them. We're gonna transcend them. |
|
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14654 Credit: 200,643,578 RAC: 874
|
So, where few apps possible old testing methods by relative time comparison between platforms could be used.Absolutely. In general, each individual app generates very stable and consistent runtimes for its own class of work - though I haven't checked the recent situation - so even simple scraping of historical task times from the website would do. There's no direct equivalent of the Angle Range variability here. |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
So, where few apps possible old testing methods by relative time comparison between platforms could be used.Absolutely. In general, each individual app generates very stable and consistent runtimes for its own class of work - though I haven't checked the recent situation - so even simple scraping of historical task times from the website would do. There's no direct equivalent of the Angle Range variability here. It simplifies things indeed :) BTW, maybe you could know? Are there sources for binary searches available for E@h? In old days I recall having some E@h app GPU sources (where was fight with iGPU app flavour those days) but most probably they are quite outdated already and even don't recall what search was that.... SETI apps news We're not gonna fight them. We're gonna transcend them. |
|
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14654 Credit: 200,643,578 RAC: 874
|
BTW, maybe you could know? Are there sources for binary searches available for E@h?The link is in the footer of each of their web pages: APPLICATION SOURCE CODE AND LICENSE |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
BTW, maybe you could know? Are there sources for binary searches available for E@h?The link is in the footer of each of their web pages: Unfortunately link broken there (that page I found of course ;) ) @ PAGE NOT FOUND The requested page could not be found. @ Though I didn't touch GW link so far. Maybe there are all sources now... Will look... EDIT: nope, htis one https://wiki.ligo.org/Computing/LALSuite strongly looks like GW-search only. So the question remains - sources for binary pulsar searches.... The one of those app is the single available for Android and another causes issues on my hardware. GW is too big to fit into small smartphone ;) :)))) SETI apps news We're not gonna fight them. We're gonna transcend them. |
Sirius B  Send message Joined: 26 Dec 00 Posts: 24881 Credit: 3,081,182 RAC: 7 
|
I thought I'd hit Emerald this morning. :-( It could have teased a bit closer. :-) 0:364:23:02:30 If they don't go into pending, should hit it on tonight's stats. |
|
Sirius B Send message Joined: 26 Dec 00 Posts: 24881 Credit: 3,081,182 RAC: 7
|
No problem. I've got some serious thinking to do. Thought I would be back at work on reduced hours, but back on furlough. Still on 100% pay for this month. However... ...with the furlough scheme extended to March, wondering how long it will stay at 100%. Reduced to 80% will hurt but liveable. Reduced to 70% will seriously hurt which will certainly force a budget re-think. |
Keith Myers   Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873 
|
Unless you grabbed a copy of the BRP source code back when the link and location was good, you are out of luck now. I have the BRP source compiled as a gpu app for my Jetson Nano from a developer at Einstein. Gaurav Khanna So he must have had the BRP source code at some time. Maybe try contacting him and seeing if he still does and is willing to share. He hasn't responded to any of my PM's in quite a while after the initial flurry of PM's in figuring out how to get his app working on my Nano. It works very well though and I can't complain. https://einsteinathome.org/host/12775352 Seti@Home classic workunits:20,676 CPU time:74,226 hours   A proud member of the OFA (Old Farts Association) |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
Unless you grabbed a copy of the BRP source code back when the link and location was good, you are out of luck now. I have the BRP source compiled as a gpu app for my Jetson Nano from a developer at Einstein. Thanks! I'll try that route when restore dev environment at least somewhere EDIT: Also I'll try to write web-admin about broken link. If it's under GPL perhaps it should be there... SETI apps news We're not gonna fight them. We're gonna transcend them. |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
So, where few apps possible old testing methods by relative time comparison between platforms could be used.Absolutely. In general, each individual app generates very stable and consistent runtimes for its own class of work - though I haven't checked the recent situation - so even simple scraping of historical task times from the website would do. There's no direct equivalent of the Angle Range variability here. And seems that GW tasks broke that rule One complication in my usual advice to "test, test, test", is that unlike the current GR GPU tasks, the current GW GPU tasks vary quite a lot in how long they take to complete because of internal task characteristics. https://einsteinathome.org/content/vega56-much-more-efficient-fgrp-gw-o2-md SETI apps news We're not gonna fight them. We're gonna transcend them. |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
And it was right move! The link is fixed already! Very fast! SETI apps news We're not gonna fight them. We're gonna transcend them. |
|
Keith Myers Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873
|
Great news Raistmer. Are you going to play with the BRP code to gain any insights into the Einstein crunching algorithm? I was always curious why they never bothered to make a CUDA or ATI OpenCL app when they pushed one out for Intel iGPU's. Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
Great news Raistmer. Are you going to play with the BRP code to gain any insights into the Einstein crunching algorithm? Definitely I would like to study sources as time allows. I was always curious why they never bothered to make a CUDA or ATI OpenCL app when they pushed one out for Intel iGPU's. Maybe because Arecibo data array is limited so they decided to improve other stuff... So there are Gamma-ray search sources somewhere, third one... Well, think radio ones is enough for now anyway. SETI apps news We're not gonna fight them. We're gonna transcend them. |
Stephen "Heretic"    Send message Joined: 20 Sep 12 Posts: 5557 Credit: 192,787,363 RAC: 628 
|
. . Well the advent of summer temps here have hit my rigs hard. On Monday it reached the mid 40's {C} in the equipment room and one machine shut itself down due to the heat and 3 others were way overtemp despite all attempts to cool them so I shut them down manually. When things cooled down I tried restarting them and one (my best producer) is now taking about 1 minute longer per task which says to me it has suffered some form of degradation, while my middle performance machine will not restart, it is giving Incorrect (invalid?) licence as an error. Not being a Linux guru I am unable to restore from that. It is actually only middle spring here and true summer will be much hotter so I have decided to give up on being a good guy and pull the plug on the project before it costs me too much. I am looking forward to the savings in my power bills (shrug). It has been a lot of fun but all things have an ending ... Stephen |
|
Keith Myers Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873
|
I can't for the life of me figure out what kind of "license" is expired in Linux. It's FOSS in the first place. No licenses required. Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
|
ML1 Send message Joined: 25 Nov 01 Posts: 20406 Credit: 7,508,002 RAC: 20
 |
... while my middle performance machine will not restart, it is giving Incorrect (invalid?) licence as an error. Not being a Linux guru I am unable to restore from that... That sounds very strange... There are no 'Linux' licenses to stop you booting... Is that machine an ex-Windows machine and you are inadvertently trying to boot into a Windows bootup? Or is there some BIOS/EFI silliness and confusion? Or is there some proprietary RAID being used that is dongled by some license? Or is the motherboard battery flat and you have a totally scrambled set of BIOS/EFI settings? Or... Any further clues as to when you get the message and the exact wording?... Clean out the dust?? Good luck? Happy cooler crunchin'! Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) |
|
Stephen "Heretic" Send message Joined: 20 Sep 12 Posts: 5557 Credit: 192,787,363 RAC: 628
|
I can't for the life of me figure out what kind of "license" is expired in Linux. It's FOSS in the first place. No licenses required. . . Yep that confuses me too. I can boot the machine from another drive, but not from the linux drive with BOINC on it. I could set up BOINC on the other drive but I don't want to lose the history this machine has created ... . . When I get time I might strip the machine and check things out a little more ... Stephen |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.