Cool, thanks. So far SoG is faster on my Mac, couple of inconclusives. Not sure if it will end up being faster than what I'm running on main as I've only seen one AR (~0.41) so far so we shall see.

Chris

I will also note that I've seen one odd behavior. Boinc reports the wu is still running for about 30s after it says 100% complete. Normally I'd just mark that up to boinc but I've noticed that the app is actually unloaded from memory at this point, so it's not still running, but the time does seem to be reported as computation time anyway.

Chris

Last night when I had some time to check what was going on I noticed that all of the 8.09 Sog's on my system were starting out just like the 8.07 full core for about 15-20 seconds then minimal till the end but the 8.09's were not using the full core at the end and when the elapsed time reached the estimated time the est. time suddenly spiked way up some of them jumped to 400+ days but within less than a minute dropped and the wu finished.

---

ME AND MY BOY LOOKING FOR E.T.

I will also note that I've seen one odd behavior. Boinc reports the wu is still running for about 30s after it says 100% complete. Normally I'd just mark that up to boinc but I've noticed that the app is actually unloaded from memory at this point, so it's not still running, but the time does seem to be reported as computation time anyway.

Chris

lINK TO SUCH RESULT?

---

News about SETI opt app releases: https://twitter.com/Raistmer

I will also note that I've seen one odd behavior. Boinc reports the wu is still running for about 30s after it says 100% complete. Normally I'd just mark that up to boinc but I've noticed that the app is actually unloaded from memory at this point, so it's not still running, but the time does seem to be reported as computation time anyway.

Chris

lINK TO SUCH RESULT?

It's all of them but here are a couple...

http://setiweb.ssl.berkeley.edu/beta/result.php?resultid=23278603



http://setiweb.ssl.berkeley.edu/beta/result.php?resultid=23278651

Chris

A bit off topic, but some images here are broken.

This is in both Firefox and in Opera, could somebody fix this?

Been broken for about a month now. I just put it down to DA "fixing" the server code yet again.

emailed the crew.
I'd class it as cosmetic only, so I wouldn't hold my breath for getting fixed soon.

Thanks to the crew.

---

I will also note that I've seen one odd behavior. Boinc reports the wu is still running for about 30s after it says 100% complete. Normally I'd just mark that up to boinc but I've noticed that the app is actually unloaded from memory at this point, so it's not still running, but the time does seem to be reported as computation time anyway.

Chris

lINK TO SUCH RESULT?

It's all of them but here are a couple...

http://setiweb.ssl.berkeley.edu/beta/result.php?resultid=23278603



http://setiweb.ssl.berkeley.edu/beta/result.php?resultid=23278651

Chris

One SoG other usual and both have no issues in stderr. Perhaps BOINC thing.

---

News about SETI opt app releases: https://twitter.com/Raistmer

--

---

News about SETI opt app releases: https://twitter.com/Raistmer

http://setiweb.ssl.berkeley.edu/beta/apps.php
SETI@home v8 (AMD/ATI apps):

Windows/x86 8.09 (opencl_ati5_cat132)
Windows/x86 8.09 (opencl_ati5_nocal)
Windows/x86 8.09 (opencl_ati5_sah)
Windows/x86 8.09 (opencl_ati5_SoG)
Windows/x86 8.09 (opencl_ati5_SoG_cat132)
Windows/x86 8.09 (opencl_ati5_SoG_nocal)
Windows/x86 8.09 (opencl_atiapu_sah)
Windows/x86 8.09 (opencl_atiapu_SoG)
Windows/x86 8.09 (opencl_ati_cat132)
Windows/x86 8.09 (opencl_ati_nocal)
Windows/x86 8.09 (opencl_ati_sah)

(In green the 5 apps which my PC have downloaded until now.)

Which apps will get my FuryX VGA cards additional, all apps except the 'cat132' apps?
(Crimson 16.3 Hotfix (Beta, Mar/09) installed with v2004.6 driver in it.)

Thanks.

BTW.

The cmdline .TXT files are with:

*8.09_windows_intel__opencl_*

They shouldn't look like:

*8.09_windows_intelx86__opencl_* ?

The apps have the x86 in their names.

http://setiweb.ssl.berkeley.edu/beta/apps.php
SETI@home v8 (AMD/ATI apps):

Windows/x86 8.09 (opencl_ati5_cat132)
Windows/x86 8.09 (opencl_ati5_nocal)
Windows/x86 8.09 (opencl_ati5_sah)
Windows/x86 8.09 (opencl_ati5_SoG)
Windows/x86 8.09 (opencl_ati5_SoG_cat132)
Windows/x86 8.09 (opencl_ati5_SoG_nocal)
Windows/x86 8.09 (opencl_atiapu_sah)
Windows/x86 8.09 (opencl_atiapu_SoG)
Windows/x86 8.09 (opencl_ati_cat132)
Windows/x86 8.09 (opencl_ati_nocal)
Windows/x86 8.09 (opencl_ati_sah)

(In green the 5 apps which my PC have downloaded until now.)

Which apps will get my FuryX VGA cards additional, all apps except the 'cat132' apps?
(Crimson 16.3 Hotfix (Beta, Mar/09) installed with v2004.6 driver in it.)

Thanks.

Application details for host 76831
http://setiweb.ssl.berkeley.edu/beta/host_app_versions.php?hostid=76831

SETI@home v8 8.09 windows_intelx86 (opencl_atiapu_sah)
SETI@home v8 8.09 windows_intelx86 (opencl_atiapu_SoG)
SETI@home v8 8.09 windows_intelx86 (opencl_ati5_nocal)
SETI@home v8 8.09 windows_intelx86 (opencl_ati5_SoG_nocal)
SETI@home v8 8.09 windows_intelx86 (opencl_ati_nocal)

Downloaded to the project folder:
*opencl_ati_sah.exe
*opencl_ati5_sah.exe
*opencl_ati5_SoG.exe
*opencl_atiapu_sah.exe
*opencl_atiapu_SoG.exe

Just two apps are 'identically' - is this all OK or is there something wrong?

---

Just two apps are 'identically' - is this all OK or is there something wrong?

Did you check their MD5 sums to say they are identical ones?

---

News about SETI opt app releases: https://twitter.com/Raistmer

Application details for host 76831
http://setiweb.ssl.berkeley.edu/beta/host_app_versions.php?hostid=76831

SETI@home v8 8.09 windows_intelx86 (opencl_atiapu_sah)
SETI@home v8 8.09 windows_intelx86 (opencl_atiapu_SoG)
SETI@home v8 8.09 windows_intelx86 (opencl_ati5_nocal)
SETI@home v8 8.09 windows_intelx86 (opencl_ati5_SoG_nocal)
SETI@home v8 8.09 windows_intelx86 (opencl_ati_nocal)

Downloaded to the project folder:
*opencl_ati_sah.exe
*opencl_ati5_sah.exe
*opencl_ati5_SoG.exe
*opencl_atiapu_sah.exe
*opencl_atiapu_SoG.exe

Just two apps are 'identically' - is this all OK or is there something wrong?

With 'identically' I meant the 'plan_class'.

The host apps details overview show 5 apps, in the project folder are 5 apps.
But just two 'plan_class's are similar (shown in my last/above message).


On the apps overview are shown AMD/ATI apps:
Windows/x86 8.09 (opencl_ati5_cat132)
Windows/x86 8.09 (opencl_ati5_nocal)
Windows/x86 8.09 (opencl_ati5_sah)
Windows/x86 8.09 (opencl_ati5_SoG)
Windows/x86 8.09 (opencl_ati5_SoG_cat132)
Windows/x86 8.09 (opencl_ati5_SoG_nocal)
Windows/x86 8.09 (opencl_atiapu_sah)
Windows/x86 8.09 (opencl_atiapu_SoG)
Windows/x86 8.09 (opencl_ati_cat132)
Windows/x86 8.09 (opencl_ati_nocal)
Windows/x86 8.09 (opencl_ati_sah)

Which, how much apps (shown on the host apps details overview & project folder?) should get my FuryX VGA cards?

At the host apps details overview & in the project folder should be 11 apps?

---

Where you see identical plan classes if they supply different binaries??

---

News about SETI opt app releases: https://twitter.com/Raistmer

Getting a bit of strangeness on status reporting (shown in Boinctasks screen format).
About 15 minutes into task, Progress percentage will jump from~50% to .001 percent complete. At same time, time remaining will change from 1-2 minutes remaining to over 20,000 hours remaining.
After another minute or so, task will show complete and upload normally. Don't see anything in Stderr raising an issue, and tasks seem to validate OK, but something in the last minute or two of processing blows away the percentage and time statuses.

Computer Project Progress% Application CPU% Use
Time left Elapsed time Deadline Name Work unit
77530 SETI@home Beta Test 0.001 8.09 SETI@home v8 (opencl_nvidia_SoG) 8.50 0.04C + 0.5NV (d0) 21339:07:19 00:14:50 (00:01:15) 487:24:39 24no10ab.19773.6611.6.40.126_0 8401455
77530 SETI@home Beta Test 0.001 8.09 SETI@home v8 (opencl_nvidia_SoG) 8.51 0.04C + 0.5NV (d1) 21288:37:40 00:14:47 (00:01:15) 487:24:39 24no10ab.19773.6611.6.40.189_0 8401518
77530 SETI@home Beta Test 0.001 8.09 SETI@home v8 (opencl_nvidia_SoG) 7.98 0.04C + 0.5NV (d0) 21288:15:14 00:14:47 (00:01:10) 487:36:37 24no10ab.19773.6611.6.40.180_1 8401509
77530 SETI@home Beta Test 0.001 8.09 SETI@home v8 (opencl_nvidia_SoG) 8.82 0.04C + 0.5NV (d1) 20821:50:21 00:14:28 (00:01:16) 487:39:54 24no10ab.19773.6611.6.40.159_2 8401488
77530 SETI@home Beta Test 0.001 8.09 SETI@home v8 (opencl_nvidia_SoG) 8.30 0.04C + 0.5NV (d0) 20858:07:12 00:14:29 (00:01:12) 487:40:12 24no10ab.19773.6611.6.40.138_0 8401467
77530 SETI@home Beta Test 0.001 8.09 SETI@home v8 (opencl_nvidia_SoG) 8.43 0.04C + 0.5NV (d0) 20443:41:44 00:14:12 (00:01:11) 487:21:57 24no10ab.19773.6611.6.40.160_0 8401489
77530 SETI@home Beta Test 0.001 8.09 SETI@home v8 (opencl_nvidia_SoG) 8.29 0.04C + 0.5NV (d1) 20030:23:35 00:13:55 (00:01:09) 487:17:31 24no10ab.19773.6611.6.40.75_0 8401404

Sorry I can't get the listing format better...

[Edit] Negative impact of this is that once Time Left exceeds Deadline time, BOINC sets subsequent tasks to High Priority. This may trigger alarms or other changes in processing or priority, depending on configuration of client or other downstream processing aids such as BoincTasks.

---

If I can help out by testing something, please let me know.
Available hardware and software is listed in my profile here.

It was explained few times already please use search

---

News about SETI opt app releases: https://twitter.com/Raistmer

It was explained few times already please use search

OK, I searched and found:

Non-linear progress was ALWAYS, even with CPU Multibeam apps. It's inherent property of types of data processing distribution inside processing chain. SoG just makes this difference more visible.

But what Jimbocous described isn't just non-linear, I'd call it broken - it's the behaviour I'd expect if <fraction_done> never made it through the BOINC API reporting chain into boinc_task_state.xml.

Jimbocous, which version of BOINC are you running? I can look it up myself, but just wanting to be sure whether it was one of you v7.6.9 hosts, or the v7.6.22 one?

I'll try running a few myself, and see what I can turn up if I poke around a bit.

Edit - I acknowledge that all the tasks Jimbo has listed come from the same batch

24no10ab.19773.6611.6.40.xxx

which have the same very unusual

WU true angle range is : 136.583901

but that shouldn't affect <fraction_done> over a 15 minute run. The app has to cope with the hard cases, too.

It was explained few times already please use search

OK, I searched and found:

Non-linear progress was ALWAYS, even with CPU Multibeam apps. It's inherent property of types of data processing distribution inside processing chain. SoG just makes this difference more visible.

But what Jimbocous described isn't just non-linear, I'd call it broken - it's the behaviour I'd expect if <fraction_done> never made it through the BOINC API reporting chain into boinc_task_state.xml.

Jimbocous, which version of BOINC are you running? I can look it up myself, but just wanting to be sure whether it was one of you v7.6.9 hosts, or the v7.6.22 one?

I'll try running a few myself, and see what I can turn up if I poke around a bit.

Edit - I acknowledge that all the tasks Jimbo has listed come from the same batch

24no10ab.19773.6611.6.40.xxx

which have the same very unusual

WU true angle range is : 136.583901

but that shouldn't affect <fraction_done> over a 15 minute run. The app has to cope with the hard cases, too.

On SoG, all WU's with extremely high ARs behave exactly like that. Something I pointed out on main from the very beginning of SoG. They start out, counting up to something like 45% done, and then falls back down to 0.001% or so, where they sit, until they all of a sudden finish, after a couple of minutes, poof, 100%. They go to high priority, when they fall back to 0.001%, and time remaining can go up as high as 600 days :-)

Doesn't bother me much though, since it's more of a cosmetic issue...

Doesn't bother me much though, since it's more of a cosmetic issue...

It may only be cosmetic during testing, but it's the sort of thing that annoys volunteers: if a successor to this build is ever to be considered for stock use, it's one of the things that needs to be on the ToDo list to fix, somewhere down the line (agreed it isn't top of that list at this stage).

Doesn't bother me much though, since it's more of a cosmetic issue...

It may only be cosmetic during testing, but it's the sort of thing that annoys volunteers: if a successor to this build is ever to be considered for stock use, it's one of the things that needs to be on the ToDo list to fix, somewhere down the line (agreed it isn't top of that list at this stage).

More info to add. It's Boinc Tasks that counts up to around 45% before falling back to 0.001-0.002%. Boinc Manager doesn't count up at all, it sits at 0.001-0.002% all the time on these extremely high AR's, until they all of a sudden finish with poof 100%. However the remaining time goes up on Boinc Manager also, to more than a year or so.

I didn't notice that until now, because I do not use Boinc Manager at all, only Boinc Tasks.