Not a good excuse. Sorry.

If it is such a burden for him, then shut this project down...

---

It's not a project, it's a test facility. The clue is under your name: "Volunteer tester". It's entirely up to you whether you wish to continue testing.

Nice one Richard was thinking along same lines.

---

A computer program will always do what you tell it to do, but rarely what you want it to do.

Happy?

Nope. Not that it's about my happiness anyway.

If there is some greater issue taking up the admin's time, fine. No problem with that. Just say so plainly, for all to know. This is what the front page of projects is for. It would take less than 60 seconds to do so.

---

Dublin, California
Team: SETI.USA

How many times do you have to be told, its a test facility. And take a look around, and ask yourself,

"What is been tested, at this moment in time, that is being controlled by the Seti staff?"

It would take less than 60 seconds to do so.

Actually front page messages are a PITA to put in place, blame DA as per ....

---

How many times do you have to be told, its a test facility. And take a look around, and ask yourself,

"What is been tested, at this moment in time, that is being controlled by the Seti staff?"

Just currently - nothing; so feel free ( IMO ) to set project to NNT. Just as I did for own hosts.

But we have something pending for testing in near future.

1) Update in OpenCL AstroPulse plan class. Issue that prevents testing from start - AP database broke again, more possible info in News section.

2) Update in OpenCL MB. In development/alpha for now, ETA ~month-two months, but perhaps not earlier than updated OpenCL AP will go on main.

3) MultiBeam v8 - AFAIK in development by Eric, guessed ETA few months (IMHO, hardly earlier).

4) Updated MultiBeam app for Android 5
5) Updated MultiBeam app for Android x86
Both these 2 items heavily rely on volunteer porting team, if any. So no foreseeable ETA.

There are parts that require Berkeley's part of team participation (I don't think that word "control" is appropriate in this context at all, control as the ability to switch server off or what type of "control" did you mean??) but some can receive good speedup from volunteers too.
Ask your friends (and yourself ;) ), maybe there are Android-oriented programmers around who could help (for example)...

---

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

How many times do you have to be told, its a test facility. And take a look around, and ask yourself,

"What is been tested, at this moment in time, that is being controlled by the Seti staff?"

Just currently - nothing; so feel free ( IMO ) to set project to NNT. Just as I did for own hosts.

But we have something pending for testing in near future.

1) Update in OpenCL AstroPulse plan class. Issue that prevents testing from start - AP database broke again, more possible info in News section.

2) Update in OpenCL MB. In development/alpha for now, ETA ~month-two months, but perhaps not earlier than updated OpenCL AP will go on main.

3) MultiBeam v8 - AFAIK in development by Eric, guessed ETA few months (IMHO, hardly earlier).

4) Updated MultiBeam app for Android 5
5) Updated MultiBeam app for Android x86
Both these 2 items heavily rely on volunteer porting team, if any. So no foreseeable ETA.

There are parts that require Berkeley's part of team participation (I don't think that word "control" is appropriate in this context at all, control as the ability to switch server off or what type of "control" did you mean??) but some can receive good speedup from volunteers too.
Ask your friends (and yourself ;) ), maybe there are Android-oriented programmers around who could help (for example)...

other issues that could stand being looked at in Beta:

Multi-beam for Intel GPU, (I've never gotten a WU for this...)

AP for Intel GPU (I haven't gotten a WU since November, 2014) (yes, I know about the AP woes - but this outage started BEFORE the AP database woes did...

why some "tapes" produce nothing but errors... with
"Error on launch (ac_reducePartial<<<grid3, block3,blksize*sizeof(float3)>>>( (float *)dev_AutoCorrIn, dev_ac_partials )), file c:/[Projects]/__Sources/sah_v7_opt/Xbranch/client/cuda/cudaAcc_autocorr.cu, line 200: invalid configuration argument"

and (of course...) fixing the stats export...

BTW, Eric has been away from Berkeley for a while...

RE: test facility: how do we know that it's working, if we don't get stats?

---

Multi-beam for Intel GPU, (I've never gotten a WU for this...)

Create thread about that and post BOINC logs where can be seens request didn't honored.

AP for Intel GPU (I haven't gotten a WU since November, 2014) (yes, I know about the AP woes - but this outage started BEFORE the AP database woes did...

the same.

why some "tapes" produce nothing but errors... with
"Error on launch (ac_reducePartial<<<grid3, block3,blksize*sizeof(float3)>>>( (float *)dev_AutoCorrIn, dev_ac_partials )), file c:/[Projects]/__Sources/sah_v7_opt/Xbranch/client/cuda/cudaAcc_autocorr.cu, line 200: invalid configuration argument"

search these and SETI main forums, it was explained many times already. In short - broken WU config in part of splitted tasks.

RE: test facility: how do we know that it's working, if we don't get stats?

By checking completed results perhaps. And completion times. Stats don't needed for current test goals. It's only for testers eye candy right now.

---

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

RE: test facility: how do we know that it's working, if we don't get stats?

Can't you get what you need from the stats right here on the Beta site?

---

David
signature sent back to alpha testing

RE: test facility: how do we know that it's working, if we don't get stats?

Can't you get what you need from the stats right here on the Beta site?

Logic would say yes. Given that the external stat sites are not a requirement of the BOINC server operations. In that respect generating the external data dump is also not a a required component of the server(s).

RE: test facility: how do we know that it's working, if we don't get stats?

Can't you get what you need from the stats right here on the Beta site?

Logic would say yes. Given that the external stat sites are not a requirement of the BOINC server operations. In that respect generating the external data dump is also not a a required component of the server(s).

On the other hand, Matt has been back working full time for about three months now, so perhaps he could have fixed it. I can't help thinking it will only take five minutes to fix.

---

David
signature sent back to alpha testing

Maybe, if everyone's working so hard on everything else, fixing the stats export could be a relaxing break...

Current stats show a difference of 1.2 million for me... (as in the stats sites show around 6 million, and the Beta site shows 7.2 million...)

As can be seen under my avatar, I've been with Beta since 2007 - I think I know if (and when...) I'm requesting Intel tasks (I am, daily since before the big outage...)

---

Just passed the 300k difference between actual score and reported.

---

Stats export is stopped.

I do not run S@H Beta tasks because I don't get credit for doing it.

Somebody will say now that 'It's for science'

For me, it's about credits. And I believe it is for several users.

And My humble bow for those who really crunch workunits for purpose of better science. I do not have real life time to do it.

But my point is: Points, credits, RAC..............

Stats export is stopped.

I do not run S@H Beta tasks because I don't get credit for doing it.

Somebody will say now that 'It's for science'

For me, it's about credits. And I believe it is for several users.

And My humble bow for those who really crunch workunits for purpose of better science. I do not have real life time to do it.

But my point is: Points, credits, RAC..............

Actually Beta is for TEST OF APPS & systems related to apps. Not directly for science.

Stats export is stopped.

I do not run S@H Beta tasks because I don't get credit for doing it.

Somebody will say now that 'It's for science'

For me, it's about credits. And I believe it is for several users.

And My humble bow for those who really crunch workunits for purpose of better science. I do not have real life time to do it.

But my point is: Points, credits, RAC..............

You still get the credit here on the Beta site. Someday, it will be fixed and all the 130,000+ credits I have racked up since export stopped will be exported and will show up on the external reporting sites.

What I want to know is if Eric is really learning anything from all the work my 3 Androids are doing lately. I've had my new phone on Beta for less than a week and it already has 11 invalids to go with its 19 valids. I don't mind doing it if it serves a purpose, but if Eric's not paying attention then what's the point?

---

David
signature sent back to alpha testing

The point depends on what involvement you wanna have/allow in testing process.
As you post on these boards (most participants never post anything) your involvement rather high to make positive use of that involvement.

Just consider 2 reports:

1) Out of 19 tasks 11 failed.

2) Out of 19 tasks being send on device XXX android version YYY 2 failed with error ZZZ, 3 with WWW and rest with VVV. App plan classes used for tasks were: ..., ... ,...


Nobody will look in your results particularly, they will be used in project-wise statistics.
But if you pareticipate actively and consider your particular results worth looking at them - then please specify what is so unusual in your particular results that they worth looking into?

---

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

The point depends on what involvement you wanna have/allow in testing process.
As you post on these boards (most participants never post anything) your involvement rather high to make positive use of that involvement.

Just consider 2 reports:

1) Out of 19 tasks 11 failed.

2) Out of 19 tasks being send on device XXX android version YYY 2 failed with error ZZZ, 3 with WWW and rest with VVV. App plan classes used for tasks were: ..., ... ,...


Nobody will look in your results particularly, they will be used in project-wise statistics.
But if you pareticipate actively and consider your particular results worth looking at them - then please specify what is so unusual in your particular results that they worth looking into?

Nothing is especially interesting (that I know of) about my particular results. What I'm asking is, is Eric or anyone else looking at those project-wide statistics right now and gaining any insight from them?

---

David
signature sent back to alpha testing

AFAIK current priority would be SETI 8.
Android SETI needs mostly volunteers for porting than for testing of already released apps.
Android devices can be successfully used on main either with already released stock or Mateuzs' opt apps.

---

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