Let's see if anyone from the team is paying attention.

The question is: Why are all the hyb_******* tasks only giving a fixed credit of 20?

This is messing with my RAC and is starting to annoy me.
Why can't you guys make them give credit more like a real task instead of some toy task?

I am into the science for you guys, but at the same time it makes me feel like I am wasting CPU time on such ridiculous amounts of credit.

Why are all the hyb_******* tasks only giving a fixed credit of 20?

I don't see any issues with credit on any of my hyb_* tasks.

hyb_ag_bench_3rj8A_SAVE_ALL_OUT_IGNORE_THE_REST_57307_268_0: 350.69 credits
hyb_af_bench_T0547_SAVE_ALL_OUT_IGNORE_THE_REST_57108_97_0: 210.24 credits
hyb_ab_bench_4ascA_SAVE_ALL_OUT_IGNORE_THE_REST_53963_276_1: 338.10 credits
hyb_ad_bench_3qc7A_SAVE_ALL_OUT_IGNORE_THE_REST_56437_170_0: 338.89 credits
hyb_ab_bench_4dweA_SAVE_ALL_OUT_IGNORE_THE_REST_53978_808_0: 310.02 credits
hyb_ac_bench_3qphA_16_SAVE_ALL_OUT_IGNORE_THE_REST_54638_86_0: 363.21 credits

Nothing of that is even near to fixed 20 credits per WU...

---

.

I see this on your sterr out for your 20 credit workunits:

WARNING! cannot get file size for default.out.gz: could not open file.
Output exists: default.out.gz Size: -1
InternalDecoyCount: 0 (GZ)
-----
0
-----
Stream information inconsistent.
Writing W_0000001


A project reset may be in order.

---

Funny, you guys have more information quicker than the team will ever have.
I don't think they even bother watching the boards anymore.

Looking at my stats I see a small climb happening in my RAC so I must have worked through whatever junk they had stored in the queue.

and as I figured...NO ONE from the "Team" reads these threads anymore.

I don't see any issues with credit on any of my hyb_* tasks.
Nothing of that is even near to fixed 20 credits per WU...

I had a lot of 20 Cr WUs (many dozens).
I report it in this message: https://boinc.bakerlab.org/rosetta/forum_thread.php?id=6055&nowrap=true#73741

Fresh WUs examples:
hyb_ag_bench_T0601_SAVE_ALL_OUT_IGNORE_THE_REST_57488_1754_0
hyb_ag_bench_T0537_SAVE_ALL_OUT_IGNORE_THE_REST_57427_2581_0
hyb_ag_bench_4dg9A_SAVE_ALL_OUT_IGNORE_THE_REST_57389_1152_0
hyb_ai_bench_3rr5A_SAVE_ALL_OUT_IGNORE_THE_REST_57977_3138_0
hyb_ag_bench_3stoA_SAVE_ALL_OUT_IGNORE_THE_REST_57327_1992_0
hyb_ag_bench_4aimA_SAVE_ALL_OUT_IGNORE_THE_REST_57376_1381_0
hyb_ag_bench_3rr5A_SAVE_ALL_OUT_IGNORE_THE_REST_57313_2495_0
hyb_ag_bench_3slkB_SAVE_ALL_OUT_IGNORE_THE_REST_57323_2116_0
hyb_ag_bench_3rdeD_SAVE_ALL_OUT_IGNORE_THE_REST_57305_2685_0
hyb_ag_bench_4abnB_SAVE_ALL_OUT_IGNORE_THE_REST_57368_1526_0
hyb_ag_bench_2yeqB_SAVE_ALL_OUT_IGNORE_THE_REST_57261_3224_0
hyb_ag_bench_3rj8A_SAVE_ALL_OUT_IGNORE_THE_REST_57307_2399_0
hyb_ag_bench_3zyyY_SAVE_ALL_OUT_IGNORE_THE_REST_57364_1631_0
hyb_ai_bench_3zyyY_SAVE_ALL_OUT_IGNORE_THE_REST_58028_1718_0
hyb_ah_bench_3rj8A_SAVE_ALL_OUT_IGNORE_THE_REST_57650_2148_0
hyb_ah_bench_3qd9D_SAVE_ALL_OUT_IGNORE_THE_REST_57633_2296_0
hyb_ai_bench_3rr5A_SAVE_ALL_OUT_IGNORE_THE_REST_57977_911_1
hyb_ah_bench_3zyyY_SAVE_ALL_OUT_IGNORE_THE_REST_57707_1335_1

Many participants also encounter an error in this series of tasks (while all the others WUs work fine).
So now I abort all WUs from this series in the BOINC job queue.

You also have those weird error messages in the std_err:

BOINC:: CPU time: 25550.8s, 14400s + 10800s[2012- 9- 8 3: 6:41:] :: BOINC
WARNING! cannot get file size for default.out.gz: could not open file.
Output exists: default.out.gz Size: -1
InternalDecoyCount: 0 (GZ)
-----
0
-----
Stream information inconsistent.
Writing W_0000001

Maybe it's because of BOINC v7? You and Greg_BE are both using v7.0.28, I'm on 6.12.34.

---

.

don't understand how Version 7 would foul up just some specific tasks.
i will downgrade and see what happens.

what is also weird is the cpu time goes past my cutoff time by an additional 14k seconds.

don't understand how Version 7 would foul up just some specific tasks.

I don't understand that too, but since according to other threads v7 is known to be responsible for some errors on rosetta, downgrade to v6 would be the first thing I'd try (unless I'd need to stay with v7 for some other reason).

---

.

don't understand how Version 7 would foul up just some specific tasks.

I don't understand that too, but since according to other threads v7 is known to be responsible for some errors on rosetta, downgrade to v6 would be the first thing I'd try (unless I'd need to stay with v7 for some other reason).

I did 2 things and am now able to crunch for Rosie again...1st I downgraded to the version 6 series of Boinc, 2nd I STOPPED all gpu crunching on those machines that ALSO crunch for Rosie. I have had 1 or maybe even 2 funky units since then but ALL the rest are finishing just fine! Knocking on wood here!!!!!!!!!!!!!!

don't understand how Version 7 would foul up just some specific tasks.

I don't understand that too, but since according to other threads v7 is known to be responsible for some errors on rosetta, downgrade to v6 would be the first thing I'd try (unless I'd need to stay with v7 for some other reason).

I did 2 things and am now able to crunch for Rosie again...1st I downgraded to the version 6 series of Boinc, 2nd I STOPPED all gpu crunching on those machines that ALSO crunch for Rosie. I have had 1 or maybe even 2 funky units since then but ALL the rest are finishing just fine! Knocking on wood here!!!!!!!!!!!!!!

downgraded to V6 as well and it handled a hyb task in the normal fashion.
I thought Ralph was supposed to see these errors before they stuck the tasks on Rosie? Guess they did not do that or there are no V7 users on Ralph.
In any case, we solve the problems on our own with no support from the team.
In the old days we had them or modsense actually pay attention. No I would call this project a "unsupported" project in terms of any feedback from the team. I posted a error message once in POEM and in Einstein and had feedback from some higher up folks. Not to mention their news updates either on the page or via notices here in BOINC.

In any case, thanks for the info guys...very useful.

don't understand how Version 7 would foul up just some specific tasks.

I don't understand that too, but since according to other threads v7 is known to be responsible for some errors on rosetta, downgrade to v6 would be the first thing I'd try (unless I'd need to stay with v7 for some other reason).

I did 2 things and am now able to crunch for Rosie again...1st I downgraded to the version 6 series of Boinc, 2nd I STOPPED all gpu crunching on those machines that ALSO crunch for Rosie. I have had 1 or maybe even 2 funky units since then but ALL the rest are finishing just fine! Knocking on wood here!!!!!!!!!!!!!!

downgraded to V6 as well and it handled a hyb task in the normal fashion.
I thought Ralph was supposed to see these errors before they stuck the tasks on Rosie? Guess they did not do that or there are no V7 users on Ralph.
In any case, we solve the problems on our own with no support from the team.
In the old days we had them or modsense actually pay attention. No I would call this project a "unsupported" project in terms of any feedback from the team. I posted a error message once in POEM and in Einstein and had feedback from some higher up folks. Not to mention their news updates either on the page or via notices here in BOINC.

In any case, thanks for the info guys...very useful.

Awhile back they said they HAD tested it on Ralph and everything was 'fine', but not to use ver 7 as they do NOT support it. My question is then WHY does Boinc KEEP telling me to update?!!!! They NEED to TALK to each other!!!! Rosie and Berkeley I mean!!!

don't understand how Version 7 would foul up just some specific tasks.

I don't understand that too, but since according to other threads v7 is known to be responsible for some errors on rosetta, downgrade to v6 would be the first thing I'd try (unless I'd need to stay with v7 for some other reason).

I did 2 things and am now able to crunch for Rosie again...1st I downgraded to the version 6 series of Boinc, 2nd I STOPPED all gpu crunching on those machines that ALSO crunch for Rosie. I have had 1 or maybe even 2 funky units since then but ALL the rest are finishing just fine! Knocking on wood here!!!!!!!!!!!!!!

downgraded to V6 as well and it handled a hyb task in the normal fashion.
I thought Ralph was supposed to see these errors before they stuck the tasks on Rosie? Guess they did not do that or there are no V7 users on Ralph.
In any case, we solve the problems on our own with no support from the team.
In the old days we had them or modsense actually pay attention. No I would call this project a "unsupported" project in terms of any feedback from the team. I posted a error message once in POEM and in Einstein and had feedback from some higher up folks. Not to mention their news updates either on the page or via notices here in BOINC.

In any case, thanks for the info guys...very useful.

Awhile back they said they HAD tested it on Ralph and everything was 'fine', but not to use ver 7 as they do NOT support it. My question is then WHY does Boinc KEEP telling me to update?!!!! They NEED to TALK to each other!!!! Rosie and Berkeley I mean!!!

Well why can't they write a code that blocks those tasks from being loaded onto computers running V7? I thought there was something that BOINC told Rosie as to the type of computer I have and how many tasks I can handle, etc.

don't understand how Version 7 would foul up just some specific tasks.

I don't understand that too, but since according to other threads v7 is known to be responsible for some errors on rosetta, downgrade to v6 would be the first thing I'd try (unless I'd need to stay with v7 for some other reason).

I did 2 things and am now able to crunch for Rosie again...1st I downgraded to the version 6 series of Boinc, 2nd I STOPPED all gpu crunching on those machines that ALSO crunch for Rosie. I have had 1 or maybe even 2 funky units since then but ALL the rest are finishing just fine! Knocking on wood here!!!!!!!!!!!!!!

downgraded to V6 as well and it handled a hyb task in the normal fashion.
I thought Ralph was supposed to see these errors before they stuck the tasks on Rosie? Guess they did not do that or there are no V7 users on Ralph.
In any case, we solve the problems on our own with no support from the team.
In the old days we had them or modsense actually pay attention. No I would call this project a "unsupported" project in terms of any feedback from the team. I posted a error message once in POEM and in Einstein and had feedback from some higher up folks. Not to mention their news updates either on the page or via notices here in BOINC.

In any case, thanks for the info guys...very useful.

Awhile back they said they HAD tested it on Ralph and everything was 'fine', but not to use ver 7 as they do NOT support it. My question is then WHY does Boinc KEEP telling me to update?!!!! They NEED to TALK to each other!!!! Rosie and Berkeley I mean!!!

Well why can't they write a code that blocks those tasks from being loaded onto computers running V7? I thought there was something that BOINC told Rosie as to the type of computer I have and how many tasks I can handle, etc.

They probably could but Rosie has said several times that they are happy the way things are now and are not making any changes! If it crashes their response is just downgrade, if you don't chose to downgrade they don't see it as their issue. If you can't crunch they feel they have enough other crunchers to make up the difference.

Right now I have 2 pc's on here, as soon as I can figure out how to put a gpu in each I am outta here too! In fact my quad core will be set to no new tasks today! My dual core may take a bit but it won't stay long either. The linux machine is off line and has been for awhile now. I in fact just aborted all cached tasks on my quad core here at Rosie and attached to a different Project. I then detached it from Rosie. I JUST did the same thing to my dual core! So I now have NO pc's attached here anymore!

If you are doing it for the points, you are doing it for the wrong reasons.

Let's see if anyone from the team is paying attention.

The question is: Why are all the hyb_******* tasks only giving a fixed credit of 20?

This is messing with my RAC and is starting to annoy me.
Why can't you guys make them give credit more like a real task instead of some toy task?

I am into the science for you guys, but at the same time it makes me feel like I am wasting CPU time on such ridiculous amounts of credit.

If you are doing it for the points, you are doing it for the wrong reasons.

If he was doing it only for the points, he probably wasn't here, there are projects paying a lot more.

The credits are a measure for the amount of job done for the project, watching them means also watching that my machines are doing useful job for the project.

---

.

If you are doing it for the points, you are doing it for the wrong reasons.

If he was doing it only for the points, he probably wasn't here, there are projects paying a lot more.

The credits are a measure for the amount of job done for the project, watching them means also watching that my machines are doing useful job for the project.

If I am going to burn electricity all the time for this project, I just want to know that my machine is processing things correctly. Its a mental thing the points. If my machine is going to crunch for x hrs or whatever then I mentally want to see that I am getting the right amount of credit for it. And the above comment about not wanting to do anything about problems like I have found here with version 7 and not putting out a notice or a thread about this problem or even bothering to reply to something about this problem shows that the team really does not give a damn.

Of course if I leave then there will be plenty others to take my place. Key thing is, why waste my resources on a malfunctioning non news no reply project? It is really bizarre how the guys at Einstein can put notices on their page when they find something interesting or have a task malfunction badly or a IT problem and also send out notifications when they have made a big discovery and this project does none of the above. never has, never will. Pretty sad really.

I understand where you are coming from, after all I am a long time folder myself.

I believe you are missing the point, a project such as Rosetta is a massive under taking and I am almost 100% sure they are under staffed.

Its not that they dont give a damn, its they dont have enough time in the day to get everything done. I run a small business and this is the case for me everyday, just imagine how the staff feel at Rosetta.

What I am essentially wanting to say, is its not as clear cut as "they are not responding so they dont care."

At the end of the day its your choice whether you stay or go, but you should put yourself in their shoes for a few days then you will understand (even if only by imagination).

If you are doing it for the points, you are doing it for the wrong reasons.

If he was doing it only for the points, he probably wasn't here, there are projects paying a lot more.

The credits are a measure for the amount of job done for the project, watching them means also watching that my machines are doing useful job for the project.

If I am going to burn electricity all the time for this project, I just want to know that my machine is processing things correctly. Its a mental thing the points. If my machine is going to crunch for x hrs or whatever then I mentally want to see that I am getting the right amount of credit for it. And the above comment about not wanting to do anything about problems like I have found here with version 7 and not putting out a notice or a thread about this problem or even bothering to reply to something about this problem shows that the team really does not give a damn.

Of course if I leave then there will be plenty others to take my place. Key thing is, why waste my resources on a malfunctioning non news no reply project? It is really bizarre how the guys at Einstein can put notices on their page when they find something interesting or have a task malfunction badly or a IT problem and also send out notifications when they have made a big discovery and this project does none of the above. never has, never will. Pretty sad really.

I understand where you are coming from, after all I am a long time folder myself.

I believe you are missing the point, a project such as Rosetta is a massive under taking and I am almost 100% sure they are under staffed.

Its not that they dont give a damn, its they dont have enough time in the day to get everything done. I run a small business and this is the case for me everyday, just imagine how the staff feel at Rosetta.

What I am essentially wanting to say, is its not as clear cut as "they are not responding so they dont care."

At the end of the day its your choice whether you stay or go, but you should put yourself in their shoes for a few days then you will understand (even if only by imagination).

I think the problem is Rosetta gets grants from places like the Gates Foundation, as in Bill and Melinda Gates, yet STILL can't hire enough people to make the Project work? WHY would Bill and Melinda Gates give money to a half a$$ Project like that? What a waste of their hard earned money and resources! Rosetta is NOT even looking for a cure for ANYTHING, they are trying to find ways of improving the PROCESS of using pc's to find cures. Then when the very people that helped Rosetta win the grant need some help and feedback and basically get told 'love it or leave it, we don't care!', it is BAD for Rosetta and its future as a viable Distributed Computing Project. ALL of us here have put in some significant time and resources to Rosetta, to abandon us after all that effort can cause each of us to say bad things about Rosetta as we move on to other Projects! For Rosetta to abandon the Boinc upgrade process is dumb, stupid and a VERY bad move! For those that also crunch the Project Albert http://albert.phys.uwm.edu/index.php it is IMPOSSIBLE to do it anymore! Albert REQUIRES Boinc 7.027 OR HIGHER, yet Rosetta does NOT support version 7 of Boinc!!

I found Dr Bakers email and also put a message out on the boards a long time ago saying why don't they hire someone from communications pool of students to come and work for the project and handle their PR and chat boards. There used to be a very skilled Mod here, but he left. Tech issues are handled by a permanent person within the team, but he does not post anything at all, even if the system "burps" badly. that's the thing here, you don't know if rosie will be offline for a few hour or a few days or a week when they have system troubles. but I think now they have a good tech backbone to work with so it will just be software bugs. The science team of Grad students working with Dr Baker on their ideas used to post stuff here telling about what they were studying with rosie and what they hoped to find. You don't see that anymore. Plus when the person that wrote the code for the stuff they were looking for read about a bug, they responded or did something about it. Now they don't do either one.

I understand where you are coming from, after all I am a long time folder myself.

I believe you are missing the point, a project such as Rosetta is a massive under taking and I am almost 100% sure they are under staffed.

Its not that they dont give a damn, its they dont have enough time in the day to get everything done. I run a small business and this is the case for me everyday, just imagine how the staff feel at Rosetta.

What I am essentially wanting to say, is its not as clear cut as "they are not responding so they dont care."

At the end of the day its your choice whether you stay or go, but you should put yourself in their shoes for a few days then you will understand (even if only by imagination).

If you are doing it for the points, you are doing it for the wrong reasons.

If he was doing it only for the points, he probably wasn't here, there are projects paying a lot more.

The credits are a measure for the amount of job done for the project, watching them means also watching that my machines are doing useful job for the project.

If I am going to burn electricity all the time for this project, I just want to know that my machine is processing things correctly. Its a mental thing the points. If my machine is going to crunch for x hrs or whatever then I mentally want to see that I am getting the right amount of credit for it. And the above comment about not wanting to do anything about problems like I have found here with version 7 and not putting out a notice or a thread about this problem or even bothering to reply to something about this problem shows that the team really does not give a damn.

Of course if I leave then there will be plenty others to take my place. Key thing is, why waste my resources on a malfunctioning non news no reply project? It is really bizarre how the guys at Einstein can put notices on their page when they find something interesting or have a task malfunction badly or a IT problem and also send out notifications when they have made a big discovery and this project does none of the above. never has, never will. Pretty sad really.