Advanced search

Message boards : Wish list : So why dont I start a NEW thread

Author Message
Mr Stupid
Send message
Joined: 7 Dec 10
Posts: 2
Credit: 0
RAC: 0
Message 9211 - Posted: 7 Dec 2010, 19:34:03 UTC
Last modified: 7 Dec 2010, 19:37:03 UTC

I cant be bothered to read anything, so I shall start all over at the beginning.


I cannot edit my profile, I wanted a silly avatar and some other rubbish. I feel angry that I cannot do this so please fix it for ME.

Some of the work units come up with "file not found" errors, I feel this is a personal insult that you are rejecting ME.

You keep sending out far too many work units, once again this is an insult to ME because I cannot process them in time.

The workunits use up about 800 megabytes of memory so when I run them on my 8 core machine with 2GB memory it crashes.




Please fix EVERYTHING for ME, NOW, or I will go away and sit in the corner sulking.

Thank you.

.clair.
Send message
Joined: 4 Nov 07
Posts: 591
Credit: 3,822,426
RAC: 5,012
Message 9216 - Posted: 7 Dec 2010, 23:01:27 UTC

Sorry to have to tell you this but at cosmo we have no contact with the admin,
if they ever read the boards they dont say so or make it known that they have.

Editing of profile - its been broken for some time now.

Download errors - we all get them, ignore them.

Work unit bombs - until you have crunched for a boinc project for a few weeks this can happen, set `no new tasks` in boinc manager for the overloaded project to give the boinc software time to sort itself out.

Memory used - we all have to live with it.

Sorry that i have no good news for you,
The problem is that cosmo is on autopilot and some of it is broken :(

Scrooby
Send message
Joined: 13 Feb 08
Posts: 77
Credit: 616,140
RAC: 0
Message 9217 - Posted: 7 Dec 2010, 23:26:05 UTC - in response to Message 9211.

I cant be bothered to read anything, so I shall start all over at the beginning.

[...]

Please fix EVERYTHING for ME, NOW, or I will go away and sit in the corner sulking.

Thank you.



Yes if you read this board you'll find all of this has been discussed for several years.

Profile Faik
Send message
Joined: 3 Oct 11
Posts: 2771
Credit: 140,700
RAC: 0
Message 9958 - Posted: 3 Feb 2012, 14:19:07 UTC - in response to Message 9216.

Hello clive G1FYE,

I hope that it is plainly clear now that the days you describe below are behind our backs! The lack of communication with the C@H administration was a genuine concern, and I hope that our efforts in the past few months have contributed, even in a small part, to bridging the gap.

Thank you for your patience. Thank you for sticking around!

Best Wishes,

Faik
---
Member of the C@H Team!


Sorry to have to tell you this but at cosmo we have no contact with the admin,
if they ever read the boards they dont say so or make it known that they have.

Editing of profile - its been broken for some time now.

Download errors - we all get them, ignore them.

Work unit bombs - until you have crunched for a boinc project for a few weeks this can happen, set `no new tasks` in boinc manager for the overloaded project to give the boinc software time to sort itself out.

Memory used - we all have to live with it.

Sorry that i have no good news for you,
The problem is that cosmo is on autopilot and some of it is broken :(

Profile cykodennis
Send message
Joined: 31 May 10
Posts: 234
Credit: 4,896,378
RAC: 0
Message 9959 - Posted: 3 Feb 2012, 15:11:51 UTC

*gives Faik a cup of coffee*

Yo man,
you are a scientist. Scientists should not do necromancy, even no thread necromancy ;-)

Profile Faik
Send message
Joined: 3 Oct 11
Posts: 2771
Credit: 140,700
RAC: 0
Message 9962 - Posted: 3 Feb 2012, 19:28:05 UTC - in response to Message 9959.

Actually that is exactly what a scientist does, especially someone who specializes in the fine science of necromantology like myself ;)

Faik
---
Member of the C@h Team!

*gives Faik a cup of coffee*

Yo man,
you are a scientist. Scientists should not do necromancy, even no thread necromancy ;-)

Michael James
Send message
Joined: 5 Jan 08
Posts: 3
Credit: 519,470
RAC: 300
Message 10003 - Posted: 7 Feb 2012, 3:50:41 UTC

Have to say I'm getting fed up with the amount of memory being hogged when the WUs get suspended, anything up tp 650mb even though my preferences set the limit much lower. Gave up this project once a while ago because of this, thought the problem might have been fixed. Probably give up again and switch to another project that's better behaved.

Profile cykodennis
Send message
Joined: 31 May 10
Posts: 234
Credit: 4,896,378
RAC: 0
Message 10008 - Posted: 7 Feb 2012, 8:20:11 UTC - in response to Message 10003.

I do not think that this is a "problem" which has to be fixed.
I just guess, that you can not describe universe with just 256 MB Ram.

Of course, thats annoying for you. I have checked your shown machine, you have a maximum of 4 GB RAM and two cores, which means that half of your memory is filled with Cosmo. Your machine needs 8-20 (?) hours per result, your system performance while daily use is not very smoothly and so, it is fully understandable when you quit.

I just want to tell you openly, i do not think that this memory thing will ever be "fixed" in Cosmo, especially in times when RAM is cheap an 64 Bit Systems become more common.

Michael James
Send message
Joined: 5 Jan 08
Posts: 3
Credit: 519,470
RAC: 300
Message 10098 - Posted: 13 Feb 2012, 8:04:23 UTC

You are missing the point, it is not acceptable for a suspended process to hog 50% of the total memory when suspended. It doesn't matter what the configuration of my machine is, the application should behave in a manner that is acceptable and conforms to the preferences I set. Impling that I should get a 64 bit machine and more ram just to voluntary support a badly behaved application simply shows what little respect or appreciation is given to those who freely give up time on their personal computers to assist in the research. There are many more worthy projects out there that could use what resources I can give them that behave in a more acceptable manner.

Profile Faik
Send message
Joined: 3 Oct 11
Posts: 2771
Credit: 140,700
RAC: 0
Message 10099 - Posted: 13 Feb 2012, 8:37:09 UTC - in response to Message 10098.

Dear Michael,


The C@H team members deeply appreciate what you and other volunteers are doing for the project. We understand and appreciate that you are donating valuable time and resources to this project and that you are here only because you care about this project and the advancement of science. Your support is vital and without it this very project could not continue.


This is an open public forum, and community members use it to interact with us and with each other. Their opinions are reflect of their personal views and they do not necessarily reflect the views of the C@H Administration. Please keep in mind that many of our community members do not speak English as a first language, which makes communication even harder. I hope that this clarifies things a bit. You are certainly very welcome here, and your vital support is deeply appreciated.


Thank you for being with us, and thank you for your understanding.


Faik

---
Member of the C@H Team!

Profile cykodennis
Send message
Joined: 31 May 10
Posts: 234
Credit: 4,896,378
RAC: 0
Message 10104 - Posted: 13 Feb 2012, 15:33:06 UTC - in response to Message 10098.

Dear Michael,
you have just misunderstood me. I implied nothing, i just told you openly, from user to user, that i fully understand when you decide to quit. What i wrote is what you can read.

Depending your last post:
Sometimes i ask myself, when i read what my fellow crunchers write:
"What do they think should define the structure of a distributed computing project?"
The scientific task itself, or fitting into the preferences which M.James (f.e.) has set?

Is the task "finding-the-best-fitting-model-of-the-universe" comparable to the task "listening-for-E.T."? I do not think so. I'm no scientist, but i can imagine that different tasks need different formulas, different workunits, different data...
And so i do not fully understand how you can decide, that C@H is a badly behaving application instead of a simply more complex project (compared to others).




Message boards : Wish list : So why dont I start a NEW thread