Welcome to MilkyWay@home

Nbody WU Flush

Message boards : News : Nbody WU Flush
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 3 · 4 · 5 · 6 · 7 · 8 · 9 . . . 22 · Next

AuthorMessage
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 73166 - Posted: 29 Apr 2022, 21:40:35 UTC - in response to Message 73163.  

Yes but the Project has died out like so many others so they are now sitting on the shelf gathering dust. It was a Boinc Project called Bitcoin Utopia and the idea was to crunch using cpu's, gpu's and or asic miners to generate money to donate to things like someone wrote a book on how to populate Mars, another time we sent money to some of the stats sites, one time we even helped fund another Boinc Project. But like alot of Boinc Projects over the years it's not gone by the wayside, you have to remember back when Bitcoins weren't were more than a dollar or so each so it took awhile to send some $1500US.
I never found mining Bitcoins to make any money. I'd be better just donating cash.
ID: 73166 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3315
Credit: 519,946,411
RAC: 22,500
Message 73167 - Posted: 29 Apr 2022, 23:34:55 UTC - in response to Message 73166.  

Yes but the Project has died out like so many others so they are now sitting on the shelf gathering dust. It was a Boinc Project called Bitcoin Utopia and the idea was to crunch using cpu's, gpu's and or asic miners to generate money to donate to things like someone wrote a book on how to populate Mars, another time we sent money to some of the stats sites, one time we even helped fund another Boinc Project. But like alot of Boinc Projects over the years it's not gone by the wayside, you have to remember back when Bitcoins weren't were more than a dollar or so each so it took awhile to send some $1500US.


I never found mining Bitcoins to make any money. I'd be better just donating cash.


At $38KUS today Bitcoin IS worth mining IF you have high speed Asic miners or some very cheap electricity like solar or wind or even water AND enough batteries to get you thru the times when whatever isn't working. One Bitcoin mining company bought a fricking power plant in Pennsylvania, USA recently because it generates electricity by water which runs on 2 sides of the power plant.
ID: 73167 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 73168 - Posted: 29 Apr 2022, 23:38:03 UTC - in response to Message 73167.  

Yes but the Project has died out like so many others so they are now sitting on the shelf gathering dust. It was a Boinc Project called Bitcoin Utopia and the idea was to crunch using cpu's, gpu's and or asic miners to generate money to donate to things like someone wrote a book on how to populate Mars, another time we sent money to some of the stats sites, one time we even helped fund another Boinc Project. But like alot of Boinc Projects over the years it's not gone by the wayside, you have to remember back when Bitcoins weren't were more than a dollar or so each so it took awhile to send some $1500US.


I never found mining Bitcoins to make any money. I'd be better just donating cash.


At $38KUS today Bitcoin IS worth mining IF you have high speed Asic miners or some very cheap electricity like solar or wind or even water AND enough batteries to get you thru the times when whatever isn't working. One Bitcoin mining company bought a fricking power plant in Pennsylvania, USA recently because it generates electricity by water which runs on 2 sides of the power plant.
Yeah you buy the high speed miner then a better one comes out next week and yours are too slow to make money.

You could always steal the power:
https://www.thesun.co.uk/news/15096237/cops-find-bitcoin-mine/
ID: 73168 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3315
Credit: 519,946,411
RAC: 22,500
Message 73170 - Posted: 29 Apr 2022, 23:56:50 UTC - in response to Message 73163.  


@mikey:
Did Boinc exist 40 years ago? I had its predecessor United Devices running in the late 90s.


Yes I was one of the very first Boinc users waaaay back in the 1980's:

SETI@home member since 17 Dec 1999

That is the current version of Boinc and these are credits I got under the previous version of Boinc:
SETI@home classic CPU time 383,036 hours

Generally speaking back then we got one credit for each task done but there was ALOT of cheating going on, not by me, where people would crunch a task to about 95% complete then suspend it and copy it to a 2nd, 3rd, 4th etc pc and finish it AND get credit for completing it. That's just one of the reasons they have ALOT more security built into the current version of Boinc. Back then we also did 'sneaker net' stuff where one pc was hooked up to the internet, it was ALL dial-up back then, and it would download a large batch of tasks and then you could copy some tasks onto a floppy disk and crunch them on a different pc then bring the completed units back to the original pc to report them. Yes I DID do that and no it was not discouraged because they were worried about the time it took for each individual pc to connect to Seti, return the old tasks and get new tasks all the while someone else was getting a back-off notice saying the Server was unavailable. With todays fast internet and MUCH MUCH faster pc's and Servers all that time has been cut by over 100 times BUT it still takes time for the Server to respond to a pc trying to connect, make the connection and then do what it needs to do all the while someone else is getting a back-off notice saying the Server is unavailable.

Back then Seti was the ONLY option for crunching so it just was what it was, one time someone stole the copper phones lines, dug them right up out of the ground and made off with them and the Project just shutdown for the 2 weeks or more it took to replace the phone lines. No notice of any reason, no NOTHING except eventually by someone on a website told us what was going on. Todays people have zero clue about Admins not being communicative!! LOL
ID: 73170 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 73171 - Posted: 30 Apr 2022, 0:31:45 UTC - in response to Message 73170.  

Argh! The fan on my new Fury just jammed. Stupid thing didn't bother to thermally throttle, it just shut off the computer power. Twice. I wondered what was going on.

Yes I was one of the very first Boinc users waaaay back in the 1980's:
I thought it wasn't called Boinc until 2000 when it was born from United Devices and Seti?

Generally speaking back then we got one credit for each task done
And now we're on a million credits a day for a few GPUs.

Back then Seti was the ONLY option for crunching so it just was what it was, one time someone stole the copper phones lines, dug them right up out of the ground and made off with them and the Project just shutdown for the 2 weeks or more it took to replace the phone lines. No notice of any reason, no NOTHING except eventually by someone on a website told us what was going on. Todays people have zero clue about Admins not being communicative!! LOL
ROFL at the copper theft. Not quite as funny as when they steal live railway wires.
ID: 73171 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3315
Credit: 519,946,411
RAC: 22,500
Message 73172 - Posted: 30 Apr 2022, 2:00:35 UTC - in response to Message 73171.  

Argh! The fan on my new Fury just jammed. Stupid thing didn't bother to thermally throttle, it just shut off the computer power. Twice. I wondered what was going on.


Whoops, at least it shut down instead of frying everything!!

[/quote]
Yes I was one of the very first Boinc users waaaay back in the 1980's:


I thought it wasn't called Boinc until 2000 when it was born from United Devices and Seti?[/quote]

I never crunched for United devices so no not that way. Let me get my main desktop up and running probably next week and I will pm you the original Boinc proposal for funding, I think it has a date on it.

Generally speaking back then we got one credit for each task done
And now we're on a million credits a day for a few GPUs.[/quote]

I now but that won't ever happen at a "credit new" Project, at least not in our lifetimes.

[/quote]
Back then Seti was the ONLY option for crunching so it just was what it was, one time someone stole the copper phones lines, dug them right up out of the ground and made off with them and the Project just shutdown for the 2 weeks or more it took to replace the phone lines. No notice of any reason, no NOTHING except eventually by someone on a website told us what was going on. Todays people have zero clue about Admins not being communicative!! LOL


ROFL at the copper theft. Not quite as funny as when they steal live railway wires.[/quote]

Yeah that would be a hard no for me too!! LOL
ID: 73172 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 73173 - Posted: 30 Apr 2022, 2:29:43 UTC - in response to Message 73172.  
Last modified: 30 Apr 2022, 2:33:03 UTC

Argh! The fan on my new Fury just jammed. Stupid thing didn't bother to thermally throttle, it just shut off the computer power. Twice. I wondered what was going on.
Whoops, at least it shut down instead of frying everything!!
Fan's fine. XFX is a very stupid crap manufacturer, do not buy from them. So an XFX label on the fan, made of paper? No? Let's make it out of a steel disk. So when it comes unstuck, it flings itself into the blades and jams them. Ridiculous. In retaliation, I have removed the label and placed it in the bin. The GPU is now operating anonymously.

I thought it wasn't called Boinc until 2000 when it was born from United Devices and Seti?
I never crunched for United devices so no not that way. Let me get my main desktop up and running probably next week and I will pm you the original Boinc proposal for funding, I think it has a date on it.
In the very late 90s when I put United Devices on everyone's machine at work, a professor already had Seti, but it wasn't called Boinc. He had a dual core machine (one of two people in the whole department, actually it was two single core CPUs!), and both Seti and United Devices only used one core, so he ran both. I don't think anyone archived the stats, but I think I got very high up, using 100s of work machines.

And now we're on a million credits a day for a few GPUs.
I now but that won't ever happen at a "credit new" Project, at least not in our lifetimes.
They give less credits? All I know is they're very inaccurate, for example you can cheat them easily on LHC by using the wrong number of cores. Hence LHC is banned from Gridcoin.

ROFL at the copper theft. Not quite as funny as when they steal live railway wires.
Yeah that would be a hard no for me too!! LOL
I have been known to short out a 100A cable. That was sore. Molten copper all up my arm.
ID: 73173 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile HRFMguy

Send message
Joined: 12 Nov 21
Posts: 236
Credit: 575,029,129
RAC: 37,181
Message 73174 - Posted: 30 Apr 2022, 3:14:00 UTC - in response to Message 73173.  

I have been known to short out a 100A cable. That was sore. Molten copper all up my arm.

i dub thee "Sir Sparky!"
ID: 73174 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 73175 - Posted: 30 Apr 2022, 4:50:13 UTC - in response to Message 73174.  
Last modified: 30 Apr 2022, 5:00:02 UTC

I have been known to short out a 100A cable. That was sore. Molten copper all up my arm.

i dub thee "Sir Sparky!"
Breaking stuff is fun. Let's melt some tools while drunk:
https://youtu.be/DJOX0c60wQE
ID: 73175 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
San-Fernando-Valley

Send message
Joined: 13 Apr 17
Posts: 256
Credit: 604,411,638
RAC: 0
Message 73176 - Posted: 30 Apr 2022, 7:03:16 UTC - in response to Message 73164.  

@Jaari:

The reason a bunch of older tasks don't validate is because their second task got cancelled, but no new task is being generated for it, it seems.
One of my own tasks to show the second task having been cancelled: https://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=411722482

As I understand it:
This task has a "quorum" of one (as thousands of others do also).
A second (wingman) task can't be cancelled, because there is none.
It wasn't needed, because of the quorum. So it was never even generated.
Tasks marked "validation inconclusive" will eventually be marked as "valid".
Might take a while, as already stated/mentioned in several previous discussions/posts.
ID: 73176 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 73177 - Posted: 30 Apr 2022, 7:08:20 UTC - in response to Message 73176.  

@Jaari:

The reason a bunch of older tasks don't validate is because their second task got cancelled, but no new task is being generated for it, it seems.
One of my own tasks to show the second task having been cancelled: https://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=411722482

As I understand it:
This task has a "quorum" of one (as thousands of others do also).
A second (wingman) task can't be cancelled, because there is none.
It wasn't needed, because of the quorum. So it was never even generated.
Tasks marked "validation inconclusive" will eventually be marked as "valid".
Might take a while, as already stated/mentioned in several previous discussions/posts.
Validation requires server calculations? I thought it had caught up now.
ID: 73177 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
San-Fernando-Valley

Send message
Joined: 13 Apr 17
Posts: 256
Credit: 604,411,638
RAC: 0
Message 73178 - Posted: 30 Apr 2022, 7:28:33 UTC - in response to Message 73177.  

@Jaari:

The reason a bunch of older tasks don't validate is because their second task got cancelled, but no new task is being generated for it, it seems.
One of my own tasks to show the second task having been cancelled: https://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=411722482

As I understand it:
This task has a "quorum" of one (as thousands of others do also).
A second (wingman) task can't be cancelled, because there is none.
It wasn't needed, because of the quorum. So it was never even generated.
Tasks marked "validation inconclusive" will eventually be marked as "valid".
Might take a while, as already stated/mentioned in several previous discussions/posts.
Validation requires server calculations? I thought it had caught up now.

Hmmm, good question !
Probably some kind of action has to be started that does the checking for quorums of one to get validated ?

No matter what, I just have to many of those hanging around in my "validation invalid" queue.
On top of that I'm getting confused.
Just checked that queue and noticed that now there are plenty "vlaidation inconclusive" tasks with a quorum of one where there are actually further tasks actively running (as _1 or _2).
So, what is the deal ?
ID: 73178 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
San-Fernando-Valley

Send message
Joined: 13 Apr 17
Posts: 256
Credit: 604,411,638
RAC: 0
Message 73179 - Posted: 30 Apr 2022, 7:30:23 UTC - in response to Message 73175.  

I have been known to short out a 100A cable. That was sore. Molten copper all up my arm.

i dub thee "Sir Sparky!"
Breaking stuff is fun. Let's melt some tools while drunk:
https://youtu.be/DJOX0c60wQE

I wonder if doing that with some old, but running, HDDs would do the trick ....
ID: 73179 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 73180 - Posted: 30 Apr 2022, 7:57:00 UTC - in response to Message 73178.  

No matter what, I just have to many of those hanging around in my "validation invalid" queue.
On top of that I'm getting confused.
Just checked that queue and noticed that now there are plenty "vlaidation inconclusive" tasks with a quorum of one where there are actually further tasks actively running (as _1 or _2).
So, what is the deal ?
The problem is Boinc is a cobbled together piece of shit. The server is worse than the client. It's written by volunteer programmers - as in those who aren't good enough to get a real paid job. I'm currently trying to have a bash at Einstein, and am getting this stupid message:

"Einstein@Home Not requesting tasks: don't need (CPU: not highest priority project; AMD/ATI GPU: )"

Specifically, no reason given for AMD, just can't be bothered. At that point I had 10 minutes of work left, running on 1 GPU. The other GPU was empty. My buffer was increased to 5 + 5 days. Only Einstein was enabled to get new work. Facepalm at Boinc programmers.
ID: 73180 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 73181 - Posted: 30 Apr 2022, 8:02:41 UTC - in response to Message 73179.  

I have been known to short out a 100A cable. That was sore. Molten copper all up my arm.

i dub thee "Sir Sparky!"
Breaking stuff is fun. Let's melt some tools while drunk:
https://youtu.be/DJOX0c60wQE

I wonder if doing that with some old, but running, HDDs would do the trick ....
Why waste them, make love, I mean music: https://youtu.be/8Hot4j97Mbk
ID: 73181 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3315
Credit: 519,946,411
RAC: 22,500
Message 73183 - Posted: 30 Apr 2022, 10:38:04 UTC - in response to Message 73180.  

No matter what, I just have to many of those hanging around in my "validation invalid" queue.
On top of that I'm getting confused.
Just checked that queue and noticed that now there are plenty "vlaidation inconclusive" tasks with a quorum of one where there are actually further tasks actively running (as _1 or _2).
So, what is the deal ?
The problem is Boinc is a cobbled together piece of shit. The server is worse than the client. It's written by volunteer programmers - as in those who aren't good enough to get a real paid job. I'm currently trying to have a bash at Einstein, and am getting this stupid message:

"Einstein@Home Not requesting tasks: don't need (CPU: not highest priority project; AMD/ATI GPU: )"

Specifically, no reason given for AMD, just can't be bothered. At that point I had 10 minutes of work left, running on 1 GPU. The other GPU was empty. My buffer was increased to 5 + 5 days. Only Einstein was enabled to get new work. Facepalm at Boinc programmers.


The Boinc Programmers never even thought about using a gpu to crunch with in the beginning, over several version of Boinc, both the Client and Server side, gpu crunching was enabled but even then it was 'who in their right mind wants to use more than one gaming gpu to crunch with' thinking. Do you have a line in your cc_config.xml file to 'use all gpu's' in Boinc? If not it won't use both of them unless you put them in different pc's.
ID: 73183 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3315
Credit: 519,946,411
RAC: 22,500
Message 73184 - Posted: 30 Apr 2022, 10:46:31 UTC - in response to Message 73180.  

No matter what, I just have to many of those hanging around in my "validation invalid" queue.
On top of that I'm getting confused.
Just checked that queue and noticed that now there are plenty "vlaidation inconclusive" tasks with a quorum of one where there are actually further tasks actively running (as _1 or _2).
So, what is the deal ?
The problem is Boinc is a cobbled together piece of shit. The server is worse than the client. It's written by volunteer programmers - as in those who aren't good enough to get a real paid job. I'm currently trying to have a bash at Einstein, and am getting this stupid message:

"Einstein@Home Not requesting tasks: don't need (CPU: not highest priority project; AMD/ATI GPU: )"

Specifically, no reason given for AMD, just can't be bothered. At that point I had 10 minutes of work left, running on 1 GPU. The other GPU was empty. My buffer was increased to 5 + 5 days. Only Einstein was enabled to get new work. Facepalm at Boinc programmers.


Boinc was originally written by PAID Programmers that due to funding cuts NOW do it on a Volunteer basis, it wasn't until a few years ago that they lost their funding, the original paper for the idea of Boinc has long ago been met and even surpassed. Several places provided funding over the years but now no place with the money is interested in keeping it going, they have other new and fresh ideas to fund. Boinc isn't going to make someone a Billion or even a Million dollars while other ideas might so the funding just isn't there. Even projects that run thru Universities have to fight to keep the budget they have let alone get increases every year.
ID: 73184 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3315
Credit: 519,946,411
RAC: 22,500
Message 73186 - Posted: 30 Apr 2022, 10:54:03 UTC - in response to Message 73173.  
Last modified: 30 Apr 2022, 10:55:55 UTC

And now we're on a million credits a day for a few GPUs.
I now but that won't ever happen at a "credit new" Project, at least not in our lifetimes.


They give less credits? All I know is they're very inaccurate, for example you can cheat them easily on LHC by using the wrong number of cores. Hence LHC is banned from Gridcoin.[/quote]

Yes "Credit New" is a Seti invention to standardize the Credits over all the Projects, the problem is each project has it's own agenda and thinking about attracting new users, some just don't have the funding to upgrade the hardware if a hundred thousand new people show up tomorrow to crunch their tasks, while others seem to be able to handle it without so much as breathing heavy. Even Seti was running in a 'closet' before they finally shut the doors and gave up due to the lack of funding. Since the original Boinc Programmers all worked for Seti "Credit new" is hard coded into every version of the Boinc Server and has to be changed if a project doesn't want to use it. AND it's not just in one place it's all over the place according to to some past Admins who had to change it. "Credit New" is based on a formula that doesn't really translate well across the different cpu's and gpu's people use because they are just different.
ID: 73186 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
San-Fernando-Valley

Send message
Joined: 13 Apr 17
Posts: 256
Credit: 604,411,638
RAC: 0
Message 73187 - Posted: 30 Apr 2022, 11:01:45 UTC - in response to Message 73183.  
Last modified: 30 Apr 2022, 11:04:50 UTC

.... Do you have a line in your cc_config.xml file to 'use all gpu's' in Boinc? If not it won't use both of them unless you put them in different pc's.

I'm sure Peter knows that very secret trick !

But, I think (as far as I'm able), that that isn't the problem ...
So, I'm also sure the more knowledgeable crunchers will explain ...
ID: 73187 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 73188 - Posted: 30 Apr 2022, 11:05:03 UTC - in response to Message 73183.  
Last modified: 30 Apr 2022, 11:08:21 UTC

The Boinc Programmers never even thought
That's accurate if you just leave the sentence there.

about using a gpu to crunch with in the beginning, over several version of Boinc, both the Client and Server side, gpu crunching was enabled but even then it was 'who in their right mind wants to use more than one gaming gpu to crunch with' thinking. Do you have a line in your cc_config.xml file to 'use all gpu's' in Boinc? If not it won't use both of them unless you put them in different pc's.
I have it set to only let gravity run on one, and only gamma on another. I think I'll try taking that complication out and see what happens.... you fixed it! Sort of. If you specify gamma not to go onto device 1, and gravity not to go onto device 0, it's too complicated for Boinc and virtually nothing gets downloaded, only 1 or 2 tasks at a time. So I've compromised and just said no gravity on device 1 (as it can't do them). I wanted to save the few gamma that appeared for the other card, but it seems I cannot. I'll put that onto Milkway.

https://github.com/BOINC/boinc/issues/4738
ID: 73188 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 . . . 3 · 4 · 5 · 6 · 7 · 8 · 9 . . . 22 · Next

Message boards : News : Nbody WU Flush

©2024 Astroinformatics Group