While I understand the WU shortage situation at MW@H but I'm wondering, why some users can get, probably (based on RAC), >1000 WUs while most of the users here get next to nothing?

This computer (as well as other computers this user has) seems to have no problem getting WUs at all. Does MW@H project has preferred list of users who would get more WUs?

I have no idea! Some of my computers can get a lot of WU's and others get zip.

For a start they are using GPUs in their machines, which get through the wu's very quickly so their machines then ask for more work more often.

Secondly they are either using a flag in older BOINC clients (-return_results_immediately) or they are using some form of script that overrides the 'normal' BOINC exponential back off system and contacts the server to get work more often. The script utilises a standard feature that comes with BOINC.

Thirdly, they are potentially just lucky....

For a start they are using GPUs in their machines, which get through the wu's very quickly so their machines then ask for more work more often.

Secondly they are either using a flag in older BOINC clients (-return_results_immediately) or they are using some form of script that overrides the 'normal' BOINC exponential back off system and contacts the server to get work more often. The script utilises a standard feature that comes with BOINC.

Thirdly, they are potentially just lucky....

where can i get one of these scripts? all my machines are dry as a bone and have been for a month.

Got 35 WUs in last 2 hours - it's almost equal as i had for last 10 days.
Seems there is no work for me now :/

I am not sure if it is luck, or client versions, or something else ... but a couple of my systems have not gotten work in a long time.

Oh, well, WCG started a new sub-project so off to mine some bronze, silver, and gold...

bob>where can i get one of these scripts?
Mine [XP] says:

@echo off
REM Boinc Automatic Project Update
:TOP
"c:\Program Files\BOINC\boinccmd.exe" --project http://milkyway.cs.rpi.edu/milkyway/ update
ECHO %TIME%
SLEEP 5m
GOTO TOP

You'll have to get SLEEP from the net, or fake it using PING as mentioned elsewhere on this board

---

Cheers,

PeterV

.

bob>where can i get one of these scripts?
Mine [XP] says:

@echo off
REM Boinc Automatic Project Update
:TOP
"c:\Program Files\BOINC\boinccmd.exe" --project http://milkyway.cs.rpi.edu/milkyway/ update
ECHO %TIME%
SLEEP 5m
GOTO TOP

You'll have to get SLEEP from the net, or fake it using PING as mentioned elsewhere on this board

I see you use SLEEP 5m (I take it m=minutes)
Stupid thing is that i've been converting minutes to seconds, hence, SLEEP 350.
As usual, you live and learn!

---

OMG..it's been posted again! The flames are going to start again now....unless, of course, the previous flammers are using the script themselves....LOL.

This script works and gives you choices as to time between requests and number of requests .... works on XP .....

--------------------------------------------------------------------------
@echo off
Echo.
echo.
Echo Request Work From Milkyway Sever
echo.
Echo.
echo.
set /p xx= Server Request Work Interval - Enter Seconds :

cls
Echo.
echo.
echo Server Request Interval %xx% seconds.
Echo.
echo.
set /p zz= Enter the number of requests :


set /a Request_limit = %zz%
set /a seconds = %xx%
set /a looper = 0
set /a timecount = %zz%*%xx%

set /a h = %timecount%/3600
set /a m = (%timecount%-(%h%*3600))/60
set /a s = (%timecount%-(%h%*3600)-(%M%*60))


cls
echo.
echo.
echo. Milkyway - CPU ONLY
echo.
echo Sending %zz% requests - One Every %xx% seconds.

ping localhost -n 3 > NULL

goto main

:main


IF %looper%==%Request_limit% goto finish

goto app

:app

C:\PROGRA~1\BOINC\boinccmd.exe --passwd 11110 --project http://milkyway.cs.rpi.edu/milkyway update

ping localhost -n %xx% > NULL

goto nest

:Nest
set /a countdown = %timecount%-(%xx%*%looper%)
set /a h1 = %countdown%/3600
set /a m1 = (%countdown%-(%h1%*3600))/60
set /a s1 = (%countdown%-(%h1%*3600)-(%M1%*60))


SET /A looper=%looper%+1


cls
echo.
echo.
echo. Milkyway - CPU ONLY
echo.
echo Sending %zz% requests - Once Every %xx% seconds.
echo.
echo.
Echo %looper% of %Request_limit% Requests
echo.
echo.
echo Total Runtime of Batch File %h%:%m%:%s%
echo.
echo Time remaining %h1%:%m1%:%s1%

goto main

:finish

cls
echo.
echo.
echo.
echo.
Echo Milkyway Request Count total is %looper% of %Request_limit% - Completed Sequence
echo.
Echo Dos Windows Closing - Have a nice day

ping localhost -n 10 > NULL
exit
------------------------------------------------------------------------

---

OMG..it's been posted again! The flames are going to start again now....unless, of course, the previous flammers are using the script themselves....LOL.

Actually, like I told boosted, that's where all of this is going, a big old hissy fit because some people are getting work and others are not. This same stuff happened over at LHC. You were there for that, so I know you know what I'm talking about...

The scripts posted all have the capability of hammering the server with RPC requests that can end up being a DoS attack if enough people start using them. If you all want to be that determined to get points, then so be it I suppose. That's what it's really about, getting points, because jealousy has raised its' ugly head... "I see so-and-so is getting work and I'm not. It's not fair!", ect, etc, etc...

Brian, I think your post about it, sums it all up quite nicely. I don't use scripts myself because I don't want to help bring down the servers, such that none of us get any work. If others want to use them that's up to them. I'm not just here for credit, with my lowly stuff I couldn't compete anyway. It's also about liking the project and the social side of it as well.

Judging by past experiences any posts of mine are generally not welcomed in MW crunchers, so I'll disappear again back to lurking mode. Just thought I'd give you a friendly vote.

Brian, I think your post about it, sums it all up quite nicely. I don't use scripts myself because I don't want to help bring down the servers, such that none of us get any work. If others want to use them that's up to them. I'm not just here for credit, with my lowly stuff I couldn't compete anyway. It's also about liking the project and the social side of it as well.

Judging by past experiences any posts of mine are generally not welcomed in MW crunchers, so I'll disappear again back to lurking mode. Just thought I'd give you a friendly vote.

Thanks Chris... The thing is, in my prior job, one of the credit card processors we used had an issue where their link would crash and cut off communication to us. They claimed they had it fixed and wanted me to test it. I put a VB app that we had already developed to test authorizations into a tight loop and brought down the link within about 30 seconds.

I know what these scripts are capable of doing. Just because one person is careful, doesn't mean that everyone will be. Also, even if everyone is careful, the aggregate total of thousands of people running scripts can end up being very bad...

This script works and gives you choices as to time between requests and number of requests .... works on XP .....

--------------------------------------------------------------------------
@echo off
Echo.
echo.
Echo Request Work From Milkyway Sever
echo.
Echo.
echo.
set /p xx= Server Request Work Interval - Enter Seconds :

cls
Echo.
echo.
echo Server Request Interval %xx% seconds.
Echo.
echo.
set /p zz= Enter the number of requests :


set /a Request_limit = %zz%
set /a seconds = %xx%
set /a looper = 0
set /a timecount = %zz%*%xx%

set /a h = %timecount%/3600
set /a m = (%timecount%-(%h%*3600))/60
set /a s = (%timecount%-(%h%*3600)-(%M%*60))


cls
echo.
echo.
echo. Milkyway - CPU ONLY
echo.
echo Sending %zz% requests - One Every %xx% seconds.

ping localhost -n 3 > NULL

goto main

:main


IF %looper%==%Request_limit% goto finish

goto app

:app

C:\PROGRA~1\BOINC\boinccmd.exe --passwd 11110 --project http://milkyway.cs.rpi.edu/milkyway update

ping localhost -n %xx% > NULL

goto nest

:Nest
set /a countdown = %timecount%-(%xx%*%looper%)
set /a h1 = %countdown%/3600
set /a m1 = (%countdown%-(%h1%*3600))/60
set /a s1 = (%countdown%-(%h1%*3600)-(%M1%*60))


SET /A looper=%looper%+1


cls
echo.
echo.
echo. Milkyway - CPU ONLY
echo.
echo Sending %zz% requests - Once Every %xx% seconds.
echo.
echo.
Echo %looper% of %Request_limit% Requests
echo.
echo.
echo Total Runtime of Batch File %h%:%m%:%s%
echo.
echo Time remaining %h1%:%m1%:%s1%

goto main

:finish

cls
echo.
echo.
echo.
echo.
Echo Milkyway Request Count total is %looper% of %Request_limit% - Completed Sequence
echo.
Echo Dos Windows Closing - Have a nice day

ping localhost -n 10 > NULL
exit
------------------------------------------------------------------------

Ok, well I'm dummy enough for not knowing how this script works or how can I utilised that. Personally I don't want to bombard server with loop requests but I just think there must be a way to fix this, perhaps WUs quota? I admitted that I get some negative feeling when certain peoples get tonnes while the rest barely get anything.

Well i have done a bit of digging to why not everyone is getting work units all the time.

The credit per day numbers come from boinc stats but this is really intresting numbers and maybe why the server is having really bad problems keeping up.

Lets start right of the top 100 users from Milkyway boinc stats

26 users have day credit of over 100,000

Now if you avarege that day credit total but say 27 credit per work unit
that adds up to 3703 work units per day.

The top user is doing about 1,500,000 credit per day.
Which works out at 55,555 work units per day.

Now since we can only have 6 work units per cpu at a time just from those few users it works out at 151,937 units per day and since a lot of those will be done through gpus.

Oh yes just a few more number there is 18013 users and out of the first 1000 users only 264 are not running the project or just cant get work but may be still making request calls for work.

---

Smoke me a kipper ill be back for breakfast

Well i have done a bit of digging to why not everyone is getting work units all the time.

I sure wish the project would say or do something in regards to the GPU project. Even if ATI isn't available right away, the potential exists for people with CUDA-capable cards that are currently processing with CPU to switch over to GPU. This will help everyone, even if it isn't the ATI cards...

Beyond that, I don't doubt that there are a LOT of scripts in use. I tend to doubt they really have that much more efficacy than just letting BOINC do its' own thing... Of course, someone will point out that I don't have the hardware so I don't know what I'm talking about...

What I do know though is that some people have "good" luck, other people have "normal" luck, and others have "bad" luck. You know the person with "good luck"...they win things frequently, have a super-high metabolism that enables them to eat whatever they want without exercising and still look borderline underweight, etc, etc, etc... Then there are those of us with "bad luck"...the people that can be as careful as they can be while eating spaghetti with a white shirt on and make it all the way to the last bite and dump it on our shirts, or find the only nail in the road and drive our car over it and puncture a tire, etc, etc, etc... Some of us are extremely capable of snatching defeat from the jaws of victory...

One might consider that there is luck involved here as well...

Once again, we have people who cannot appreciate that running scripts is in fact part of the reason many people can't get WUs. Oh well, I still won't run a script. I'll rather just let boinc get work from other projects. MW has quite a high debt on some of my hosts, but at least I'm getting some WUs fairly often.

Of course, the easy way to stop the hammering of the server would be to reduce the credits to that of *cough*seti*cough*, or even below that. Then watch how quickly people will stop crunching and hammering here - after some very unpleasant postings in the forum, no doubt.

Excuse me while I put on my fire-resistant suit!

I think that we all overlooking the fact that work units are allocated 6 per cpu- core. This is totally irrelevant to the crunching capacity of your GPU.
So you can script or manually refresh till you are blue in the face.
A 3850 installed in an i7 will earn more than a 4890 in a dual because the i7 will get 48 work units and the dual only 12

---

A BLAST FROM YOUR PAST

Anything will earn more than a 4890 because as far as I know nobody's got a 4890 to run the WU's with any Stability yet ... ;)

Once again, we have people who cannot appreciate that running scripts is in fact part of the reason many people can't get WUs. Oh well, I still won't run a script. I'll rather just let boinc get work from other projects. MW has quite a high debt on some of my hosts, but at least I'm getting some WUs fairly often.

Of course, the easy way to stop the hammering of the server would be to reduce the credits to that of *cough*seti*cough*, or even below that. Then watch how quickly people will stop crunching and hammering here - after some very unpleasant postings in the forum, no doubt.

Yep. Suddenly the demand would be greatly reduced.

Excuse me while I put on my fire-resistant suit!

Mine got burned off some time ago...