Thanks. Now THAT's how a utility download should work!

Thanks. Now THAT's how a utility download should work!

I think this utility is what saved me. I was getting errors; video driver stopped, video driver restarted, or something of the sort. I really though it was caused by my BOINC settings, because it just happened out of the blue and everything else was still running fine.

DDU beats AMDcleanup hands down.

Now running Crimson. Everything seems to be stable again and my task errors have stopped (I hope).

---

A lot have been split but I'm have a hard time getting them.

---

I thought most people used Windoze because every good game ever is directx. I know that's the only reason I'm still using Windows.


Other than that.. I'm still not fairing well on replenishing my cache. It also doesn't help that I have to keep babysitting BOINC because after about 3 or 4 "no work available" replies, BOINC goes into what I call a "phantom backoff" where it doesn't SAY it is in a backoff... but it really is. It will wait 5m 3s for 2-3 contacts, and then next thing I know, it's been 45m since it last tried, and "status" is blank and there's no indication anywhere that it is in some arbitrary exponential backoff.

If you were using Linux you could cobble together a short bash script that runs every minute using cron and checks stdoutdae.txt for the last time "Project requested delay of 303 seconds" occurred. If it is more then 245 seconds, then sleep the appropriate amount of time to make 305 seconds and then force an update. Linux is good for something :).

You could do the same thing in windows with a powershell script & task schedule. Even easier would be to schedule a task to run about every 310 seconds telling BOINC to update.
When I was running 30+ machines I would use one system to tell all of my hosts to update every 6 hours. That way I could look at my project computers list to see if any were out of communication & issue a reboot.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

I am running two AP open cl tasks on nvidia.
Has someone seen the effect, that one task is processed faster at the expense of the other?
If i start them exactly at the same time, they finish at the same time (more or less), using about the same amount of cpu.

Usually you see it with AP and MBs, where the MBs tend to slow down and the APs accelerate.

Not heard of it with 2 APs.

The common wisdom is to not run mixed aps on GPUs, they waste time loading the next app.

---

The common wisdom is to not run mixed aps on GPUs, they waste time loading the next app.

That may explain who pour molasses into my system........me......":D>

---

"Sour Grapes make a bitter Whine." <(0)>

The common wisdom is to not run mixed aps on GPUs, they waste time loading the next app.

Not sure how big a deal that really is, unless your GPU is smokin' fast. Also suspect that the app would be loaded fresh each time in any event.
But I did take advice I got from another thread about setting it so that there'll never be more than 1 AP running per GPU (set APs at .51 of a GPU and MBs at .49 in app-config, thus permitting 1 of each or two MBs). Jury's still out on that change, for me at least, but I think it will leave me in a place where I don't need to change config each time I'm running APs versus MBs. That's been getting tiresome lately even with BOINTasks loaded to enable centralized changes.

---

The common wisdom is to not run mixed aps on GPUs, they waste time loading the next app.

This depends on the GPU.

I have no problems running mixed apps.
Same for x70`s and x80`s NV cards.

---

With each crime and every kindness we birth our future.

Try it on a Mac with Two GTX 750ti. All it takes is to run a CUDA on One card and an OpenCL on the other. The OpenCL App is Slowed by around 30% or more and the CUDA card is Slowed by around 300%. Yes, about 3 times Slower on the CUDA App if the other card is running an OpenCL task. You don't even have to run the tasks on the same card, the problem happens with running a Single task on each card. Fortunately most of the APs I manage to receive are sent to the ATI card.

not helped by the usual mass of fake download buttons embedded in advertising

Don't you know of uBlock Origin?
https://chrome.google.com/webstore/detail/ublock-origin/cjpalhdlnbpafiamejdnhcphjbkeiagm
https://addons.mozilla.org/en-US/firefox/addon/ublock-origin/

https://github.com/gorhill/uBlock
https://en.wikipedia.org/wiki/UBlock


Path to DDU from link in your post:
http://www.wagnardmobile.com/
http://www.wagnardmobile.com/?q=node/51
"Click here for Info and Download" - http://www.wagnardmobile.com/forums/
http://www.wagnardmobile.com/forums/viewforum.php?f=5

http://www.wagnardmobile.com/forums/viewtopic.php?f=5&t=164

"Official Download Here" gets:
http://www.wagnardmobile.com/DDU/download/DDU%20v15.7.3.1.exe

---

 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 

The number of channels of MB is dropping maybe they will reload the tapes and APs soon will be split, that would be a good thing.

---

Well they split a few and then quit but I did get 3.

---

There is a new dataset splitting some AP just now....
Good luck.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

Received my first AP on my Surface 3. 2 days to process? Probably not, but interested to see...

---

Hopefully your Surface will work better than the $400 Million Official NFL Sidelines version(which the TV people keep referring to as IPADS, much to M$ chagrin)......";>)

---

"Sour Grapes make a bitter Whine." <(0)>

Hopefully your Surface will work better than the $400 Million Official NFL Sidelines version(which the TV people keep referring to as IPADS, much to M$ chagrin)......";>)

I won't go into all of the reasons I didn't purchase the Pro. It is a very handy little table though. I also have a company issued iPad. Not an Apple fan or basher, but the iPad is not nearly as useful, not even close.

---

I can`t run multiple task simultaneously on 780ti, now i only run 1 gpu per Ap task and 0.3 on regular. I get 80% usage on one AP task so it`s ok.

---

APs continue to be split and as has been the pattern of late I'm not getting many.

---