Others discuss users holding onto work, is this because they have their system in burp set to store for example 20 days of work or more?

Yes, many users have their machines set up to download the maximum amount of work and in the case of a 64 core machine and it is finishing tasks in 2.5days means it could be downloading 512 tasks which when a batch only contains 2,000 or less is a substantial chunk!

Others discuss users holding onto work, is this because they have their system in burp set to store for example 20 days of work or more?

Yes, many users have their machines set up to download the maximum amount of work and in the case of a 64 core machine and it is finishing tasks in 2.5days means it could be downloading 512 tasks which when a batch only contains 2,000 or less is a substantial chunk!

Can this be stopped in the same way other projects do? Or should I say is this going to be stopped?

Can this be stopped in the same way other projects do? Or should I say is this going to be stopped?

I have no idea. I don't know how other projects manage it. WCG always seems to have work 24/7 as do the other ones I have looked at. Which other projects have regular periods with a lack of work?

For CPDN, it probably isn't a problem as these multi-core machines will get through the work in a reasonable length of time. In some ways, perhaps it is perhaps machines like my atom netbook with a cpu right at the minimum 1.6GHz which will take 50 days or more to get through the larger of the current tasks that are the problem. I don't know how many minimum spec machines there are attached to the project right now. The only real problem for the project with these machines would be if after downloading lots of work they were switched off a lot of the time but I suspect that is more likely with less task hungry machines.

I can understand it is a problem for crunchers who like myself are only really interested in CPDN but I don't know what percentage of the crunchers we make up.

Large batch 499 of EAS50/13 now available ...

Large batch 499 of EAS50/13 now available ...

Is it only for WIN machines (Weather At Home 2 (wah2) v8.24)?

Large batch 499 of EAS50/13 now available ...

Is it only for WIN machines (Weather At Home 2 (wah2) v8.24)?

WAH2 is listed on the applications page as Windows-only, here.

Large batch 499 of EAS50/13 now available ...

Is it only for WIN machines (Weather At Home 2 (wah2) v8.24)?

WAH2 is listed on the applications page as Windows-only, here.

Sorry for being unclear. I meant is it only v8.24 (WIN) in the hopper and is there a way to figure it out - besides hitting update on non-win machine? Both Linux machines answered no task needed :) so I couldn't determine is the batch WIN only or not.

No, there's no way of seeing what's in there. It's what used to be called a "lucky dip".


edit
Actually I think that it may be possible.
On the Applications page, I think that the wah2 near the "short" is just wah2, and the one at the bottom of the list is wah2-RI.

And the ones that I've received are V8.24, which are Windows only, and also the latest batch are all from the top of the Status page list, making them Windows only.

Thanks Les,
It would've been probably easier if those two WAH 2 categories have different names :) Next time if bottom WAH 2 has unsent it would mean non-win models are in the hopper.

I'm sure that I asked for the name change ages ago, but I'll try again. :(

Is it only for WIN machines (Weather At Home 2 (wah2) v8.24)?

I have just checked my one machine running Linux and it has been told there are not tasks available. - 1.6GHz netbook resurrected for the purpose of checking. It does still have a task on it from last year when I thought it broke but having risen from the dead I will let it finish.

Edit:Not 100% sure now - they may all have been gone when linux machine updated.

... and now lots of AFR50/13, also WAH2 v8.24, so I assume Windows-only.

... and even more EAS50/10. Likewise WAH2 8.24.

Hi Bernard

The wah2 had two different paths; the original wah2_ri was region independent, whereas the wah2 was tied to a fixed region. This has now been changed, so that both are region independent.

So the one with the earlier app date is now obsolete, and will be removed sometime.
Putting everyone back to only Windows.

Thanks Les, I got back to WINE so all crunching now.

These new work units (eas50 and nawa25) can take reboots better. I haven't had a problem on my Win7 64-bit machine yet. This is my main PC, so I have to reboot it once in a while.
My dedicated machines have been converted to Linux (Ubuntu 16.10), and don't need reboots very often, but I can't do CPDN on them now.

My dedicated machines have been converted to Linux (Ubuntu 16.10), and don't need reboots very often, but I can't do CPDN on them now.

All my machines are Linux (Whatever you see when you look at the computers on this site) running tasks under WINE. It seems to have improved massively since I tried it a few years ago though getting Skype to work still eludes me.

Hummm. That could be interesting, though at the moment one machine (four Haswell cores) is enough, as long as it works. Thanks.

@Jim1348 and Dave Jackson and Les:
My experience with Linux on older machines has been unexciting,and,
because of that and a lack of WU's in late November I suspended CPDN and turned my attention from Oxford U to Cambridge where the Computer Lab was busy with development of projects for RaspberryPi (www.cl.cam.ac.uk/projects/raspberrypi)
In January an article appeared in the Foundation magazine MagPi: "Share your Power With BOINC" !! (Tutorial by Rob Zwetsloot, magpi.cc / @TheMagPi)
RaspberryPi runs on a Linux Debian or Raspian version, so I loaded up, but got the disappointing Notice From (CPDN) server that "This project doesn't support computers of type arm-unknown-linux-gnueabhf".
Disappointing because I have just started receiving some WAH2's on a Windows machine (@ 75 Watts) and was hoping to transfer to Raspberry Pi mini's (@ 5-6 Watts).
Q1: Do you experts think that CPDN Oxford (who are also involved in
raspi project work) could hack the code to operate on ARM cpu's?
( the ARM will process other BOINCs, including SETI)
Q2: Will there be a more constant flow of WU's for both Linux and Win10 ??
John Hopkinson

---

Q1: Do you experts think that CPDN Oxford (who are also involved in
raspi project work) could hack the code to operate on ARM cpu's?
( the ARM will process other BOINCs, including SETI)

This has been covered in a number of posts over the years. The main problem is that the cpdn code is in FORTRAN over a million lines of it and it does not belong to Oxford Uni but is licensed from the UK Met Office. Because it isn't open source, hacking it in that way would be illegal. I also suspect that there is no one at Oxford with the spare time to do that work even were the Met Office to grant permission.

I posted a question about this early in the Pi's life and got this answer which was repeated the other day when someone asked about being able to run CPDN work on Android phones.

I personally think this is a great pity but unless the Met office decide to release the code I don't think there is much anyone can do about it.