Actually it does work but we would get ALOT fewer credits for each task at most projects which is why the nickname

Actually it's usage at SETI@home proved it doesn't work as optimized applications and hardware getting faster and faster were both pushing the credit down for the same amount of scientific work done, i.e. users were "punished" for buying better hardware and using applications, that were able to use that hardware efficiently. IIRC there was the sugestion to run beta for a while with completely unoptimized applications (i.e. only i386 instructions, no MMX, SSEx or whatever and no GPUs) and see where the credit than goes. Unfortunately it never happened, but they expected 2-4x higher credit or so.

---

Hello, Tom,

My apologies. This is all Greek to me. Does that mean my laptop will no longer be crunching numbers? I noticed there have been no more tasks running.

Is there something else I might do to contribute?

John

the n-body needs more than 1 core as it is multi core app and it currently does not run on mac os

Hello, Tom,

My apologies. This is all Greek to me. Does that mean my laptop will no longer be crunching numbers? I noticed there have been no more tasks running.

Is there something else I might do to contribute?

John

John,

Go to your account page and the Milkyway Preferences section and select to receive Milkyway@home N-Body Simulation tasks. You will then be eligible to receive work again.

Bill F

Actually it's usage at SETI@home proved it doesn't work as optimized applications and hardware getting faster and faster were both pushing the credit down for the same amount of scientific work done, i.e. users were "punished" for buying better hardware and using applications, that were able to use that hardware efficiently. IIRC there was the sugestion to run beta for a while with completely unoptimized applications (i.e. only i386 instructions, no MMX, SSEx or whatever and no GPUs) and see where the credit than goes. Unfortunately it never happened, but they expected 2-4x higher credit or so.

I don't understand what you mean. If we both buy the same video card, and I make the effort to run an optimised app and you don't, presumably I get more credits. What's unfair about that?

---

The above was double spaced between sentences, I apologise for the forum software ruining my post.

Really? I have some slow 1 core machines that I keep running for sentimental reasons they always ran “Separation”. So they won’t be able to run “N-Body” ?

They might. I just instructed my 4 core machine to only use 1 core for Nbody, and it obeyed me. I put the below in app_config, but you don't need to, since you only have 1 core. Just try getting them and see what happens. They're fine from 2 to 16 cores, so I don't see why they can't use 1. If the server refuses on the grounds your CPU is 1 core, I guess there's a workaround somewhere but it might involve some tinkering.

<app_config>
    <app_version>
        <app_name>milkyway_nbody</app_name>
        <plan_class>mt</plan_class>
        <cmdline>--nthreads 1</cmdline>
        <avg_ncpus>1.000000</avg_ncpus>
    </app_version>
</app_config>

---

The above was double spaced between sentences, I apologise for the forum software ruining my post.

Really? I have some slow 1 core machines that I keep running for sentimental reasons they always ran “Separation”. So they won’t be able to run “N-Body” ?

Yes they can run the NBody tasks as you have Windows pc's, just go in and select the NBody tasks instead of the Separation tasks, now deadlines could be a problem, they probably were before on the Separation tasks too so be sure to keep the cache small to start with or you could end up with a few hundred tasks you will never finish before their deadlines.

Yes they can run the NBody tasks as you have Windows pc's, just go in and select the NBody tasks instead of the Separation tasks, now deadlines could be a problem, they probably were before on the Separation tasks too so be sure to keep the cache small to start with or you could end up with a few hundred tasks you will never finish before their deadlines.

I get that problem with 4 core machines! Boinc isn't very bright. I have one machine which takes up to a day and a bit to run an Nbody on 4 cores. So for my 0.1+0.1 day buffer, it decided to download 100 tasks?! It predicted each one would take 20 minutes, who knows where it gets it's silly ideas from.

And on my gaming machine, which I play games on (and therefore pause Boinc) for 3 hours every day, it leaves panicing until there's 2 hours left on a task. Then I play a game and it misses the deadline by an hour. Or it panics and runs 30 cores of work in 24 cores and slows the GPU to a crawl.

---

The above was double spaced between sentences, I apologise for the forum software ruining my post.

What if you run game without pausing boinc?

Actually it's usage at SETI@home proved it doesn't work as optimized applications and hardware getting faster and faster were both pushing the credit down for the same amount of scientific work done, i.e. users were "punished" for buying better hardware and using applications, that were able to use that hardware efficiently. IIRC there was the sugestion to run beta for a while with completely unoptimized applications (i.e. only i386 instructions, no MMX, SSEx or whatever and no GPUs) and see where the credit than goes. Unfortunately it never happened, but they expected 2-4x higher credit or so.

I don't understand what you mean. If we both buy the same video card, and I make the effort to run an optimised app and you don't, presumably I get more credits. What's unfair about that?

Nothing, but that's not the issue, you should get more credits per day. But because even the stock applications were optimized to some extent AND lots of people were using even better optimized apps, CreditNew "compensated" all those optimizations, and instead of people running optimized apps getting more credits per day for doing more work per day than they could if using a simple i386-app, everyone got less. CreditNew was simply not able take any optimizations into account, wether hardware or software, it simply assumed there was less work to do and gave less credit for the work.

---

I have some slow 1 core machines that I keep running for sentimental reasons they always ran “Separation”. So they won’t be able to run “N-Body” ?

As long as they are 64-bit (also the OS), they should be able to run them, like others said, just set your cache very low, perhaps something like 0.1 + 0.01, no app_info.xml needed. And abort all your Separation tasks as they are completely worthless and you might even not get credit for them.

Your Celeron won't be able to run them, the other systems have 64-bit OS and should run nBody.

---

What if you run game without pausing boinc?

If you crunch on GPU, the game might run slow, if you crunch only on CPU, probably nothing, the task will simply need a bit longer.

---

What if you run game without pausing boinc?

Then game get jerky and annoy me. GPU Boinc can also cause computer to crash. If I happen to spot something urgent running, I can reduce Boinc to a few cores instead of pausing it, but I don't always notice.

---

The above was double spaced between sentences, I apologise for the forum software ruining my post.

Nothing, but that's not the issue, you should get more credits per day. But because even the stock applications were optimized to some extent AND lots of people were using even better optimized apps, CreditNew "compensated" all those optimizations, and instead of people running optimized apps getting more credits per day for doing more work per day than they could if using a simple i386-app, everyone got less. CreditNew was simply not able take any optimizations into account, wether hardware or software, it simply assumed there was less work to do and gave less credit for the work.

I don't understand, how is it calculating your credits? If we are both given 1 identical task and run it on the same hardware, but you optimise your code, do we get the same credit per task because it's on the same card, or do we get different amounts because we took different lengths of time?

---

The above was double spaced between sentences, I apologise for the forum software ruining my post.

What if you run game without pausing boinc?

If you crunch on GPU, the game might run slow, if you crunch only on CPU, probably nothing, the task will simply need a bit longer.

Some games are affected by CPU Boinc. The games I play use up to 6 cores. If Boinc is using all 24, the game doesn't get as many cores at it should.

And my GPU hates doing two things at once. If I run two different projects on it at once, or a game and a project, or three tasks from one project, the machine locks up and I get coloured patterns on the screen. Strange as it's a good card otherwise. It's powering 4 screens without a problem and rips through Boinc tasks at the correct rate. It's a Radeon R9 Nano. I have cranked the power limit to +50%, but it uses all of that fine in Boinc without crashing. I had to hard wire the power lines to it as it melted those flimsy stupidly designed PCI-E connectors which couldn't handle 50% more current. I'm probably drastically shortening the life of the metal oxide semiconductor field effect transistors in the voltage regulator module.

---

The above was double spaced between sentences, I apologise for the forum software ruining my post.

Erlier up this thread i made the joke of how to switch of a server

What was that about "the simplest of things"
Now then , how could some dim bint like me `shutdown` a server ;-)
"where does that wire go" , Oop`s I have tripped over it and it`s gone kinda quiet .
I need a socket for the vacume cleaner . . . .
. . . . . . . . . . . . . . . my hair drier
. . . . . . . . . . . . . . . My phone needs recharged
The list is as long as a todo list for a husband .
And those are just the none destructive ways
A nice conductive cup of tea , coffee or that pint from the other thread .

BUT , a cleaner at Rensselaer Polytechnic Institute in New York , witch milkyway is connected with did it for real , switched off a cryo unit [fridge] and destroyed a lot of important work
https://www.bbc.co.uk/news/world-us-canada-66028401
It was`nt me i don`t work there .
sometimes a joke can creep up behind you from the past a give a nasty kick up the ass .

Thanks, everyone for the help. One of the 2 slow machines has a 64 bit operating system and the other one has a 32 bit operating system. From what I understand the 64 bit unit will run N Body (very slowly), but the 32 bit computer won’t? Even though I run a lot of machines, I don’t know a whole lot about them.

a cleaner at Rensselaer Polytechnic Institute in New York , witch milkyway is connected with did it for real , switched off a cryo unit [fridge] and destroyed a lot of important work

You'd think they'd have that on a UPS.

---

The above was double spaced between sentences, I apologise for the forum software ruining my post.

Thanks, everyone for the help. One of the 2 slow machines has a 64 bit operating system and the other one has a 32 bit operating system. From what I understand the 64 bit unit will run N Body (very slowly), but the 32 bit computer won’t? Even though I run a lot of machines, I don’t know a whole lot about them.

Yes, you need 64bit for Nbody. There are other astrophysics projects which work on 32bit, someone will add a list in a moment. I'm not sure how to tell.

---

The above was double spaced between sentences, I apologise for the forum software ruining my post.