Posts by jujube
|
21)
(Message 279)
Posted 21 Sep 2012 by jujube Post: the credits are a representation of my contribution to the project's science Credits are numbers that mean nothing and represent nothing. How can that possibly be? It's because you, me or anybody can steal as many credits as we want and some projects give out huge amounts of credits for minimal work. If I have 100 million credits you can never know whether I actually earned them or whether I stole them or whether I received them for a mere 10 hours CPU time from some renegade project that uses ridiculously high credit payouts to steal crunchers from other project. If I donate cycles of my CPU, electricity and my time to a project, I want this work being used in real science, not waiting eons to be used. Eons is a ridiculous claim. I doubt you'll convince anybody to change anything if you use ridiculous claims and exaggerations. Try sticking with facts instead. The fact is that it takes years for projects to produce any useful conclusions. It doesn't happen overnight, or in 2 weeks or even 2 months. The fact is a certain amount of data needs to be crunched and shortening deadlines doesn't speed up that process. It speeds up with more hosts attaching to the project, faster hardware, science app optimization, greater resource shares assigned to the project by volunteers and probably other factors I've not thought of. And a smaller deadline will certainly reduce this gap. Which gap are you referring to? The gap between the time you return your result and the time it gets verified and awarded credits? That has nothing to do with how much time the research takes. The fact is that after your result gets verified and awarded credit the result gets dumped into a database where it might sit for several weeks or months before it gets further analysis and it may be years before any of that analysis yields conclusions that are useful to anyone. It seems like you have little understanding of how real science is accomplished and how much time it requires. You seem to think that your result plus a matching result from your wingman is all the scientists need to call the newspaper and tell them they've discovered something new and fantastic. Nope! That ain't how it works. It takes hundreds of thousands of results plus many months of further analysis of those results and shortening deadlines doesn't speed that up by even 1 second. |
|
22)
(Message 277)
Posted 20 Sep 2012 by jujube Post: And the results would return faster, reducing the pending tasks. I think that's a misconception. The rate at which any given host crunches a given project's tasks is determined, in the long term, by the resource share allocated to the project, the host's hardware capabilities and host on time. In other words if I assign 20% resource share to this project and my host produces 100 billion (just for example) CPU cycles in a day then over the long term and on average this project will receive 20 billion cycles per day from my host (minus some cycles for my own computing needs and other factors but let's ignore those to keep it simple). BOINC might and often does give more than the allocated resource share to a project that needs extra CPU cycles to complete a task that is in danger of missing deadline (and perhaps in a few other circumstances too) but that's a short term measure. There are certain other "aberrations" that can and do happen but after "the flow" becomes established (in other words after BOINC learns what to expect from your projects regarding deadlines, task durations, etc.) any given host produces only so much work for a project on any given day and the length of the task deadlines has zero effect upon that over the long term, always over the long term. So what we notice wrt pending tasks, over the long term, is that they build to a certain point and then stay at about that level (don't bother quibbling over what about means cuz you'll get nowhere) regardless of the length of the deadlines. I will admit to the fact that anyone can and most likely will (to prove that the system will seem to fail in exactly the way I have said it will seem to fail, thinking they've proven me wrong) micro-manage their cache and other prefs until they have 6 trillion pendings but to them I say in advance RTFM and RMFP. And a smaller deadline would benefit us all, IMHO. How does it benefit anyone? Is it because we get credits sooner? Well, that doesn't happen for reasons I explained above. Besides, even if we did get credits sooner the credits are worthless. How does it benefit me to receive something worthless sooner? I mean suppose you want to send me a condom with a hole in it. The condom is worthless. How would I benefit from receiving the worthless condom in 2 days as opposed to 10 days? |
|
23)
(Message 269)
Posted 19 Sep 2012 by jujube Post: hi jujube. Can you quantify the difference? I didn't write the numbers down and now I don't remember. The reason I did not write them down is because they were no worse/better than the benchmarks I obtained on Ubuntu. Tonight I tested with Debian 32 bit and 64 bit. The 32 bit gave 3,023 and 10,880 but the 64 bit gave 3,640 and 22,513 which is very near the 3,770 and 22,028 benchmarks from the host OS. For me 64 bit Debian is the clear winner so I think I will phase out my Ubuntu VMs and replace them with 64 bit Debian VMs. Ccandido, do I understand correctly that 64 bit Debian did not give good benchmarks for you? In all cases the host OS on my system is Linux not Windows which some of you may have noticed and which might make you wonder why I bother with these tests since I can run the Asteroid app directly with no need for a VM. One of the reasons I test is because in order to run more than 1 T4T project task at a time one needs to use VMs. The other reason is that all BOINC projects draw their water (CPU cycles) from the same well so it is in the best interest of each project to keep the well (the hosts owned by us crunchers) clean, healthy and efficient, for the good of all parties. Maybe these tests help only a little bit but crunchers know better than anyone how every little bit counts. Debian repositories seem to make only BOINC 6.10.58 available which might be adequate for Asteroids. If a 7.x version is required then someone in the community will compile a 7.x and make it available or else I will publish instructions for compiling your own. It's easy. |
|
24)
(Message 265)
Posted 18 Sep 2012 by jujube Post: So far I've tried a Fedora 17 32 bit VM and a Fedora 17 64 bit VM and both gave very poor BOINC benchmarks. I shutdown all other CPU intensive apps while the benchmarks ran and there was little difference between benchmarks from 32 bit and benchmarks from 64 bit. Next I'll try a Debian VM and a Gentoo. |
|
25)
(Message 256)
Posted 17 Sep 2012 by jujube Post: Stop boinc and anything else that is processor intensive in windows before running benchmarks in the VM if you want accurate benchmark results. That doesn't help, I still get poor benchmarks in the VM. Normal benchmarks for my i7 2600K at stock clock (3.40 GHz) are 3,770 and 22,028. In the Ubuntu VM benchmarks are 3,227 and 13,884. Wow, a huge hit on the IOPs. I installed 32 bit Ubuntu in the VM along with XFCE desktop because that combination seemed to make the system more stable and responsive. Does 32 bit BOINC benchmark lower than 64 bit BOINC? I'm burning a 32 bit Fedora 17 Live DVD as I post this. I'll report the benchmarks as soon as I get them. I might have to try a 64 bit Fedora 17 VM for comparison. |
|
26)
(Message 253)
Posted 16 Sep 2012 by jujube Post: It's Linux only for now but there is no intention to keep it that way. The admin(s) are working hard to spread the joy to Windows and Mac as well but try to remember this is a relatively new project and not everything can happen at once. |
|
27)
(Message 251)
Posted 16 Sep 2012 by jujube Post: I have observed the same as ccandido... the benchmarks from BOINC running on a virtual Ubuntu machine are considerably lower than normal BOINC benchmarks. I can't understand why they would be lower on a Ubuntu VM but not a VM built from a different Linux distro but I intend to verify that claim. I don't like Ubuntu either so maybe it's time to investigate and perhaps find a better distro. Or maybe Ubuntu needs to be configured a certain way when installed as a VM. Maybe folks at the Ubuntu forums have some advice on the subject. |
|
28)
(Message 214)
Posted 8 Sep 2012 by jujube Post: I think Kyong's message 189 in this thread is saying there won't be a Windows app. Instead they are going to install a virtual Linux machine on Windows hosts (perhaps OS X and Linux hosts as well) so that all hosts will run the same application... the Linux application. That's just my interpretation of Kyong's post, I could be wrong, I suggest waiting for an admin to verify/deny. BTW, if it's true it won't require any work or knowledge of Linux on the part of Windows users. BOINC would install and manage the virtual Linux machine automatically. |
|
29)
(Message 198)
Posted 6 Sep 2012 by jujube Post: There are surface mount components but I've been etching PCBs at that resolution and soldering surface mounts for some years. I used to do it manually with an iron but there are much easier and faster ways. |
|
30)
(Message 192)
Posted 5 Sep 2012 by jujube Post: Its useful if you don't have the manpower to develop apps for every platform. The drawback (if you see it as that) is it only works with BOINC v7... The wrapper under development by the BOINC devs requires BOINC v7 and recent server code/patches. The wrapper in use at T4T works with BOINC 6.x and 7.x so 7.x isn't a strict requirement but it enables some nice features. |
Previous 20