Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
World Community Grid Forums
Category: Completed Research Forum: The Clean Energy Project - Phase 2 Forum Thread: Appeal to CEP2 staff to react to wrong wu behavior and volunteers vote |
No member browsing this thread |
Thread Status: Active Total posts in this thread: 18
|
Author |
|
SekeRob
Master Cruncher Joined: Jan 7, 2013 Post Count: 2741 Status: Offline |
Runs fine on my Q6600 stock 2.4Ghz, BOINC data-dir on 64GB 3.0 USB stick [+ a 15 Euro adapter card and some bios updating to support the PCI half size slot use].
Hibernating perfect, even under the dog OS going by the name of W10-Pro [Hilarious], and coming out of Hibernation makes big resume difference... very slick resume, loss-less. VM is a superfluous solution lest you just want do an instant and take time to zero-power using hibernation mode. Not sure if VM has shutdown-backup functionality, but see no need to add further overhead. Not exactly original idea as this has been discussed multiple times and probably works too on the appropriate hardware. How late it is for CEP2, don't know, but symlinking instead of unzip>copying 6700 files to each job slot is top of my wishlist. The it becomes multi-concurrent runnable in a ram-drive on an 8GB system. BTW, Iguess the 'SP2' was another poor memory [typo] and it was SP3...the only thing MS supports FAIK. A brief hack described on the internet actually restores update maintenance through 2019 or so, but who bothers with this OS, lest there's absolutely no way around. Same ancient HW and Linux and you got years of crunch-fun ahead, at no charge. |
||
|
Jim1348
Veteran Cruncher USA Joined: Jul 13, 2009 Post Count: 1066 Status: Offline Project Badges: |
How late it is for CEP2, don't know, but symlinking instead of unzip>copying 6700 files to each job slot is top of my wishlist. The it becomes multi-concurrent runnable in a ram-drive on an 8GB system. I don't know if VirtualBox has built-in ramdisk capability, but that would be the easy way to it. Of course, it would set a larger minimum memory requirement, but you need something to play in this game anyway. [Edit 1 times, last edit by Jim1348 at Nov 27, 2015 6:00:19 PM] |
||
|
hristo_h_m
Cruncher Joined: Dec 10, 2008 Post Count: 22 Status: Offline Project Badges: |
I am graduated in physics and I am well aware of software which tackle such problems. To sume the picture it is more smoke than fire.
----------------------------------------q6600 is good chip and q9450 is his scaled version at around equal performance and thermal range. My operating is server 2003 64 bit and must handle intensive IO. Personally I like win98 even win95 they just fly under 100 Mhz. Even Linux can hardly beat this today. I bet that even pentium dual core will do the job if there are good programmer on the other side. And not just about CPU but overall, believe me. To add in one post. Boinc already is using VirtualBox to isolate the application, and ask regularly for checkpoint but the other side answer as rogue or zombie process. So as you can see crunchers lead the way not just with ideas but workaround to buggy program. Qxxxx line of processors lack EPT flag which means it is not suitable to run boinc into your Virtual box, which run again in virtualbox... Simply because this program has bilions of page fault and overall big rate of page fault index. It is pity that someone of King's men not answer to this post topic. For now i am waiting. [Edit 1 times, last edit by hristo_h_m at Nov 27, 2015 7:07:02 PM] |
||
|
deltavee
Ace Cruncher Texas Hill Country Joined: Nov 17, 2004 Post Count: 4842 Status: Offline Project Badges: |
It is pity that someone of King's men not answer to this post topic. For now i am waiting. hristo_h_m, We have been working on this project for over five years now. Everything you mention above has been discussed and rehashed ad infinitum. The "King's men" have responded and have explained and have joined in the discussions. This project is what it is. Lot's of members have quit in disgust and anger after not being able to effect change in this project. You will not get much sympathy, only time worn and patient explanations. |
||
|
Jim1348
Veteran Cruncher USA Joined: Jul 13, 2009 Post Count: 1066 Status: Offline Project Badges: |
You will not get much sympathy, only time worn and patient explanations. You found a way to put it much more nicely than I was able to come up with. |
||
|
theodolite
Advanced Cruncher Joined: Apr 11, 2014 Post Count: 119 Status: Offline |
The thing I find most annoying in this community is a Johnny-come-lately-to-the-party who trolls into the forums espousing demands, and most often one who has failed to do any research in the forums for past information. The truly special ones toss insults right and left like a MAC-10 that is out of control. They usually don't even deserve a single response, never mind a thread.
The really annoying ones throw in errors (l**s), such as suggesting that a Q9450 is equivalent ("equal performance and thermal range") to a Q6600 chip when Passmark clearly states that the Q9450 is 27% faster and has a 10% better TDP. This example is one of the worst. a member for nearly 7 years who has begrudgingly devoted the time and effort to contribute about .25 work units per day, and now threatens to quit. Does it cover it if I say 'will be sadly missed' or am I overdoing it? Guess what we are waiting for? |
||
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7545 Status: Offline Project Badges: |
SekeRob*:
----------------------------------------I did not mean to imply that CEP2 would not run on the Q6600, but that it is probably not a good fit if the machine is only on for fairly short periods of time. My Q6600 runs Windows Vista and has accumulated more than 16 years of cpu time on many of the WCG projects, but I do not have the luxury of leaving this machine on 24/7. It is reassuring to know CEP2 does run acceptably on this platform. I agree the Q9450 is faster than the Q6600, I just did not calculate by how much. I appreciate the feedback. Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
hristo_h_m
Cruncher Joined: Dec 10, 2008 Post Count: 22 Status: Offline Project Badges: |
I was thinking to not post this but anyway this is my opinion and it is opened to discussion.
All processing tasks until now from mentioned comp (but who knows maybe this is true for the last five years) are for molecules which are not charged. You see, the current is a flow of electrons, which solar cell need to provide. It can happen that one molecule looks dumb but when this molecule is integrated in suitable matrix the whole thing can rocks, because solar cell is not just molecule but solid state device. There is not molecular orbital anymore but solid states, bands and ... read the books. As you see this is a simple argument. Another simple argument. The project screens organic molecules for the light to energy transform. I think that the end result of this project would be the very famous molecule which gives color of the green plants. But I think that this project is very far from this yet. Someone knows the efficiency of this conversion? So what is going on. Q-chem Inc is commercial software which for now using free resources to tunning his code to produce something that has meaningful output. That's all. I think this is enough for the big name of this project which someone are proud that it is running for five yards. I can assure you that for the whole ten yards there will be no difference between someone with 1 wu at month and another with 100 wu per day and badges and so on and so forth. Oh, I would miss to reply to the second lieutenant about why posting here my so called first post. https://www.youtube.com/watch?v=dijVbM9DpxU Johnny is very calm person, in most cases he feels nothing. He feels nothing for seconds and nothing for unix. But he does laughs at all to see you picking first things first. So he used given link to facebook where everybody can see happy smiling people. These people have access to some big computing nodes so called mainframe and they shines. But desktops are very different from nodes, and smiling happy people not want to massively rewrite software to fit desktops so they make here and there patches. And more importantly they are reluctant to give openness to the source code to the point that they are ready to lose many desktops but not to do what is need to be done. So let me ask you what to discuss with such people. They already made their minds. And that they not responded to bugs can be understood easily from this point of view. And why they don't publish gross time for running the project instead the net time? So everybody see for itself. And you? You volunteers, will fight long and hard for a bit of colored ribbon. And to finish. Most good Q6600 are overclocked, good chips are cold chips. G0 stepping are of particular interest... As everybody knows ( it means if you don't it is meaningless to continue the read) Every E2140 run at 333 the math gives 2666 ~ Q9450. The whole line of core processors depends very nonius from the size of the L2 cache because they are very aggressive in prefetching. These are things around, comrades. |
||
|
|