Posts by Christoph Jansen

41) Message boards : Number crunching : BOINC isn't helpful to science? (Message 32362) Posted 9 Dec 2006 by Christoph Jansen Post: I would tell "HIM" this: The quantum mechanical foundation of chemistry and physics is what brought us all our modern devices and technologies, be it lasers, solar cells, safety systems in cars and planes, PCs, MP3 Players, Microwaves, MRTs and CTs and any other advanced medical device or whatever has a silicon chip inside. It all depends on us being able today to do the right material design. Today, we do in principle understand any chemical element and its behaviour and reactions on a mathematical basis. On the same basis we do in principle know how to calculate the structures of biomolecules, just that they are a lot harder to do than "simple" solid state chemical or physical calculations. As SOAN said, there are a lot of medically relevant projects Rosetta is helping with its research. It is definitely at the front of what quantum chemistry can do and that always meant and means doing valuable research that will aid or bring forth future technologies.
42) Message boards : Number crunching : Report Rosetta screensaver problems here (Message 32263) Posted 8 Dec 2006 by Christoph Jansen Post: Applications run flawlessly. But when opening the screensaver the graphics sometimes becomes unresponsive and has to be terminated from Task Manager. It then restarts the model and all is well. Right now I tried to solve it by just waiting for the graphics to respond again and that is what happened: 08.12.2006 09:35:45\|rosetta@home\|Unrecoverable error for result FRA_t103_test_LARS_constraints_oldfrags_barcode_enforced_hom001_12_IGNORE_THE_RESTS_00001_0010849_0.pdb_1429_ 272_0 ( - exit code -1073741819 (0xc0000005))
43) Message boards : Number crunching : AMD 64x2 dual core computer not showing two CPU's (Message 32236) Posted 7 Dec 2006 by Christoph Jansen Post: Hi Clyde, go to the device manager, open the "computer" in the list and it should read "ACPI multiprocessor PC" or similar. If it doesn't, click on the "drivers" tab and click "update driver" or similar (I've of course got the German version here, so I am not sure about exact labels). You can now choose between two HALs, one for singleprocessor and one for multiprocessor machines. Select the multiprocessor and click ok. Then you need to restart the PC. Regards, Christoph
44) Questions and Answers : Unix/Linux : Rosetta screensaver crashes Ubuntu 6.10 (debian) (Message 32235) Posted 7 Dec 2006 by Christoph Jansen Post:
45) Questions and Answers : Unix/Linux : Rosetta screensaver crashes Ubuntu 6.10 (debian) (Message 32206) Posted 7 Dec 2006 by Christoph Jansen Post:
46) Message boards : Cafe Rosetta : Moderator contact thread archive (Message 32061) Posted 4 Dec 2006 by Christoph Jansen Post: Can somebody please remove this item from the Q&A board? The visible text is so long that you cannot see the right "Answers" bar indicating the number of answers and the time since they were given anymore. To see the effect you need to go into the board rather than the post itself. Very annoying when trying to browse that board. Thanks.
47) Questions and Answers : Windows : Any speed difference in versions of Windows? (Message 32060) Posted 4 Dec 2006 by Christoph Jansen Post: Hi, an addition: I just found in another discussion on a team board that Win98 is no longer officially supported by BOINC. This does not mean that BOINC won't run, but there is no specific debugging anymore for 98. So if any new feature of BOINC will not work under 98 nobody will officially care about that any more. Sorry I did not know that, I was using 98 myself till a short while ago and did not notice the change in supported platforms. But I thought I'd better mention it here.
48) Questions and Answers : Wish list : Glossary (Message 32002) Posted 3 Dec 2006 by Christoph Jansen Post: Hi Edwin, welcome to Rosetta! You might use the inofficial BOINC Wiki you can find here. Many terms and basics of BOINC are explained in that collection. Regards, Christoph
49) Questions and Answers : Windows : Any speed difference in versions of Windows? (Message 31956) Posted 2 Dec 2006 by Christoph Jansen Post: Hi, I do not think there is a measurable difference. I had a machine that had Win98 and 2000 installed besides each other and benchmarks were the same on both. I did not notice any difference in the credits I got per hour when running one or the other. As your 98 machine is a PIII with comparatively little memory I'd stay with 98 if there is no other need to upgrade. I think it has a smaller memory footprint. That said, there is an important exception if your machine is not just used for some hours a day but for longer times on end: If you would like to have your PC run 24/7 for days or weeks then at least 2000 is a must in my opinion as it gives you way less problems in continuous operation. 98 is not the star of stamina in that respect, not to say it can be a nightmare when trying server-like operation. Regards, Christoph
50) Message boards : Rosetta@home Science : amyloid fibril structure prediction (Message 31852) Posted 30 Nov 2006 by Christoph Jansen Post: Yes, that is correct, and she probably contributed much more to the succes of Watson and Crick than most sources tell. She did all the experimental work that made the determination of the DNA structure possible, from preparation of DNA samples in their A and B form to taking the best x-ray images ever achieved until then. And she was a very good interpreter of features that have to underly x-ray data too. Watson and Crick were, apart from getting access to the famous photograph 51, lucky to find an unpublished article which gave them clues Franklin and her colleagues had overlooked - and pretty quickly solved the problem. It would just have taken some time to have that done by somebody else, maybe Franklin could have done it, as she already was certain it had to be a double, triple or quadruple helix. And she found out that the bases had to be inside with a sugar and phosphate framework outside. Watson and Crick had that from the article and as they were sure it was a double helix solved the problem rather straightforward by accounting for the bonding of the bases to one another based on hydrogen bonding. It was like a knot that had gotten unravelled and everything fitted. Watson and Crick though insisted for a long time they did not know those unpublished data. The reason was that the King's College published their own article to at least get the credit for the experimental work, if not the discovery itself. By denying to having known the facts stated there they made it look like it was all their own conception. Watson later confessed he did know the data in his book "The Double Helix" in 1968 and even Crick admitted that that was true after the publication. Regarding the Nobel Prize, Rosalind Franklin died in 1958 and Watson and Crick got the prize four years later, six years before the book. Had Waston and Crick told the truth about that article while Franklin was alive, she might have gotten larger credit. But her part in the story was underestimated by withholding the truth, to state it diplomatically.
51) Message boards : Rosetta@home Science : [idea] forum again, please make it easier (Message 31831) Posted 29 Nov 2006 by Christoph Jansen Post: Hi, I mostly drop in there when I am on the forum, but as I am a Windows user only I cannot say much regarding other OSs and have little knowledge on certain network issues. And it has been pretty quiet there recently. [Although I do not think the reason is that there are no questions anymore. I rather think that even Newbies pretty quickly realize that the Q&A area is rather one where warm desert winds may flow unheeded, sometimes rolling some dead bush along.] I would also like the whole Q&A thing made more like a "real" forum (or several fora) with the same hints on new and unread threads.
52) Message boards : Number crunching : OT: A closer look at Folding@home on the GPU (Message 31758) Posted 28 Nov 2006 by Christoph Jansen Post: Hi all, just a thing that came to my mind when browsing for current graphics cards' prices: It looks very much like the focus of crunching on GPUs concentrates on ATI based solutions, which is also picked up in this thread. How does that fit with the problems that quite some users run into when using ATI cards with BOINC screensavers and graphical frontend in general on several projects? There might be a train that pulls out of the station and leaves BOINCers behind, at least to some degree. Is the problem rare enough to just neglect it or are the people reporting these problems just the tip of an iceberg, most of which is concealed by just terminating BOINC without further comment? Just intended as a cautious question, as I do not have the slightest notion about problems associated with trying to reconfigure BOINC to cope with that and the value of it in comparison to the effort. [Note]Of course I do not think that is a generic train of thought only realised by me. I am sure the programmers themselves are quite aware of all that. It is just that it has not been mentioned in this thread before and I wanted to throw it in to the arena.
53) Message boards : Number crunching : CASP 7 Results are in! (Message 31722) Posted 27 Nov 2006 by Christoph Jansen Post: Hi, another place to look at is here. [EDIT]OOops, a little caveat: that page seems to be very big in transfer volume as it loads pretty slow here, so if you do not have DSL you might run into problems. Maybe it is just a slow connection, but I thought I'd rather warn you.[/EDIT] What you see below each target is its structure and, the interesting part, a plot with orange lines. The upward axis is the deviation from the original structure in Angstroms, the horizontal line shows the percentage of residues within that deviation. The Baker team is always highlighted (lowlighted in that case?) as a black line in the link I put here. The best possible plot would of course be a flat line at 0 Angstroms = perfect match for all percentages (or rather just a dot at 0 Angstroms / 100 percent). In reality you see curves for which a simple rule is valid: Lower is better. So everybody that is lying farther left and upward has performed worse than the Baker team ( = has got higher deviations for the same percentage of residues), everybody who is further down and right has done better ( = has lower deviation for the same percentage of residues). Rosetta is in the lower right portion for a lot of targets, sometimes it looks like it is best overall and there are also some "dropouts" that are rather a little worse than average. But on the whole I'd say it looks pretty good. [Edit2] Look for the T0299's and you're in for some eye-candy for Rosetta. Very nice curves there.[/Edit]
54) Message boards : Rosetta@home Science : Can computional biology play an important role ? (Message 31687) Posted 26 Nov 2006 by Christoph Jansen Post: What I wonder is: with that range of variation, how much variation occurs in ourselves compared to our parents and siblings? When you take a genetic "fingerprint", you just look exclusively at certain, reliable genetic positions ("gene loci") that will, under normal conditions, not differ from the genetic material inherited from our parents. So variation is kind of excluded systematically and the method does not tell anything about true variability. Does the variation take place in everybody by individual mechanisms, e.g. by copying errors during the production of our germ cells, or is it strictly a question of inheritance after random and rare mutations that add up in your bloodline? To clarify that question one would have to make a screening with numerous groups of related people and then sort it out. Probably data that need massive computing power. Fascinating new results.
55) Message boards : Number crunching : Sony: Consoles can aid medical research (Message 31588) Posted 22 Nov 2006 by Christoph Jansen Post: Hi all, not sure if that has been mentioned anywhere else, but though: Sony describes an Other OS installer on its online help page. And they also introduce the first PS3 capable Linux Kernel, see this page using which the Fedora Core 5 Linux distro for PPC can be installed on a PS3. To the pros among us: what does that mean regarding applications that run on this platform? Are they comparable to what runs on other Linuxes or does the PS3 require e.g. a special version of BOINC?
56) Message boards : Rosetta@home Science : Sharing Account Key (Message 31524) Posted 21 Nov 2006 by Christoph Jansen Post: Hi Eric, you can run Rosetta (and any other BOINC project) on as many other PCs as you like. Just install BOINC on the machine and in BOINC manager select "Tools" => "attach to project" and here you answer "attach to an existing account". Welcome to Rosetta, by the way. Regards, Christoph
57) Message boards : Rosetta@home Science : amyloid fibril structure prediction (Message 31253) Posted 16 Nov 2006 by Christoph Jansen Post: It's more challenging to derive detailed structures from fiber diffraction data but perhaps this could be done. But you can see features in the data that correspond to structural repeats -- eg the 4.8A repeat that corresponds to stacked beta-strands. Hi phil, a good example is the famous "photograph 51" from which Watson and Crick derived the double helix structure of the DNA: Photograph 51 This diffuse pattern of a non-crystalline specimen was enough to tell that the DNA is a helix by the equidistant blackening and the angles involved. If you look close, you see that there should be 5 such patterns from the inside out. There are three [the center does not count, it is referred to as zero'th order] clearly visible and the fifth, most outward one, is very faint. But the total lack of the fourth told Watson and Crick that they had to deal with a double helix, one of which is displaced by 3/8th of a turn against the other. That seems to be little information, but if you know a sequence and thus how the thing is composed in general, that can give you a lot of insight into possible structural motifs. For Watson and Crick it was enough for a breakthrough.
58) Questions and Answers : Web site : Server Status (Message 31249) Posted 16 Nov 2006 by Christoph Jansen Post: Ho oddess-e, no, they do not need to run all. A short explanation of what each server does is given at the bottom of the Server Status Page. Regards, Christoph Oh, I forgot, sorry: Welcome to Rosetta! ;-)
59) Message boards : Number crunching : Report problems with Rosetta version 5.36 (Message 31134) Posted 14 Nov 2006 by Christoph Jansen Post: And the same here, all "DOC" WUs seem to just not want to validate. And I noticed that the project's RAC has dropped from 38.5 to just 33.5 TeraFLOPs today. Probably it is the effect of all those WUs erroring out in validation? [EDIT] Yes, I've browsed through the results on some of the top machines. All those WUs that errored out with "validate error" I saw since Sunday were "DOC" WUs. So it looks like somebody ought to look into that, as it is also lost capacity to the project if they are useless -or lost results if the validator just is wrong on those special WUs and results are worth looking at. Could we please get an instruction of what to do with those WUs?[/EDIT]
60) Questions and Answers : Windows : Frozen at 71.5% (Message 31049) Posted 13 Nov 2006 by Christoph Jansen Post: Hi TomFox, what do you mean by "freeze"? Does the "processor time" freeze or is it just the "progress" percent that freezes? If it is the progress percent, then there is no need to worry as that only changes when a breakpoint is reached or Rosetta has completed a decoy and starts computing the next one. Decoys can take from some minutes to hours, depending on processor speed and protein size (the number of amino acids involved). If the processor time also freezes then your WU might be defective or cause trouble on your special configuration (never happened to me, but other people report it). Then you'd best abort it, load a new one and look if that one finishes correctly. If you still have errors, ask again. An overview over what PC you have would help, especially type of CPU, Mainboard, Graphics Adapter, Memory and of course OS.

Previous 20 · Next 20