Apr 19 at 6:30p Same story here: I am busy, the school year is ending, so this project takes a back seat. I found a webpage I haven't remembered the location of for a while: a climate modeler to-be. The project currently consists of some professor-types, a webpage, and a growing list of people willing to participate, but seems to be on course to eventually become something valueable. I like the idea of distributed computing a lot, and this project may be the next big thing. Apr 13 at 12:30a I haven't forgotten to update this site or anything, just I'm busy and there isn't a lot happening. Perhaps we can think of this as me training for the summer when I am probably not going to look at this page more than once a week. Apr 7 at 12:30a Well these BP6's sure do suck. If bishop hasn't already shown that to us, I'll tell you all again. After my BP6 still crashed (after a few days) in X at 2x 366mhz, well I got tired of that and just clocked it back up to 523mhz again and kept right on going. Woooo man X doesn't like that. Onto the news... I downloaded the 2.4 linux seti client and put it on skug and skud. Skud sucks to much to tell yet, but skug (my 458mhz 366 celeron) dropped its times by about half an hour per packet, so that it does them now in about 8:45, or just a few minutes better than each of Dan's P2 463's. Hum, we see again the benefits of the large cashes for SMP. My dual celeron at 523 does two 13:45 or so packets, compared to Dan's box which runs at 88% the clock speed and does them 53% faster. The bus speeds are 95mhz for me and 103mhz for him at those speeds, which isn't a huge difference. For the record, a dual 500mhz Xeon with 1 meg cashes should do two packets at 4:15 each, or 224% faster than my celerons at 96% the clock speed. Remeber, celerons and Xeons have the same core, and so this is all differences that cash makes. I do admit that not every application is going to be effected like that, but heavy multi-tasking should always cause the celerons to fall behind. Dan has also installed the new client, and we will see tomorrow if its helped over the past few packets. Those of you who run seti clients and read this mighnt look into grabbing the new client too. As of Tuesday, our 7-day output was 197 packets, as of today its down to 189. Apr 4 at noon Well we did in fact get passed by turbo, and he's pulling away. I grabbed a few more weekly output graphs from the team website: week ending the 2nd, week ending the 26th, and week ending the 19th. Too bad he doesn't label the "little" guys like us with an actual output number, but I happen to know that we average around 200 a week. I remember when WE were top 10 on our team, with less output too. In other news, I am sure anyone that reads this page from time to time has seen that the updates have slowed, and nothing new has been done. Get used to it. My classes are keeping me good and busy. Over the summer I except I will find the time to play with this page (for fun, I don't exactly expect tons of hits, especially during the summer when everyone is away). This page will also almost certianly exist next semester, and I will probably launch a campaign to get everyone helping out who can, and then enumerate them (right now, I really dont have a clue who's doing what for seti outside me and Dan). This semester we set new records for sustained performance, and I very seriously expect to do better next semester, especially with people upgrading and finding that they have more than enough power to throw some at seti. But for now, things are gona be in standby. I will still do news updates. Mar 31 at 11:45p Well we broke 7 years yesterday, or today, or something. Right now we're sitting at 7.05 years of processing time. Which is a lot. Other than that, our fellow Ars Technica team member Turbo is 39 packets behind us, and closing, of course. Its only a matter of time. Mar 30 at 2:00a We did exactly 200 packets the past 7 days, for an average of 28.6 per day, and our average time keeps on dropping and is now down to 13:27:46.4. Haven't recruited anyone. No news. Except that skud is now running 2 setis again, after it crashed again last night and by doing so pissed me off, so no more nice me thinking that low work load is good for poor little skud and its fragile little BP6 motherboard. Damn thing. Of course who am I to complain, since bishop has had way more problems with his BP6. Anyway, more news: ort isn't running seti these days cause windows screwed his system over reeeally well. Oh and twig across the hall installed windows 2000 and should be burned at the stake. Here's some stuff that makes me happy though: prices on some Xeons from pricewatch. First, why Xeons, you ask? Well, cause a 1-meg Xeon goes through packets faster than a equal-clock Mac G4 (with a 1-meg L2). A dual 500mhz 1-meg Xeon can do almost 12 packets a day, more than twice what our local dual P2-450 can pull, and if we had one of those in our team now, it would be doing 42% of our weeklly 200 packets by itself. Two processors. I don't think anyone can argue with that kind of performance, which is not going to just show up it seti. Sure, not everything is gona go faster than a standard dual P3-500, but some things are really gona go like stink. For the record, Quake will not see that speed up. But about the prices (from various sources, bound to change, may be pulled from servers, go look youself)... 400mhz 1-meg for $279, 450mhz 1-meg for $425, 450mhz 2-meg for $499 (holy crap! 2 megs L2!), 500mhz 1-meg for $500, and from there its into the thousands. But those are some nice prices... very very nice. With SMP, large cashes are king, (more cash=less bus traffic) and these suckers have all the L2 you need. The 2 meg 450 is really neat... thats no less than 8 times a new Pentium 3, or 16 times a celeron. With the direction that the new P3's have gone, 256k fast L2, SMP may not be as efficient as the older P3's and the P2's that had 512k of slower L2. As we have seen from dual celerons (with 128 of fast L2), if the L2 is too small we start to see the bus speed become a major limiting factor. The dual P2-450 around here is faster than my dual celeron, even when its a 450mhz vs 533mhz battle (4.5 x 100 and 5.5 x 97), and sometimes its a lot faster. It sees virtually no drop from running 2 setis vs one (a 100% performance increase), whereas my celeron sees only a 40% or so increase in overall seti production running 2 vs 1. Some other applications will likely show this also, although not all will. In any case, if you wanted to, you could get a nice dual Xeon for $2000, including everything but a monitor, I bet. Mar 28 at 2:00a Well we're 17th on our team now, having passed Panders. Of cource Turbo is still back there, along with a horde of other nasties who are gona toast us some day. On the positive side of things, our seti effort is approaching 7 years of time... yup, we've done some good work for the cause. Mar 26 at 1:30a Our 7 day average packet rate is 27.4, which isn't too shabby. We could do better, and we have. We are 18th on our team now, and that Turbo is still catching up, although we'll possibly hold him till school gets out. Then we die. Average time per packet is still dropping a little, and is down to 13:29:06.3. I'd also like to say again that our team is really compeditive these days. Check this out: weekly output of the 30 top members of our team. If you don't see us at first, just keep looking down the list. We were top 10 when we joined last semester. I sat down and did some figureing, and it turns out that the linux boxes in this room I'm in actually account for 40% to 45% of our daily production. They also yield an average time of about 11:30 per packet, which is brought down a lot by whoremachine and modzer0 (that was not a complaint). If we were to inlude seamonkey, our linux running G3, linux accounts for around 55% of our packets, and it turns out that we do have a few more linux boxes on it, so we can say very safely that we are a mainly linux group. Does it impress anyone that hal9000 accounts for a little less than 20% of our packets by itself? And we're not just talking daily packets, that box is also the first box to run this stuff. By now, its got to be worth some 900 packets by itself. Mar 23 at 3:00a Well I'm up too late... gona get bad grade on paper I gotta write... but anyway I did figure out how to do some stuff with the gimp that has kept me up. Dunno why I link that page, I've never actually used it. I should, I guess. Ok, the update to the main page is done. Other than what looks like poor quality in the red background .jpg's, we're in busness. Small news post: our packets are hovering just below 30 again, and somehow when I checked the stats tonight big-bad PGS hadn't even passed us (tied). Oh, and OoklaTheMok is gona get us soon too. Like I've said before, when we joined the team it was nothing like this... its really picked up. Mar 21 at 1:45a Well we are gona get passed by the really fast moving PGS tomorrow, and will then be 17th, with Turbo still slowly gaining. We are still gaining on Panders, but we've still got 148 packets to go. Our packets have been suffering since the break began, though. We will be launching a little "couterstrike" to keep ahead here... the plans are laid.... hehe. I had been noticing our average time was dropping like a stone this past week, and I sat down a figured out that we've been doing packets in an average of 10:41:41 the past six days, compared to our current overall average of 13:31:26.8, which compares pretty well to just 7 days ago and it's 13:37:11.6. So anyway something has happened, either to our slow computers, or our fast ones. Not sure which, since it only started up towards the end of spring break. Mar 19 at 2:00a If the day before yesterday's good news was that Turbo only gained 4 packets in two days, and yesterday's bad news was that he gained 30 something in one day, then its fitting that today's good news is that he only did 2 packets. Total. Mar 18 at 3:30a Well if yesterday's good news was that Turbo only gained 4 packets on us in two days, today's bad news is that he just gained 30-something in one day. Oh well. Mar 17 at 2:30a Well the good news of the day is that Turbo only gained 4 packets on us in two days, so he's not beating is that bad. It appears his production dropped off at the start of spring break also, kinda odd. I was also noticing that our average time has dropped down to 13:35:03.9 per packet, which is the lowest its been since the 18th of Janruary, and I guess that's a good thing, although it doesn't matter too much. For the record, the lowest averge we ever have had is 12:48:47.9 per packet, on the 27th of October, 1999. With one swift move by Bishop, we have grabbed up (back) two Linux boxes: one is the now one-processor but stable Ion, and the other is the computer Twig, which, too, is crippled in comparison to its past. I think we're gona hit 30 packets a day before everyone gets back... Mar 15 at 2:00a Our throughput has been hovering around mid-20's each day, which isn't too bad, but is letting Turbo gain on us. I expect it to come back to about 30 a day when the break ends. Hum, other than that, the only news is that we've been just barely breaking the top 1500 user mark overall, which is pretty good. In fact, I'd say real good. Mar 12 at 1:00a Good thing we got that magical 7th day in when we did, the next day was only 29 packets and today was ever lower. People have gone home for spring break, you see, and some of them have turned their computers off. I do not expect to break 30 packets a day again until everyone is back in a week. Not a big deal though, the world goes on. That 7-day 30+ stretch was a record breaker. Not sure by how much, but I know we've never done anything like that before. Mar 10 at 1:30p As of last night's measurement, we have now maintained our 30+ packets a day for 7 consecutive days. This is quite good. For the past week (ending the 5th), however, our average of 28.6 packets a day was only good enough for 19th on our seti team. Wow. That is a very compeditive team we are on. In just a few days when this week's totals are posted, we'll see if we have moved that up at all. We should have. Mar 9 at 1:00a Our six day average packet output is still hovering just about 31, which is still excellent. Very nice. We will take a hit of some sort, though, cause sadly ion is still having trouble and won't be running seti at this time. Hum. I hope that gets sorted out eventually. On tuesday (the 7th) I added what is basically a rough draft of the "about dan" page to the site, the link is above the other page links. If you have anything to say about it, make sure to tell me. Mar 7 at 4:00p Well after an interesting period where nedit would segfault whenever I launched it on modzer0, things are back to normal. Dunno what was going on. Anyway, no news, but if you ever wondered what a quad processor mobo mighnt look like: here ya go. Yeah, I want one. But those damn Xeons cost so much, and besides the architecture on these boards sucks. All four processors share one system bus... ug. When AMD comes out with their chipset... it gona hurt. But we're not greedy and if someone was to give us a quad xeon board we mighnt just take it. Dan Dan The Linux Man would in fact be delighted to take it. Mar 7 at 11:00a We have maintained a 30+ packet a day production for 4 days now, which is probably either the best we've ever done or getting dang near it. Of cource we're not talking much above 30, but it IS above 30. Our #2 seti-cruncher is no longer skud, because skud now only runs one seti at a time. Sadly, this means that yoda moves up a spot to #2, and remember yoda is a windows box. Oh well. Skud is still cranking a packet every 9 hours, but that's not good enough for 2nd. It may not even be good enough for 3rd, cause seamonkey the 300mhz G3 is around 9 hours a packet too (seamonkey runs linux, by the way, so we can still hold our heads high). Ion is also a strong contender for 3rd, I think, cause its another one of the duals that run one seti. I am not sure how fast its clocked right now, but if its more than 533mhz (5.5 x 95mhz) then its got skud beat. I don't think the 3rd place spot goes to any of the windows boxes because they tend to tack a few hours on. Hum. Well even if I can't name the top 3, I can say the top 5 are hal9000, yoda, seamonkey, ion and skud. Pretty diverse group. A dual P2, a P3, a G3, and two dual celerons. Of cource that brings me to another point: we suck. 1000mhz is out and our best is 585mhz (yoda). We're behind. Mar 5 at 1:30a Well today, in the day 155 AD (after Dan Dan The Linux Man's records started) we, the pround members of Dan Dan The Linux Man, are going to surpass the 6 years mark. Not too shabby, if I do say so. Also just yesterday or so, our average packets per day for the 155 days of record keeping has gotten past 22. That sounds low, but it also includes all of the good old days when it was just me and Dan, when skud was just a 300a at 450mhz. We pulled 36 packets today, and 31 yesterday. If we can do a good solid 30 again, that'd be great. The record for three day average for Dan Dan The Linux Man is 43.33 packets a day, but that involved a bit of a backlog. We also once did an averge of 40 packets a day for three days, but that too involved a backlog. 3 day averages above 30 are actually pretty rare around here. Mar 2-3 at midnight Minor news update today. First of all, we have increased our lead over ojohnson to 100 packets now, a nice margin. Remember he used to be quite a threat. We also have 632 packets to go till we get Panders. Plus how ever many he cranks in that time, but that's only 1 or 2 a day. Of cource Panders used to have some serious hardware behind him... he's got 8.60 years of processor time. In other news, I atempted to use the NT boxes (one of them, actually) in the lab to update my mega-spreadsheet. That effort ended when excell crashed. I, of cource, had not been saving it since I figured NT was stable. Uh huh. Ok, lets see, if you make the OS and the program, where exactly does the stability problem come in? How could there be a problem? Mar 2 at 1:15a Well well well, where to start... We are cranking a pretty dang respectable 29.4 packets a day for our 5-day average. This is excellent, in fact. I guess I'll credit this increase in performance over our old steady of mid 20's to the fact that ion and the ASD webserver are helping us these days. But it just may get better than this... I have talked to our long-lost member who moved down south (that is, out of Alaska) via ICQ. Apparently he doesn't even have internet at his house right now, but he's gona fix that. Cha-ching... his dual booting 504mhz P3 is gona give us probably 2.5 packets a day again some day. That would be the push we need to stay at or just above 30 packets a day for long periods of time... very nice. Depending on how I divide me time between AI studying/AI programming/web stuff tomorrow, we could possibly see a "hardware" section go up in Dan Dan The Linux Man's empire page, and we also hope to get some links up soon on the links page. One thing I forgot to make a big deal out of yesterday is that this web page is now one month old. Pretty neat. I would also like to note that Dan Dan The Linux Man is always looking for new recruits, especially people who we know. If you have the desire, just go ahead and email me at the (far) above address. Mar 1 at 2:15a News has come my way today that the 333mhz celeron down at ASD is once again helping Dan Dan The Linux Man, adding a packet and a half each day, I bet. Other that the fact we like as many packets as we can get, this is also a Linux running web server... the best kind of box. Maybe. Dan Dan The Linux Man only has 682 more packets to go before catching our next victim: the one known as Panders. Current estimate is 27 days. Feb 28 at 6:45p I'm working on the page some today, so if it seems broken, thats probably why. I think I also figured out enough stuff so that IE will display this page as well as netscape, and so I am removing my little IE note from the top. Feb 27 at 1:45a Well Dan Dan The Linux Man logged 31 packets today, and 34 yesterday, which is pretty good. I would say real good but for days before that we were doing mid-to-low 20's, so this is possibly some sort of backlog getting pushed through. Dan Dan The Linux Man is ranked 1514th out of all the seti users at the moment, with 3630 packets. No new recruits to speak of, although I intend to pass through the computer lab at the physics building and aquire a few iMacs for us. Six 333mhz iMacs would add a LOT... as much as 18 a day, based on the processing rate of seamonkey, our 300mhz G3 that does around 3 a day. Of cource seamonkey has a 1 meg L2, whereas the iMacs have 512k, so that would hurt them somewhat in comparison. I also do not know if they are left on at night, since I've never been up there at night. News has come my way that the only busness machine that is working for Dan Dan The Linux Man, and also possibly to only pre-built machine in Dan Dan The Linux Man's army, a 300mhz pentium 2, does it packets in something like 12.5 hours each. This strikes me as pretty darn good. This thing is running a GUI windows client! 12 hours isn't too far off what the 450mhz to low 500mhz celerons can do in windows. Ok, maybe its not all that special, but it seems pretty fast. Anyway it is certianly going to add to the glory of Dan Dan The Linux Man. Oh, and on today's sad note that athlon we were going to recruit (actually build for someone) has vaporized. You see, the people who were gona get it decided to just make their 120mhz box go a little farther with another 32mb of RAM and more hard drive space (it can't even print without running out of swap right now). Well this is not a defeat for Dan Dan The Linux Man. There is no such thing. We'll, uh, run seti again on that 120. Yeah. 70 hour packets, baby! I do not expect to do any major work with the page this week since I am getting into some pretty tough coding for my AI project. But eventually things will happen to this page. Feb 24 at 9:30p With a helpful push from bishop, we've now got a simple counter up. It keeps track of the last ip to load the page, and so hitting reload a bunch won't effect the counter unless someone else gets their ip in there between yours. I'll be the first to admit that I'm learning how to do all things webpage. It also doesn't count either of the ip's I may view it from. The counter is that tiny number at the bottom. When it is a better counter, I'll make it more prominent. At this moment, it is reporting 6 hits, of which three are mine, two are from some guy (with an ip that starts 24.237... if you happen to be reading), and there's another some guy from somewhere. Feb 24 at 8:00p Well whoremachine is back, now with a neato hard drive that grows every time we turn around... first we thought is was 200meg, but it was 500, then a gig, and now its 1.3gig. No, we have no idea why. I guess our memory sucks. But anyway whoremachine is back, still at its smoking-fast overclocked non-mmx 200mhz. I hope to be able to run 64mb of RAM in it and then use it as another skud terminal, mainly. In other, possibly more important, news, Dan Dan The Linux Man has gotten his tape backup to work under linux today. Windows, unaware of the danger, continues to lurk on his hard drive. Dan's copy of windows is the only one left in the room. Feb 24 at 4:00a Well I finally updated the page's layout. The Empire of Dan link will probably not go anywhere for a while, but maybe we'll see some pictures and stuff on that page soon. Feb 24 at 2:00a Forgot to mention yesterday that that new athlon we hope to add will be our teams's very first AMD processor, assuming that fischer up the hill isn't still running our stuff on his K6-2... I've never asked. It not that we're huge Intel fans, find you, the celeron is just a lot better of a deal than anything else out there. This situation should change next semester though, cause as always there is talk floating around about athlons, but this time it dual athlons... 851 packets to Panders. Modzer0 crashed today, which is sad, but its not our fault. Seems to have something to do with a brute-force program for a crytography cource by a modzer0 user. Feb 23 at 1:00a Well whoremachine got its (borrowed) hard drive taken away today so we have lost one of our most faithful seti boxes. A moment of silence... actually without that whiny hard drive, its gona be daya of silence in here. On the positive side of things, it looks like a 650mhz athlon will be on our team in a week or two, and actually we can probably bring whoremachine back with a 500meg drive thats around here also. 874 packets to Panders. Feb 22 at 12:30a The problems with the linux clients are continuing, this time we found modzer0 striken also. Today both of skud's clients were dead when I came back from class, but they restarted without problem. Hum. As far as how we're doing, our output is down to the low 20's, probably partially do to the problems we've been having. We are now 905 packets behind our next victim: the one known as Panders. I have come to the conclusion that I will never ever get this page updated to quite what I originally pictured. The lack of photoshop is somewhat to blame, but also the fact that I've got other things to do. Feb 20 at 11:30p Both me and Dan Dan The Linux Man himself have been having problems with our computers getting "unexpected end of file" errors from the seti servers. He has spotted it recently on hal9000 three times, I've seen it twice on skud this week (and never before) and just today saw it on skug. Each time it happens seti stops in its tracks, and so if we don't notice we loose some pretty good hours. We are not, as you may imagine, accustomed to our computers not just doing what we tell them day in and day out. All the afflicted computers run linux, so I guess the linux client has a problem. All of us running linux and seti better check on our clients. Feb 18 at 4:30a Well this is what I like to see... our past three daily packet outputs have been 33, 34 and 30. Some of this, about 2.5 to 3 of those per day, can be credited to bishop and the fact he got ion (or modx) stable and doing seti on one processor. Because its not going on both, yoda will continue to be our 3rd best seti box, and remember folks, its a windows box. 1st, 2nd, and 4th are all linux, at least, and I never have figured out who #5 is. Perhaps it's skug, or maybe yog-sogoth, or maybe spoon, all of which are celerons from 458mhz to 541mhz with bus speeds of 83mhz and 75mhz. Hum, could even be book_of_mudd, another celeron of simular specs, or maybe wilbur, our only 300a...... Well anyway we also added another pre-made today. Its a 333mhz celeron compaq. A friend of null_pointer's. Lets all give him a sloppy kiss next time we see him. Its not everyone that has converted a friend to our cause. When I claimed we were within a thousand of the next guy to pass a few days ago, I apparently lied. That seti-doer is now 995 packets distant. If we could maintain a rate of 30 packets a day, it would take us about 34 days to catch him, plus one or two more to account for the packets he's doing. So if we could maintian this excellent packet rate, Dan Dan The Linux Man would be happy. Small changes to the page today (actually yesterday). Hopefully I'll do a little more this weekend, including breaking this page up and making and old news section to get rid of some of this length. Feb 16 at 1p I think we are within one thousand packets of the next guy we are going to pass now. Given that he is basically not moving, it should take us 40 days or so. No news on the recruiting front. No news on the spreadsheet front either, the spreadsheet being important so I can calculate things like about how many computers are working for us as well as the rest of the good ol stats. I would like to do that, but it requires windows. Ug. Perhaps I'll find my way to yeti's box and borrow that some day when he isn't using it, which seems to be often enough. Perhaps I'll do a bi-weekly super-mega-big-update. I haven't gone door to door asking about v2.0 seti clients yet, but you better get em soon if you don't have them already. The v1.x clients will stop working before too much longer. I am getting back into the programming (and hopefuly soon quake) cycle again, and so I won't be doing to many wild and wacky things around here. With ojohnson safely behind and no other immediate threats, we can just sit back. I would like to do some page upgrades, but who knows when that will happen. For one thing, this page is getting pretty long. Feb 14 at 1a We averaged less than 20 packets a day these past two days due to the server outages, which isn't so great but not really a big deal. We are in fact in 16th place now on our team, and 1597th out of all the seti users (there are 1.7 million accounts, many idle). So anyway we're doing pretty well. It also appears that we're about the 13th best on our team in terms of all-out packet production per week, which is pretty good I think. Of cource I am still looking to improve that number any way I can. We are working on getting another linux box, a cyrix so-called 200 that runs at more like 150mhz. Still, a box is a box, linux is linux, and packets are packets. In addition to that box, named atrus or something, that guy runs a 466mhz celeron windows box named the_realm, or something, for us. I didn't even know this guy was helping us. A digital camera has fallen into my hands recently and soon pictues of many of our boxes will start to pop up. Right now I have suitable pictures of modzer0, whoremachine, hal9000, skud, and skug. I would like to add yoda, yog-sogoth, spoon, book_of_mudd, seamonkey, wilbur, and the styrophoam computer to the list tomorrow. Ion (or modX) is probably out of range, as is jaskug, hal_jr and a number of other boxes not in the dorms. Dan Dan The Linux Man's trusty servant, his little brother, is still busily working away at trying to get us all old pentiums that his school has around. Hopefully we will see them in a few weeks. Adding a few of those, no telling how many well get, would be pretty neat. Like Dan Dan The Linux Man has tought me: a packet is a packet. Or actually maybe that's my saying. Important news item: The 2.0 clients are fast, everyone should upgrade to them. Feb 13 at noon The servers are back, at least a little, now. Skud and hal9000 each have one packet going and skug also got its packet. Lets also rejoice that ion (or is it modX?), bishop's home box, has been up for two days in linux, a new record, possibly. For those of you who are unenlightened about this box, it is an Abit BP6 based computer, and those BP6's sometimes seem to have big problems with linux. Anyway lets hope it keeps running, cause we all like dual-processor linux computers. Especially those running seti for Dan Dan The Linux Man. Feb 13 at 10:30a The seti servers are down. Skug has been trying to connect since last night, I think one of hal's setis got done like 9p last night and couldn't connect. I was also unable to get to the webpage for statistics. Assuming that this isn't another UAF-only outage, when we come-to we should find ourselves ranked above cthulu and in 16th place. Next guy to destroy: 1000 packets ahead of us, but at least he's flat-lining. The real concern for Dan Dan The Linux Man now-adays is people getting us from behind, and right now I don't see any that are going to do that real soon. Well I tried out staroffice's excel-speadsheeting ability last night on our seti records, and it did ok, but its ability to handle graphs is sub-par. Now I miss two applications from windows: I miss photoshop and now microsoft's very own excel program. On the good side of things, Dan Dan The Linux Man himself is going to see if some new linux drivers for his tape-backup mighnt work now so he doesn't have to boot into windows to back up his data. I would never slight Dan Dan The Linux Man, yet somehow it strikes me as pretty funny he boots into windows to backup his linux stuff. Well anyway if those drivers work, no more windows! Haha. That would mean 5 linux-only computers. Unless I follow a plan I've sort of been forming to boot whoremachine into windows, despite its faithful service, so I can maintain a proper speadsheet and do some graphics for the page. I think it would be pretty stable there, since I wouldn't install very much for windows to trip over. Of cource whormachine has been up for 14 days or something and we don't want to make it stop. If only I hadn't broke that little peice of plastic off of pez's processor clip I'd run windows on the 100mhz box, heck, I'd clock it DOWN to 66mhz. Feb 12 at 8:30p I lied. I do that. I found out we have a computer made by Gateway that the dad of Dan Dan The Linux himself runs seti on. Its a 300mhz or maybe 333mhz, and he runs it at work. I think this is also the only workplace machine we have working on us. Don't know what its called, but I've asked. Feb 12 at 1:30a Our packet throughput was good today at 26 packets, which was enough to keep us pulling away from ojohnson, but not enough to catch cthulu who is still 11 packets ahead somehow. We've been accumulating 1/20th of a year of processing time per day for the past few days, which is pretty impressive it seems. We do a year in 20 days... almost. Well recruiting these days isn't a real lucerative busness. We got turned down by the guy upstairs who has the 600mhz P3 Hewlett Packard... of cource he does have a HP so we didn't really want him anyway, right? Interesting fact: Dan Dan The Linux Man, outside of Macintoshes, has NO pre-assembled computers working for him. Yes, its true. Well this page is still up for improvement. I am still working on that counter idea and also adding the section for mission statement and a list of computers by name, along with their hardware, OS, and distigishing features. Such as being a peice of crap. The main thing holding me back from just going out and making things go here is that I don't want to learn how to use Gimp... I liked Photoshop fine and I don't want to go changing that... now if only they made it for linux. Oh well perhaps I'll do the graphics from someone else's machine. Feb 10 at 12:15a Well it appears that ojohnson is once again closing in on us, now 10 packets behind, so we better watch out. On the positive side of things, cthulu is now only 56 packets ahead of us and just barely moving at all, so in two days we should have him. And in not so interesting news, I was noticing that our average time went up pretty good again, and is about to break back into the 13 hour 42 minute range again. Interesting tidbit of the day: we have spent about 156514685.849821 seconds on seti as of sometime today when I was installing seti on skug. That translates into about 4.9597 years. Looks like tomorrow (the 10th) will be the day when we beat the 5 years mark. Feb 9 at 9:30p Whew... got linux installed on skug and it works now. This is of cource a proud moment for Dan Dan The Linux Man, cause after a week, three (or four) linux installs, one windows install, four network cards and three video cards, Dan Dan The Linux Man's armies have emerged victorious over the vile OS known as windows. The windows CD goes back to its deep, dark drawer were its tortured soul will remained imprisioned for all time. heh. They could make a game like diablo about this. Hopefully tomorrow this page will get a hit counter divided by team memeber hits, outsider hits, and total hits. Another project I want to work on but haven't yet even made a master plan for is making a sub-page for a mission statement of sorts, and a page listing and talking about all of Dan Dan The Linux Man's serfs. Feb 9 at 2:00a Dan Dan The Linux Man is fast approaching two important events: one is surpassing the 5 YEARS of processing time mark, and the other is passing a fellow ars-technica guy named cthulu for 16th place, probably the last time we'll increase in ranking for a long, long time. Our team is a serious team... let me tell you. Our average time is swelling up pretty quick again... no idea why. Jumped by something like a minute on average. Thats a big change given 3140 or so packets already there. Much like before when it dropped, I am uncertain why it now climbs. As far as getting more computers on the job for us, well, it takes time and right now I am a bit busy. In all honsety, I am sick of hardware, reinstalls, things not working and the load of it. Still, I think when I get linux back on skug I'll install it on pez also. There is also a guy upstairs with some pentium that he has got in linux these days that says he'll work for Dan Dan The Linux Man, but he doesn't have the RAM yet. Also, I periodically see if I mighnt get it running on twig's windows box, but for some reason he won't do it. Feb 8 at 1:00a Well today was a bad day for me, mainly because my linux on skug went to hell. But at least it took my entire windows install there with it. The only happiness I can get from this is that one more install of windows has been destroyed and not (yet) replaced. We did 28 packets today and logged about 1/20th of a year of processing time, our averge packet time went up a little but is still pretty good. I am going to bed. I hate windows. I do not like mandrake linux very much anymore. Feb 6-7 at midnight We are now ten packets ahead of ojohnson, so it looks like we've got him beat. He could still catch up, of cource, we're not outrunning him by that terribly much, and we aren't pulling ahead cause we sped up... he slowed down. We did an ok 25 packets in this past day, but our average time dropped pretty fast and is into the 13hr 40min range now, the lowest its been in a while. Not completly sure why. It seems like the newer clients must be helping, along with yoda, and wilbur also does packets in 12 hours or so, and so its helping bring it down too. On a bit of a sad note, our 133mhz pez is now a 100mhz pez, cause I thought that our floor's RA mighnt just need that extra 33mhz more than a seti box did... this is a sad moment for Dan Dan The Linux Man, but since we gave that 133 to a windows box, well, I guess we're in the running for some kinda Nobel Peace Prize or something. Of cource bishop may snap and kill either me, the RA, both of us, the windows box, or perhaps all three. It is his processor... does it help that me and wilbur's owner (are we calling him null_pointer?) built it on peice of styrofoam (from a box of my 750a case)? You know, I just have to post of picture of that computer. One of these days I'll get this page fixed up with a few more features, but don't hold your breath... I'm coding again. (at modzer0's LILO prompt) linux init 1 pico /etc/rc.d/rc.local (arrow to bottom) (type:) cd /seti nice -n -19 ./setiathome & (cntrl X) (Y) reboot **ahem** "oh me oh my," "i think im gona die," "oh poor poor me," "modzer0s idle is 99.3!"(percent) Feb 6 2000 at 2:30a Busy day (the 5th). Dan Dan The Linux Man, by means of terrorizing and bullying the local peasantry, has aquired yet another computer. This time its pez, the computer that was replaced by book_of_mudd. Minus the p200 which went somewhere else last semester. So bishop, being the kickass guy he is, brought over a p133 for Dan Dan The Linux Man, and the two (the processor and mobo) are about to be united on my floor. This union will create a new linux box, our 6th to inhabit this very room. The stats for today (the 5th) reveal that we pulled ahead of ojohnson and are now six packets ahead, for a total output of 33 packets (the backlog was reduced to 5 packets). This has been a great day, according to our 30-packet line. It is highly unlikely that we will complete so many packets tomorrow: I forcast a mid-20's output. Our average packet time is more-or-less stable now at 13 hours and 41 to 42 minutes a packet. I hope to see this number decrease, since it seems like the vast majority of our computers should be doing sub-12 hour packets. Of cource pez won't help that much with all its 133mhz... Feb 5 2000 at 8:00p Looks like we in fact are leaving ojohnson behind... by a lot, suddenly. We pulled something like 34 packets so far, including the elimation of some of the backlog I think. Not sure exactly where the packets came from, but packet production always has, and probably always will, vary from day to day with no apparent pattern. In other news, with the conversion of skug from windows to a dual boot (with most of its time in linux) we have gained a linux box. This is, needless to say, of great importance to the collective ego of Dan Dan The Linux Man. Feb 5 2000 at 11:45a Well we finished two packets ahead of ojoshnson yesterday, so looks like the recent addition of winbur and the couple of packets ion did (at risk of crashing itself) have kept us from falling a rank. This is most excellent. I also installed the new 2.0 GUI client on skug and it did a 10 hour packet, which I believe is better than its average old packet by two hours. If the 2.0 client makes this big of a difference, we certainly need to get it on yog-sogoth, spoon and book_of_mudd, which are all high-clock low-bus computers simular to skug except faster (and in same cases, with a lower bus speed). Our performance, aside from being good relative to ojohnson, was pretty good at 28 packets logged with a ten packet backlog yesterday. Assuming some of that backlog starts to get churned through today, we should get credit for another nearly 30 packet day. For the record, a day with great performance is at 30 packets or higher. There was once a time when we could do that every day, although it was brief. And like I keep saying, when ion comes back from its sickness, we're gona get there again. For those of you who may have tried to get here late yesterday and couldn't, well we installed mandrake 7 on modzer0 and that took a bit of doing. For one thing... it thought the harddrive it was installing on was bigger than it was and so we would install it and it wouldn't work... but eventually we got that fixed. Feb 4 2000 at 1:00p According to the seti page we have put two more packets between us and ojohnson, for a grand total of 3. It remains to be seen if we are still that far ahead tonight. Feb 4 2000 at 12:00a Just got the seti stats for tonight. Looks like we've done an average of 26 packets a day for the last two days, which isn't too bad. Our 10 day average number of processor*days is hanging just below 15 still, also not too bad. ojohnson is just one packet behind us, however, so we're gona need to do a bit better to stay ahead. Today Dan Dan The Linux Man gained a servant. A guy down the hall joined up and brought his dualboot win98/linux celeron 300a at 450mhz (wilbur) along. He is even going to run seti dedicated for us. He also has helped me to learn how to do web page things, so I can do this page. Back to the packet-crunching: In addition to wilbur, the mac seamonkey (running mk linux) is also back on the job, after a downtime of some length I am not entirely certain. Seamonkey is a 300mhz G3, and does packets at a rate of one every 8 hours I think, making it a top performer amoung the computers of Dan Dan The Linux Man. The fact that it also runs linux is a object of great pride, since that means 3 of our top 4 boxes run linux. (It is not clear what computer has the 5th best packet throughput, but should ion ever become stable, it will be the 2nd fastest (but VERY simular to skud) and skud, yoda and seamonkey will take a step backwards and linux will count for 4 of the top 5 computers.) On a sad note, I don't even know if one of our old top-5 is still running: a box named Whee-Knee (or something like that) that moved down to the lower 48 somewhere with its owner, Aaron. He said he'd still work for us, but I haven't heard from him yet... sad story. Well anyway his box was/is a dual booting 504mhz pentium 3, and did the packets pretty fast. Another sad story is a box named Buddha owned by my old roomate. That thing is a prime example of why we don't go by Dan Dan The Windows Man, although it does live in one of those big 'ol Supermicro sc750a cases and have lotsa SCSI stuff. Its another 300a at 450mhz, when its running, that is. Perhaps some day he will once again lend a hand to our cause. Overall, I'd have to say that the story of 300a's around here is a tradjedy. Not that they aren't fast, no, they all have gotten to 450mhz, but the fact is that we've had no less then 5 of them in our effort at one time or another, and at this moment I think the only one we have is wilbur (who was added today). The others were sambra, a linux box with a disinterested owner, kore with a hard-core linux owner (that sucker did the packets FAST for a 300a) who sees seti as unessasary, buddha, and twig's other machine named twig (it may run seti again some day soon). The only two things around here more common than 300a's are celerons in general and Matrox video cards... Feb 3 2000 at 7:30p Oh no! Last night I actually forgot to collect my seti data! This is terrible. A sighn of weakness in Dan Dan The Linux Man. Day Of Statistics #123 will go down in history as a big, black, mark. And speaking of weakness, the mortal known as ojohnson is hot on our tail. Still. Feb 2 2000 at 8:00p Last time I checked, I guess last night, ojohnson was just 3 packets behind us. This is bad. He also had gained 3 packets since the night before. Also bad. We may soon find ourselves at 18th place. Feb 1 2000 at 11:30a Suppose I should have talked about it last night, but anyways we've been in a tight race with a guy named "ojohnson" for 17th place in Team Ars Technica, our seti team. Looks like we are pulling ahead again, after several days of loosing ground an in fact being behind him briefly. Also, our average time per packet had been growing steadily, but in fact last night it went down. We can probably credit that drop to yoda, the newest box we have, I think. Its a 585mhz pentium III owned by none other than yoda's owner, and it does packets in about 7.5 hours a piece. Sadly, this newest box is a windows box, and this marks the biggest threat to the local supremesy of Linux yet, although hal9000 still slaps it silly and skud is a little faster. If ion ever becomes stable, then this little usurper windows box will really be put back in its place... 4th place. Oh, and when the Linux boxes get their upgrades next semster... Feb 1 2000 at 1:15a Dan Dan The Linux Man's first news posting! All you ho's around here not already working for me... get with it! Well OK, perhaps I'll forgive you if you don't. Its a bit hard to tell exactly, but we've had an average of 15 processors working for us over the past 10 days. Of cource, four of those are in me and Dan's dual boxes, so that leaves 11 of the rest of you... not too bad but we were better before spring break. One word about this processor measurement: its measured in processor*days, which is a processor spending 24 hours working on seti. This means that we have more than 15 processors actually working for us, just many of them spend only nights working and so it mighnt take two or three of these screensaver boxes to add up to a single processor*day. Some boxes, such as our linux boxes, run 24-7 of cource, and so they add up time much more rapidly... and two processors rack up the time twice as fast as one, needless to say. This means that the linux boxes- hal9000, skud, modzer0, and whoremachine, account for about 2/5th of the time by themselves. When you add in skug, yoda and scott's mac (what's it's name, anyways?) we arrive at 11 processor days amoung the hard-core alone, the other 4 or so go to the screensaver fleet. Just thought I should share that with you all. We've been doing about 24 packets a day over the past 15 days, which is headed roughly down, sadly. If you, for some reason, want to talk to me, I can be reached at fsras@uaf.edu. Yes, I would love to hear from anyone who wants to bodily sign himself or herself over to me for the glory of Dan Dan The Linux Man. We do try to stick to only people we know, however, because we are not creating a mini-team here: only looking for those who will run for Dan Dan The Linux Man. Not that I forsee a rush to join.