From Bartosz.Lew w astri.uni.torun.pl Mon Mar 3 13:53:41 2003 From: Bartosz.Lew w astri.uni.torun.pl (Bartosz Lew) Date: Mon, 3 Mar 2003 13:53:41 +0100 (CET) Subject: new cmb code... (fwd) In-Reply-To: <3E59E487.2040208@ncac.torun.pl> Message-ID: ah ta filozofia ;) > a jakie masz parametry? pewnie nie musisz nic przerabiac. moj program > jest bardzo modulowy poza tym. ja uzywam tak: Omega_b, Omega_CDM, Omega_Lambda, H_0, tau, n i warunki poczatkowe (tylko 2 wart. - adiabatyczne albo isocurv.) liczeniaa jest od groma i jeszcze troche to potrwa. ok. w czym napisany jest ten Twoj program do interpolacji i czy ten sam program robi ta parametryzacje widm (i co to jest za parametryzacja)? Na Twojej stronie w kazdym pliku (w tych o nazwach z wartosciami parametrow) jest chyba z kilkanascie roznych wartosci - czy to jest wlasnie ta parametryzacja ? teraz mam czas wreszcie zeby sie zabrac do roboty. zastanawiam sie ostatnimi czasy jak to zrobic zeby przypasowac docelowo ok 280 tys widm teoretycznych do danych z mapa. (choc zaczymam myslec ze chyba lepiej bedzie zrobic to kilkuetapowo - metoda kolejnych przyblizen). Jest taki programik co liczy prawdopodobienstwo pasownosci danego widma do danych mapa. na stronach lambda. obczajales go moze ? I jeszcze jedna kwestia. Jak wiadomo wielka bolaczka sa dlugotrwale obliczenia widm. ale ten proces mozna znacznie przyspieszyc mniej wiecej o rzad stosujac nowa metode ksplit - pewnie jak pisales magistere to jej pewnie jeszcze nie bylo (a moze byla). jesli ja uzywalas to co o niej sadzisz ? (daje dosc mala mozliwosc ingerencji w rozne parametry) pozdrawiam Bartek > ps. imponujace podroze na Twojej www.... moja www to juz historia, from mysterious reasons. moze ja przeniose na astronomie. From amr w astro.uni.torun.pl Tue Mar 18 10:04:47 2003 From: amr w astro.uni.torun.pl (Andrzej Marecki) Date: Tue, 18 Mar 2003 10:04:47 +0100 (MET) Subject: Gazeta Wyborcza o odkryciu Tegmarka Message-ID: <200303180904.KAA10483@galileo.astro.uni.torun.pl> Artykuł o odkryciu ("odkryciu" ?) Tegmarka: http://www.gazeta.pl/nauka/1,34148,1376101.html Krytyka Marka Demiańskiego: http://www.gazeta.pl/nauka/1,34148,1376526.html -- Andrzej From boud w astro.uni.torun.pl Wed Mar 19 20:06:15 2003 From: boud w astro.uni.torun.pl (Boud Roukema) Date: Wed, 19 Mar 2003 20:06:15 +0100 (MET) Subject: topology - WMAP Message-ID: Please follow the WMAP thread on cosmo-torun http://www.astro.uni.torun.pl/sympa/cosmo-torun/2003-03/msg00001.html Thanks boud From szajtan w poczta.onet.pl Wed Mar 19 22:52:55 2003 From: szajtan w poczta.onet.pl (szajtan odwieczny) Date: Wed, 19 Mar 2003 22:52:55 +0100 (CET) Subject: files,files,files Message-ID: hello everyone, Does anybody know theoreticaly how many files can there be on ext2 fs in one directory, on on fat32 filesystem in one directory ? bartek From tomlacz w go2.pl Thu Mar 20 00:05:20 2003 From: tomlacz w go2.pl (Tomasz Laczkowski) Date: Thu, 20 Mar 2003 00:05:20 +0100 Subject: NTG Re: files,files,files In-Reply-To: References: Message-ID: <200303200005.20659.tomlacz@go2.pl> Dnia śro 19. marzec 2003 22:52, szajtan odwieczny napisal(a): > Does anybody know theoreticaly how many files can there be on ext2 fs > in one directory, on on fat32 filesystem in one directory ? On ext2 it depends on partition size. There is limited number of inodes (=files, links, directories...) for a whole partition not for a single directory. Do as a root: dumpe2fs /dev/hd?? | head -35 and you will see. Look for inode count and free inodes. Usually number of inodes=blocks/2 or blocks/4 and a block is 4096B. On FAT32 there is also limit for whole drive and not for single directory (on FAT16 there was a limit for a root directory=512). It depends on total number of clusters. Do you need so many files in one directory? It may slow down the reading. [TL] From michalf w ncac.torun.pl Thu Mar 20 12:44:56 2003 From: michalf w ncac.torun.pl (Michal Frackowiak) Date: Thu, 20 Mar 2003 12:44:56 +0100 Subject: files,files,files References: Message-ID: <3E79A9B8.2030605@ncac.torun.pl> szajtan odwieczny wrote: >hello everyone, > >Does anybody know theoreticaly how many files can there be on ext2 fs in >one directory, on on fat32 filesystem in one directory ? > >bartek > > > > no limit as long as you do not exceed disk's capacity. if you encounter problems this is because of the soft reading directory content, not ext2 itelf. pozdr - michal From amr w astro.uni.torun.pl Thu Mar 27 12:56:45 2003 From: amr w astro.uni.torun.pl (Andrzej Marecki) Date: Thu, 27 Mar 2003 12:56:45 +0100 (MET) Subject: Matched circles on WMAP data Message-ID: <200303271156.MAA01996@galileo.astro.uni.torun.pl> Paper: astro-ph/0303580 From: Alain Riazuelo Date: Wed, 26 Mar 2003 17:20:20 GMT (641kb) Title: Cosmic microwave background constraints on multi-connected spherical spaces Authors: Jean-Philippe Uzan, Alain Riazuelo, Roland Lehoucq, Jeffrey Weeks Comments: 4 pages, submitted to PRL \\ This letter describes the Cosmic Microwave Background anisotropies of multi-connected spherical spaces in a closed universe with density parameter Omega_0 close to 1. [...] -- A. From michalf w ncac.torun.pl Fri Mar 14 13:16:45 2003 From: michalf w ncac.torun.pl (Michal Frackowiak) Date: Fri, 14 Mar 2003 13:16:45 +0100 Subject: free f90 compiler for linux References: Message-ID: <3E71C82D.6060808@ncac.torun.pl> fortran compiler we have been talking about is at: http://www.psrv.com/linux.html good luck ;-) From michalf w ncac.torun.pl Fri Mar 14 13:22:47 2003 From: michalf w ncac.torun.pl (Michal Frackowiak) Date: Fri, 14 Mar 2003 13:22:47 +0100 Subject: free f90 compiler for linux References: <3E71C82D.6060808@ncac.torun.pl> Message-ID: <3E71C997.4020208@ncac.torun.pl> Michal Frackowiak wrote: > fortran compiler we have been talking about is at: > http://www.psrv.com/linux.html > > good luck ;-) @#$(%&#!!!! "This product is no longer being offered." a nice thing about non-opensource soft. From szajtan w priv.onet.pl Fri Mar 14 13:47:26 2003 From: szajtan w priv.onet.pl (Szajtan Odwieczny) Date: Fri, 14 Mar 2003 13:47:26 +0100 (CET) Subject: free f90 compiler for linux In-Reply-To: <3E71C997.4020208@ncac.torun.pl> Message-ID: huehue, thats right ;) I've been there. dont know what happened is it commerical from some time on or what ? but there are some trial verisons over the net but need to get some licence key from different server. ******************************************************************************* Bartosz Lew e-mail: szajtan w astri.uni.torun.pl szajtan w phys.uni.torun.pl Centrum Astronomii UMK szajtan w poczta.onet.pl TORUN, POLAND www: http://www.phys.uni.torun.pl/~szajtan - no longer ******************************************************************************* On Fri, 14 Mar 2003, Michal Frackowiak wrote: > Michal Frackowiak wrote: > > > fortran compiler we have been talking about is at: > > http://www.psrv.com/linux.html > > > > good luck ;-) > > @#$(%&#!!!! "This product is no longer being offered." > a nice thing about non-opensource soft. > > > From szajtan w poczta.onet.pl Sat Mar 15 17:40:19 2003 From: szajtan w poczta.onet.pl (szajtan odwieczny) Date: Sat, 15 Mar 2003 17:40:19 +0100 (CET) Subject: map data Message-ID: well this is the thing I've been talking about on last thursday. Look at these two pictures. ;) Maeybe this kind of analysion is useless but surely it shows what problem we are dealing with here. On my plot (green dots) I chop off all negative values, all the values above 10000 for multipoles greater that 300 (which isn't a big crime), and throw out all points of which "errorbars" were bigger that the values itself. The red dots are raw data as it goes from LAMBDA. Well perhaps (I hope ;) somewhere among green dots are those we see at the picture from NASA. bartek -------------- następna część --------- Binarny załącznik wiadomości został usunięty... Nazwa: map.eps.gz Typ: application/x-gzip Rozmiar: 175521 bytes Opis: Adres: -------------- następna część --------- Binarny załącznik wiadomości został usunięty... Nazwa: f12.jpg Typ: image/jpeg Rozmiar: 134327 bytes Opis: Adres: From boud w astro.uni.torun.pl Sun Mar 16 00:36:10 2003 From: boud w astro.uni.torun.pl (boud) Date: Sun, 16 Mar 2003 00:36:10 +0100 (MET) Subject: map data In-Reply-To: Message-ID: hi Bartek, everyone, On Sat, 15 Mar 2003, szajtan odwieczny wrote: > well this is the thing I've been talking about on last thursday. Nie dokładnie rozumiałem, ale nie ma sprawy ;). > Look at these two pictures. ;) > Maeybe this kind of analysion is useless but surely it shows what problem > we are dealing with here. > On my plot (green dots) I chop off all negative values, all the values > above 10000 for > multipoles greater that 300 (which isn't a big crime), and throw out all > points of which "errorbars" were bigger that the values itself. > The red dots are raw data as it goes from LAMBDA. The good news is that it looks something like typical C_l curves. The bad news (depends on your point of view) is that i don't understand quite what is being plotted. Did you calculate the C_l spectrum directly from the "linear" WMAP map, or are these from your model calculations? pozd boud > Well perhaps (I hope ;) somewhere among green dots are those we see at the > picture from NASA. > > > bartek > > From michalf w ncac.torun.pl Mon Mar 17 17:47:36 2003 From: michalf w ncac.torun.pl (Michal Frackowiak) Date: Mon, 17 Mar 2003 17:47:36 +0100 Subject: new cmb code... (fwd) References: Message-ID: <3E75FC28.5040906@ncac.torun.pl> There is some draft description of the code. For anyone interested.... - perhaps only Bartek ;-) http://coyote.ncac.torun.pl/cmbrecon/ From boud w astro.uni.torun.pl Mon Mar 17 17:39:54 2003 From: boud w astro.uni.torun.pl (Boud Roukema) Date: Mon, 17 Mar 2003 17:39:54 +0100 (MET) Subject: new cmb code... (fwd) In-Reply-To: <3E75FC28.5040906@ncac.torun.pl> Message-ID: On Mon, 17 Mar 2003, Michal Frackowiak wrote: > There is some draft description of the code. For anyone interested.... - > perhaps only Bartek ;-) > > http://coyote.ncac.torun.pl/cmbrecon/ Hey, I'm interested too! I just don't have time (yet). I'd rather do a circles analysis. Look up http://de.arXiv.org/abs/astro-ph/0302496 From szajtan w poczta.onet.pl Tue Mar 18 01:38:22 2003 From: szajtan w poczta.onet.pl (szajtan odwieczny) Date: Tue, 18 Mar 2003 01:38:22 +0100 (CET) Subject: map data In-Reply-To: <3E75F5C7.3010203@ncac.torun.pl> Message-ID: I know this looks bad ;) and ... discouraging. But non of is comes from my code. I just ploted the data from internet - from lambda. They get a nice spectra because they bin it every 25 multipoles (on average), and as Boud noticed with some weight, which I have to chack. In this way we get that few point like in the first picture I sent but the errors become smaller. They also drop all data above l approx 600. Well there is no easy way out, and much to do. Now I'm also thinking about calculating the power spectra directly from the map from MAP. Any ideas with aid of what to do it ? (witting a new code I give up - from time reasons ) bart. oh btw. that "vast f90" you mentioned is also commercial now. On Mon, 17 Mar 2003, Michal Frackowiak wrote: > you must have missed something important.... it's more than a game of > "show the diffrences in 2 pictures". it's a game of "check your code and > do it again..." ;-) > > szajtan odwieczny wrote: > > >well this is the thing I've been talking about on last thursday. > >Look at these two pictures. ;) > >Maeybe this kind of analysion is useless but surely it shows what problem > >we are dealing with here. > >On my plot (green dots) I chop off all negative values, all the values > >above 10000 for > >multipoles greater that 300 (which isn't a big crime), and throw out all > >points of which "errorbars" were bigger that the values itself. > >The red dots are raw data as it goes from LAMBDA. > > > >Well perhaps (I hope ;) somewhere among green dots are those we see at the > >picture from NASA. > > > > > >bartek > > > > > > > > ------------------------------------------------------------------------ > > > >------------------------------------------------------------------------ > > > >LISTNAME: shape-univ > >HELP: send an email to sympa w astro.uni.torun.pl with "help" > >WEB ARCHIVE: http://www.astro.uni.torun.pl/sympa/shape-univ/ > >UNSUBSCRIBE: email to sympa w astro.uni.torun.pl with "unsubscribe shape-univ" > > > > > > > > >