Domain: penguin.cz
Stories and comments across the archive that link to penguin.cz.
Comments · 9
-
Re:What about large files and new WordPerfect?There are TeX distributions for Windows (the most commonly used being MiKTeX at http://www.miktex.org/ ). For learning LaTeX, you could start with "the (not so) short introduction to LaTeX2e" to get hold of the basics. For learning ConTeXt, there is a manual available online and some tutorials over at the ConTeXt wiki http://wiki.contextgarden.net/Main_Page
There is a WP2LaTeX tool http://www.penguin.cz/~fojtik/wp2latex/wp2latex.htm available for both Windows and Linux which does a pretty decent conversion. BTW inserting accented characters in TeX & friends is rather straightforwards: \`e, \'e, \"e etc so you can usually get by with just a couple of extras keystrokes on a standard US keyboard.
-
I just set up such scanner at my dad's linux box
He had bought another model of that series (3170). But anyway yours should be similar if not the same setup. I used:
-SuSE 9.0 (distro does not matter)
-Gimp 1.2
- a matching sane along with xscanimage (though now used)
-iscan package installed and linked as a gimp plugin, from epkowa: http://www.epkowa.co.jp/english/linux_e/index.html
-littlecms library from:
http://www.littlecms.com/
-gimp color management plugins from:
http://www.freecolormanagement.com/color/gimp.html
(do not give a damn for the "alpha" status) ... but make sure to copy all icc profiles into /usr/lib/color (hardcoded in main.c ... I guess you can name some files to: scanner.icm, printer.icm and display.icm (if you have according inputs) ... but I way able to install some scanner icm's under different names and was able to select them anyway
- grab some scanner drivers for Windows from the Epson site. The EXE Files are self extracting Zipe-Archives, so you can simple do an unzip .exe
- look for the *.ic_ files
- install msexpand (package mscompress) from your distro (or from here: ftp://ftp.penguin.cz/pub/users/mhi/mscompress
- then do an expand .ic_ and rename the result to
- alternative
+ if no *.ic_ files found, look for *.cab files
+ install cabextract from your distro (look here http://freshmeat.net/projects/cabextract/
+ look for *.icm file
- the file program should tell you that the icm-file is a:
# file Per317_r.icm
Per317_r.icm: Kodak Color Management System, ICC Profile
First I installed a icm for the scanner for prints mode and a seperate icm file for slides/negatives mode. Per317_r.icm for prints (r=reflective) and
Mine were Per317_t.icm (t=transparent) -- that made the most sense.
If you have access to a Windows Box or do run wine you can grab the following windows program (ran under wine pretty flawlessly):
iccinspect.exe from: http://www.littlecms.com/iphoto/inspect.htm (Open Source !)
This program enables you to closely examine the contents of the icm files. It prints shows all the stuff like vendor and color-space infos and whatnot :-)
Second I scanned a picture with the scanner and applied the input icm to the scan with using the gimp (There is a command line tool for non-gimp users).
Then I adjusted my monitors settings, resulting in my setting the color temperature to 9300K since that resembled the picture the best.
I tossed in the Windows own icm file "sRGB Color Space Profile.icm" that I found under C:\Windows\system\color (Windows 98) as sRGB.icm and monitor.icm into /usr/lib/color so I have some kind of setup.
If you do have a disk with the correct icm file on it, use that file instead. Only use sRGB.icm if you want to exchange files with others.
Third I send a icm file for my dads printer (an epson :-) ) to my dad and he just tossed it into /usr/lib/color (as root) and has not complained yet :-D
I keep on ignoring slight color glitches while working on the pictures since I beleive the correct colors will come out of the printer anyway ...
---------
There is plenty of more reference out there:
http://www.efg2.com/Lab/Library/Color/index.html
---------
But feel free to ask if nessecary. -
Re:How about the other way around
-
Re:Commodity Hardware
If it was only the hardware that was evolving, then yes this would be the case.
This is unlikely to be the case. Should it get to the point where the hardware is too many and too complicated for everybody to program for, you'll find generic interfaces to the hardware being implemented in generic assembly languages like table assembly.
Or perhaps firmware will develop further to ease driver creation.
There are many areas in which layers can develop to keep developing drivers possible for mortals. The industry isn't going to make things too difficult for itself.
Ever since the computer industry started, each extra level of complexity has seen an extra layer added to keep software creation manageable, and drivers are no exception.
Whilst Kristian's fears have foundation, there'll be ways around them even if it's as extreme as hobby OSes having to cooperate on driver development. -
Re:It's freaky
A 2.2 Ghz P4 (I assume) is not 88 times faster than a 25 MHz 386/486. You're buying into the megahertz myth; the modern processor has many more optimizations that make it more faste.
Yep, current CPUs do much more in one clock than 386. For example, 386 took 9-38 clocks for a single 32b multiply. Even P3 can do effectively (pipeline) a single 32b multiply in 1 clock and floating point multiply in 2 clocks (or better if you use 'multimedia' instructions). Athlon can do both in 1 clock (in fact, it can do two loads, add and multiplication with floating point numbers in one clock). I couldn't find timings for P4, but I think it can do integer multiplication in one clock but requires quite more for floating point multiplication.
-
Re:System recovery or Roxio GoBack
Ohh Jesus, WinXP home. Well, I have no advice for you!
Dude, you should have more advice, you should tell me to boot linux on a floppy, create an ext2fs partition and use the NTFS tools provided by www.winternals.com and www.sysinternals.com from a Win95 partition, or you could have referred me to the excellent Penguin archives. If you give up that easily you can't be a linux user ;-) But then again I also couldn't be bothered to set up a new partition just for getting a few Megs of files, especially since a power failure in the middle of a partition-resize or partition-move (even in PartitionMagic) wipes the partition (maybe all the partitions). Hmmm time to remove that Micro$oft bumper-sticker.Now that I've been told I lose my job or go into sales 'n' marketing of the software I just coded, I don't think my little file recovery matters any more. Ah well.
-
Re:Key points for Windows/Outlook users
If you are using Windows or Outlook, do not open an attachment if you don't know what it is.
How about Dual Boot systems? When running Windows to play games or to see how l4ame it is, a virus/worm/trojan would have root access to the linux partition under "fun" OSes like Win'98 or even Administrator on Win 2k and Win XP and could rootkit your linux. This software already does this sort of access. I don't think any inter-partition viruses exist right now. Hmmmm -
Why not dual boot?I hope this suggestion doesn't detract too much from your original intent to run Linux, but why not dual boot? Compilers under wine tend to be iffy, and there's no good way that I know of to cross-compile for Win32 using your toolset.
For a while I used VMware on my trusty U2W-SCSI PII box and found the whole experience a little bit too sluggish for daily use (I tried just about every flavor of Windows I could, including 2000). The processor benchmarked at around 333mHz inside the VM, but memory allocation and disk I/O on the virtual disk dragged down performance (yes, even on 10,000 RPM Ultra2 SCSI).
Here's my proposal, assuming you have sufficient disk space:- Divide the laptop's hard drive into at least 2 partitions; use the NT Boot Loader or LILO to dual boot between the two.
- Since dual booting every time you want to access another OS gets annoying, purchase a copy of VMWare. There's an option to mount physical partitions like virtual disks, so you should be able to access either NT from Linux or Linux from NT like you were booting to the partition (albeit slower). This could allow you to run both Linux and Windows applications at the same time, and would allow for an elementary form of filesharing using Samba and the vmnet kernel module. You get better I/O performance this way too, which you're going to want on a laptop drive.
- Finally, go grab some filesystem drivers that can read Ext2 and NTFS/FAT. Read-only NTFS (and experimental read/write) is already in the stable Linux kernel. The full list of supported filesystems for Linux is in the Module-HOWTO on your local howto site. For any other OS, check out the Filesystems HOWTO. This document is a "Filesystem map" that maps many types of filesystems to host OSs - it has just about every combination you could ask for. This will allow you to bypass VMware if you just want to view or edit files on the other OS's partition.
43rd Law of Computing: Anything that can go wr - Divide the laptop's hard drive into at least 2 partitions; use the NT Boot Loader or LILO to dual boot between the two.
-
Sun has a right to be worried
If you have ever seen Linux on Sparc you would agree. Slowaris cann't compare to Linux on Sparc. For more info on Sparc Linux check out this link