Matthew, thank you for your help. The scanner now works in regular mode. Interesting enough, it freezes in force_nibble mode.
For those who have problems with Canons N340P and FB330P. 1. Set manually the io ranges and dma mode in BIOS For me these were 378-37F, 778-77D, IRQ 7 2. Set ECC+EPP mode for your parallel port in BIOS. 3. Add the following line to the modprobe.conf (/etc/modeprobe.d/options on Ubuntu) options parport_pc io=0x378 io_hi=0x778 dma=1 4. Reload the parallel port related modules using modprobe (my choice would be to restart) 5. If the scanner still does not work, try putting a comment on force_nibble/removing the comment in canon_pp.conf. ---------------------------------------- > CC: sane-devel@lists.alioth.debian.org > From: sta...@guarana.org > Subject: Re: [sane-devel] Canon_pp freezes halfway > Date: Wed, 29 Nov 2006 23:33:15 +1100 > To: ksel...@hotmail.com > > Hi there! > > On 29/11/2006, at 11:22 AM, selma kapetanovic wrote: > ... > > I am running > > .. > > * Parallel port is in EPP mode (BIOS setting) > > I'll get back to this... > > > It scans most often (not always) in black and white. For color mode > > the number of "reads" varies from two to four. > > Hmm.. that's quite weird, but I'm guessing it's just the increased > amount of data in colour mode. > > > I spent quite some time trying to make it work. Should I have known > > it takes so much time, I would toss the scanner in the garbage. > > At this point it is more of a challenge. I know that I am not the > > only person who has the problem and I want to post the solution to > > the Ubuntu web site along with a recommendation not to buy canon > > scanners. > > Well, I'm probably not telling you anything you don't know, but it's > not so much the scanner as the port. > > Parallel ports are flakey things - it's amazing they ever work > really. They're just not really designed to push the large amounts > of data scanners need to move. > > > ... > > [canon_pp] detect_mode: Port supports ECP-H. > > [canon_pp] detect_mode: Port supports interrupts > ... > > So according to the probing done by libieee1284, the port supports > ECP-H transfers and interrupts, but no mention of DMA. Generally > speaking I've had the most success in the past using ECP and DMA mode > or pure software mode, but pure software is very slow and nasty - DMA > is much preferred if you can get it working. > > Anyway, I think the first thing to try is setting the BIOS setting to > "ECP+EPP" if it has such an option, instead of "EPP". After that you > should maybe allocate a fixed IRQ to the parallel port (yes, I know, > plug&play - but it's just more reliable this way!). Once you've done > that you can try playing with the parport_pc module settings to try > and get DMA transfers working. I wish there was an easier way, but > that's pretty much it. > > Hope that helps! > > Cheers, > > - Matthew _________________________________________________________________ Be one of the first to try Windows Live Mail. http://ideas.live.com/programpage.aspx?versionId=5d21c51a-b161-4314-9b0e-4911fb2b2e6d From stef....@free.fr Thu Nov 30 06:48:43 2006 From: stef....@free.fr (=?iso-8859-2?q?St=E9phane_VOLTZ?=) Date: Thu Nov 30 07:13:32 2006 Subject: [sane-devel] Plustek ST24 In-Reply-To: <4567a43b.2080...@gmail.com> References: <4567a43b.2080...@gmail.com> Message-ID: <200611300648.43911.stef....@free.fr> Le samedi 25 novembre 2006 03:02, Tymoteusz.Drozd a ?crit?: > hi all > recently i've discover that in backend/gensys_devices.c is placed sort > of values for ST24. > iv done small change in backend/gensys_devices.c:728 from > COLOR_ORDER_BGR to COLOR_ORDER_RGB and in same file at line 726 value > 0,8,16 to 0,18,30. > that gives correct colors on my plustek st 24 and good correction of ccd > line-distance. Sadly only for resolution > 600dpi which has wrong number of engine steps what can leads to damange > of hardware. For simple workaround i limited y size of scan area to 1/4 > of nominal size but its givs 600dpi x150dpi :/. > If sobody have any ideas how can i correct number of steps for > resolutions larger then 150 dpi plz msg me. > Also i have fears that corect number of steps dosnt resolve problem with > ccd line distance in resolutions difirent then 600 dpi(for 150 dpi those > values are difirent 2,18,32 perhaps). > Any ideas are welcome :) > > ps. sorry for my bad anglish ;) its 03:00 am and anglish isnt my native > lang :> Hello, the only way I know to find steps values and other settings is to record USB activity under windows. Then study it to understand what's going on. I use usbsniff and a few custom scripts to turn raw logs in 'higher level' format with genesys registers values and slope tables decoded. I can send them to you if you wish, or I can try them on a ST24 sniffed log to see if they need to be adapted. Regards, Stef
