[sane-devel] update

2001-11-15 Thread red...@emphanet.com
Ok..

Update on this:

Here is what I have noticed.  The following chart shows what width(x) per dpi
resolution (max), or else scanner aborts or sane receives EOF.

DPI Width Max (x)

300 2.80
295 2.85
290 2.89
285 2.95

and so on...  Sounds to me like there's a limit of data to be fed in some
operation here.  It seems to be a direct formula (+/- a few).  Since the Scanner
collects most of its data in the X coordinate (per line) as apposed to data in
the Y coordinate, the limitations seem to be on how much data per line can be
collected.  It sucks that this setup can't collect 300 dpi at a full page width
of 8.50".  In order to stay within its limits, I have to drop the dpi to 105 to
get a full page width of 8.50".  Is this a data governing problem?  buffer? 
what could this be?  Again.. help..  I'm going to bed now.. night.

Jordan



Previously what I said:

I have a Umax Astra 610S.. Seems I've had all kinds of troubles.. but here's
what I just found.

In the Geometry window from Xsane, I can set the Y variation to anything and it
will scan at 300 dpi.. but as far as the X variation, it seems I can only go
2.80 units (according to the Xsane Geometry slider bars), otherwise it won't
finish the scan.

Note:  I think 2.8 units is 2.8 inches..  not sure..  but.. anything wider than
that will cause the scanner to get an EOF.

Any thoughts?  This is more and more sounding like a library.. but I could be
wrong.  I don't think this is a mechanical problem as I can scan on any part of
the page as long as it is 2.80 units wide or less (x)..  Any measurement is fine
with height (y).

This is a curve in my progress.  

Help!  *

Jordan Peterson



[sane-devel] wow..

2001-11-15 Thread red...@emphanet.com
Ok..

I have a Umax Astra 610S.. Seems I've had all kinds of troubles.. but here's
what I just found.

In the Geometry window from Xsane, I can set the Y variation to anything and it
will scan at 300 dpi.. but as far as the X variation, it seems I can only go
2.80 units (according to the Xsane Geometry slider bars), otherwise it won't
finish the scan.

Note:  I think 2.8 units is 2.8 inches..  not sure..  but.. anything wider than
that will cause the scanner to get an EOF.

Any thoughts?  This is more and more sounding like a library.. but I could be
wrong.  I don't think this is a mechanical problem as I can scan on any part of
the page as long as it is 2.80 units wide or less (x)..  Any measurement is fine
with height (y).

This is a curve in my progress.  

Help!  *

Jordan Peterson



[sane-devel] UMAX Astra 610S (better specs)

2001-11-15 Thread red...@emphanet.com
Hi again.

ok.. my last posting about the 610S was long and .. well..  tiresome.. here's
the details to end that.

I'm running RH 7.2, Kernel 2.4.9-13, using a BusLogic SCSI host BT-930 (SCSI-2
50HD).
The chain on this bus is as follows: 
HP-9600E BURNER -> Castlewood ORB (removable 2.2GB disk drive) -> UMAX Astra
610S scanner.
>From Host card: 50HD -> BURNER -> 50HD -> ORB -> 50HD/25DB -> SCANNER.
SCSI card module BusLogic.o
Sane-Backend 1.0.6
Xsane frontend 0.80

Problems are:

Can't scan over 105 dpi in Color mode, but can scan 300 dpi in LineArt.
If attempting to scan over 105 dpi in color mode, it starts the scan and returns
with apx 1" or less of a graphic.  When in color mode, 105 dpi or less, the scan
completes the whole scan.  Check this out.  If I use the PREVIEW part of Xsane
and crop a section of the scan, the scanner moves to the start of that cropped
area, begins to scan and returns with that 1" or less blip of an image.  I get
no errors, that I know of.  Correct me if I'm wrong, but I think it's only
scanning one line and aborts, or gets a complete signal.  I don't know.  Here is
some information:

/proc/scsi/scsi --
Attached devices: 
Host: scsi0 Channel: 00 Id: 01 Lun: 00
  Vendor: UMAX Model: Astra 610S   Rev: V1.3
  Type:   Scanner  ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 04 Lun: 00
  Vendor: HP   Model: CD-Writer+ 9600  Rev: 1.0a
  Type:   CD-ROM   ANSI SCSI revision: 04
Host: scsi0 Channel: 00 Id: 06 Lun: 00
  Vendor: CWS ORB2 Model: -SE U ID 6   Rev: D20 
  Type:   Direct-AccessANSI SCSI revision: 02
--

cat /proc/scsi/sg/* --
0
dev_max(currently)=9 max_active_device=3 (origin 1)
 scsi_dma_free_sectors=112 sg_pool_secs_aval=320 def_reserved_size=32768
32768
host
chan
id
lun
type
opens
qdepth
busy
online
0
0
1
0
6
0
3
0
1
0
0
4
0
5
0
3
0
1
0
0
6
0
0
0
3
0
1
UMAXAstra 610S  V1.3
HP  CD-Writer+ 9600 1.0a
CWS ORB2-SE U ID 6  D20 
uid
busy
cpl
scatg
isa
emul
54272
0
3
128
0
0
0
0
12
64
0
1
BusLogic BT-930
iSCSI (2.0.1.8)
30120
Version: 3.1.20 (20010814)


cat /proc/scsi/BusLogic/0 --
* BusLogic SCSI Driver Version 2.1.15 of 17 August 1998 *
Copyright 1995-1998 by Leonard N. Zubkoff 
Configuring BusLogic Model BT-930 PCI Ultra SCSI Host Adapter
  Firmware Version: 5.02, I/O Address: 0xD400, IRQ Channel: 9/Level
  PCI Bus: 0, Device: 9, Address: 0xDB00, Host Adapter SCSI ID: 7
  Parity Checking: Disabled, Extended Translation: Enabled
  Synchronous Negotiation: FSF#, Wide Negotiation: Disabled
  Disconnect/Reconnect: Enabled, Tagged Queuing: Enabled
  Driver Queue Depth: 255, Scatter/Gather Limit: 128 segments
  Tagged Queue Depth: Automatic, Untagged Queue Depth: 3
  Error Recovery Strategy: Default, SCSI Bus Reset: Enabled
  SCSI Bus Termination: Enabled, SCAM: Disabled
*** BusLogic BT-930 Initialized Successfully ***

Target 1: Queue Depth 3, Asynchronous
Target 4: Queue Depth 3, Synchronous at 10.0 MB/sec, offset 15
Target 6: Queue Depth 3, Synchronous at 10.0 MB/sec, offset 15

Current Driver Queue Depth: 255
Currently Allocated CCBs:   28


   DATA TRANSFER STATISTICS

Target
Tagged Queuing  Queue Depth  Active  Attempted  Completed
==
==
===  ==  =  =
   1Not Supported 3 01  1
   4Not Supported 3 0 1060   1060
   6Not Supported 3 06  6




Target  Read Commands  Write Commands   Total Bytes ReadTotal Bytes Written
==  =  ==  ===  ===
   1  0  0  00
   4  0  0  00
   6  1  0   10240

Target  Command0-1KB  1-2KB  2-4KB  4-8KB 8-16KB
==  ===  =  =  =  =  =
   1 Read0  0  0  0  0
   1 Write   0  0  0  0  0
   4 Read0  0  0  0  0
   4 Write   0  0  0  0  0
   6 Read0  1  0  0  0
   6 Write   0  0  0  0  0

Target  Command   16-32KB32-64KB   64-128KB   128-256KB   256KB+
==  ===  =  =  =  =  =
   1 Read  

[sane-devel] Umax Astra 610S

2001-11-15 Thread red...@emphanet.com
Howdy folks..

Ok.. I took back the HP ScanJet 4300Cse yesterday and borrowed a UMAX Astra 610S
scanner instead, being that I'm needing to import images in for this lady's site
in a rush.  Anyhow.. I get it home and plugged into my BusLogic card and
everything checks out ok.  I scan a few images at 150 bpi in color no problem. 
Then after a few test scans (and being impressed that it's alive at last), I get
problems.  I notice that I can't scan over 95 dpi.  If I set it for a dpi over
95, it scans about an inch or less and it calls it done, spitting out an image
of only an inch or less of data.  Of course it scans the whole page no problem
if I set it for 95.

So I'm thinking:  Ok.. I wonder if the data connection between the scanner and
the rest of the SCSI bus is slow and getting buffer underruns, etc.  So I bypass
 my ORB and HP-9600 cdwriter and plug in the scanner directly into the BusLogic
SCSI host.  ok.. great.. now I can scan at 105 dpi (cool!  better than 95)..  So
I'm scratching my head and wondering what's up here.  For kicks I turn the SCSI
ID on the Scanner to 1 instead of 5 (for which the ORB drive is 6, and the
Burner is 4).  Then I plug everything back into the original chain (sequence). 
>From the Host adapter the chain is as follows:  HP-9600 burner, then ORB, then
UMAX scanner.  Ok.. another note:  The Umax Astra 610S uses a DB25 pin (looks
like a parallel port) setup.  My SCSI bus uses DB50(HD).  As you have guessed, I
have one of those cords that adapts from 50pin HD to 25 pin (for the scanner). 
I'm wondering if maybe the cord that I'm using is low grade or too long.  It's a
6 footer. 

Another thing.. I updated SANE-backend to 1.0.6 (since I noticed changes to a
few scanner libraries on the SANE web site (http://www.mostang.com/sane).  Hey..
just incase, right?  Also, updated Xsane to 0.80 (http://www.xsane.org).  But I
still have the same problem.  In all cases, I'm still using the same BusLogic
SCSI card and same cables.

If there are any SCSI pros out there who read this article and shake their head
in discuss of my stupidity, do share.  I'm not the pro with the SCSI specs. 
SCSI cables are expensive.  I just want to make sure I need a better cable
before I go and spend another lump sum of money on it.  I thought it was rather
odd that the scanner cable that I just bought is sort of thin for a scsi cable.
 They're usually pretty thick, but this one is dinky compared to the 50HD <>
50HD cabling I have for the ORB and BURNER. 

In Summary, I'm beliving that my dpi limitations are a direct issue with the
data transfer capabilities to the scsi host.  I have my doubts, but I gotta know
for sure if this is the issue.  LIke I said, cabling is expensive and I'm in no
shape to buy something I don't need.  I've researched the issues with the Astra
610S and I haven't seen a similar problem.  Another guy posted an ad about
having problems getting 300 dpi with LINEART setting...  He said his COLOR 300
dpi was fine, but max he could go in LINEART was 150 dpi.  Beats me.  This is
strange.  

Guess I otta send this off so I can get some answers hopefully.  

I hope sane becomes more savvy with scanners in the future.  What happens when
windoze finally dozes off?  Everyone will hit linux next right?  I hope it's
ready.  Some standardization (if a word) would be very cool with linux.  Having
to give each distribution special attention becomes time consuming.  I really
don't want to setup a seperate machine for windoze to just use a scanner here,
or a printer there.  Alright.. enuff of that, right?  I'm on it.

Any help would be fantastic.

Jordan Peterson



[sane-devel] HP ScanJet 4300C

2001-11-13 Thread red...@emphanet.com
Hi!

I've read that there isn't much of a backend for the HP ScanJet 4300C yet. 
Please let me know if there's more to know about it.  That would be awesome.

Sincerely,

Jordan