Apache-Talk @lexa.ru 

Inet-Admins @info.east.ru 

Filmscanners @halftone.co.uk 

Security-alerts @yandex-team.ru 

nginx-ru @sysoev.ru 




      :: Filmscanners
Filmscanners mailing list archive (filmscanners@halftone.co.uk)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: filmscanners: NikonScan 3.1 does not work under W2K (report ontests)

Thanks Stuart,

You've gone to a lot of trouble and we appreciate it.


----- Original Message -----
From: "Stuart Nixon" <sns@ermapper.com>
To: <filmscanners@halftone.co.uk>
Sent: Sunday, October 14, 2001 8:39 PM
Subject: filmscanners: NikonScan 3.1 does not work under W2K (report on

> Hi.
> I am sure you are aware that people have reported NikonScan 3.1 is
> unable to scan - especially batching multiple scans - without crashing.
> With rare exceptions, NikonScan 3.1 crashes for Windows 2000 users.
> My testing and feedback from other users has confirmed
> that NikonScan 3.1 does not operate correctly under Windows 2000.
> This software should never have been released in its current buggy state.
> Please feel free to distribute this report to Nikon Japan, Europe
> and USA technical support, and also to people who have performed
> reviews of Nikon scanners.  Perhaps we can finally get Nikon to
> get serious about fixing their software.
> Background:
> Some weeks ago I asked for feedback on running NikonScan 3.1
> under Windows 2000.  Many people responded, and quite a few
> were kind enough to provide additional feedback when I requested,
> including Joe Blaze, Paul Graham, Bob Kehl, Tom Scales,
> Charles Volkland and many others. Thank you.
> A couple of people reported that NikonScan 3.1 never failed
> for them.  There was no particular commonality on those systems that
> worked - one was a fast AMD single CPU, and two others were slower
> dual CPU Intel chips.  About the only interesting thing was that
> two of the working systems did not use the Nikon supplied Firewire card.
> Everyone else (about 80% of responses) reported that NikonScan 3.1
> failed constantly for them under Windows 2000, for single or dual CPU
> machines.  Most people reported having swapped to other products such
> as VueScan (which incidentally is rock solid from what I have seen),
> or if they really *had* to use NikonScan, ran it from within PhotoShop,
> scanning a single photo at a time.  Because Photoshop restarts the
> scanner software each time, this reduced the crashes to "only" 1 out 10
> scans or so.
> Ways to reduce NikonScan 3.1 crashes:
> - Use something else. If you are serious about bulk scanning,
> then basically NikonScan 3.1 is a waste of time unless you are
> one of those very rare & lucky people to have it run without crashing.
> VueScan and Silverfast are two alternative scanning products.
> - If you can, run Windows 98 on the scanner PC instead
> of W2K.  NikonScan seems more stable under W98.
> - Run a slower PC.  It is more stable on PCs less than 350Mhz or so.
> - Set the Temp drive *and* the scanner output location to a network
> drive. This will reduce NikonScan crashes by about 50%.  So if you
> are lucky, you might get 6 scans instead of 3 before it crashes.
> - On dual CPU machines, lock NikonScan to run in UniProcessor mode.
> This will reduce crashes somewhat.
> Testing performed:
> Given Nikon has been aware of these crash problems since NikonScan 3.0
> and has not fixed them, I wanted to see if the problem could be pinned
> down and perhaps a work around could be found.
> My testing basically showed that for all configurations I could try,
> NikonScan 3.1 is unstable.  I did manage to get it to batch scan 6 slides
> just once without crashing. Big deal :-(
> - Coolscan 4000ED, using the Nikon supplied Firewire controller.
> - Three different Windows 2000 machines.  Each machine had 1GB RAM,
> IDE drives, and dual CPUs.  However, each machine had a different
> motherboard, and different CPUs. I tried ASUS, Intel and Tyan
> motherboards, and Intel P3, P4 and AMD Athlon MP CPUs. Each
> ran different nVidia cards (2's and 3's). I also tested
> against a 4th machine that had been heavily used for processing
> jobs that ran into CPU-weeks, and was known to be very stable.
> Machines were 733Mhz, 933Mhz and 1.2Ghz.
> Significantly, each machine had very different motherboards
> and add-on cards from different manufacturers.
> - Each machine was clean loaded within Windows 2000 Server.
> The latest motherboard/IDE/etc drivers were updated, then SP2,
> then the latest security patches.
> - Only Photoshop 6, and NikonScan was installed on each machine.
> No printers were installed (earlier versions of NikonScan
> had problems with different printers installed).
> Testing was to simply run NikonScan stand alone, turn on ICE but
> not GEM/ROC, and try and scan a 6 photo positive strip out to
> TIFF files.
> Before giving up in disgust, I tried various other combinations:
> - Forcing NikonScan to run on one CPU
> Result: Crashed less often (every instead of 3 slides)
> - Temp and output directories network not local:
> Result: Crashed a bit less often
> - Forcing scanner driver interrupts to a single CPU:
> Result: No difference, still crashes
> - Running Windows 2000 as a UniProcessor build (so only one CPU):
> Result: No difference, still crashes
> - Trying different NikonScan options (no rotate, no ICE, etc)
> Result: No difference, still crashes
> I don't think there is much point in any more testing.  NikonScan 3.1
> is broken, at least for me and most other people using Windows 2000,
> and enough time has been wasted on it.
> Please note that some people *have* reported that NikonScan runs
> robustly for them (generally W98 users and some rare W2K users).
> So NikonScan just might run for you.  If it does, the rest of
> us would sure like to know what you did to make it run. Note
> it will generally complete scanning for a single slide, especially
> if used inside Photoshop - scanning multiple photos is where the
> trouble starts.
> Once again, I encourage distribution of this information to
> Nikon people to try and get this software sorted out.
> Apologies to the list for the length of this email.
> Regards
> Stuart

Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.282 / Virus Database: 150 - Release Date: 9/25/01


Copyright © Lexa Software, 1996-2009.