Pages: [1]   Go Down

Author Topic: Question for people who have CoolScan 4000 or V  (Read 9293 times)

AFairley

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1486
Question for people who have CoolScan 4000 or V
« on: May 09, 2012, 02:41:12 pm »

I have been trying to pick up a Coolscan V or 4000, but with each of the last 3 I have looked at (two Vs and a 4000), the Nikonscan software returns a "hardware error detected" message upon startup.  Also the preview and/or scanned colors have a significant color cast.  However, a scan made with Viewscan looks OK.  I have passed on each of them because if hardware is iffy, I figure it can only get worse.  As to whether what I am experiencing might be a software problem, I did some research on the web, and it seems that when you power on the scanner it runs a self-diagnostic during which the power light blinks.  The power light should blink slowly and steadily -- if the power light blinks rapidly during or at the end of the power up, the scanner there has been an error condition found in the the self-diagnostic.  This is the behavior I have seen in all three machines, slow blinking of the power light during the beginning and middle of boot, followed by rapid blinking at the end, with the power light on steady after boot is completed. Ergo I do not believe that what I am seeing is the result of a software problems.

My request to owners is, will you observer the light during the boot and tell me if it blinks slowly and steadily throughout the entire boot sequence before going to solid green.

(BTW, just to head some replies, I don't care if the scanner works in Vuescan, because I am not going to use it -- I want some of the features (analog gain and multisampling (for the 4000) that are only available through Nikonscan.  And I am not going to shell out the money for Silverfast, under any circumstances.)  All I want to do is confirm that the boot light sequence for the CoolScan V and 4000 does not include a change in blink speed at any point.

BTW, I have a CoolScan IV, which I use, so I am familiar both with the Nikon scanners generally, and with using Nikonscan with them.  So I'm highly confident that what I am seeing is not a case of user error.

Thanks for any and all help!
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
Re: Question for people who have CoolScan 4000 or V
« Reply #1 on: May 09, 2012, 03:40:17 pm »

I have a Nikon Super Coolscan 5000-ED. I just booted it up to test for you, and I find the light blinks steadily about once per second until the boot is complete and the light is steady on.

Nikonscan may or may not work for you depending on whether you are using a 32-bit or 64-bit operating system. For the former it will, for the latter it will not, and Nikon has vacated this line of business so there will be no upgrade. They tell users to use either Vuescan or SilverFast.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

AFairley

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1486
Re: Question for people who have CoolScan 4000 or V
« Reply #2 on: May 09, 2012, 04:12:20 pm »

Thanks for replying, Mark,

I have been some more googling, and what I am gathering is that a rapid blink at any time during bootup indicates a hardware self-diagnostic fail.  Apparently this can be caused by hardening grease on the transport rails, which would explain why a machine can give apparently good scans while still returning an error code (enough drag  to trigger an error but not enough to jump out in scans).  It also apparently can be caused if the scanner optics are dirty enough (though I would think that in that case you would see that the scans are pretty crummy).  But I am not willing to shell out hundreds of dollars on the chance that I can fix whatever the problem is.  The bummer is that an eBay purchase is ruled out since I want to test the machine before I buy, and they come up rarely on Craigslist, even in L.A.

BTW, last weekend I removed and cleaned the mirror for my IV, even though it did not look dirty from what I could see looking into the machine with the film holder out.  Turned out there was enough dust on it to noticably degrade the image, though I had not realized without an "after" image to compare.  I would recommend that Coolscan owners clean the mirror every couple of years, it's not very hard if you are handy enough to take the case off the machine and manipulate the mirror retainer.  Certainly if you picked up the scanner used.  Anyway, from now on, the dust cover stays on when not in use.

As to the Nikonscan software, I have the hacked scanners.inf file which has been letting me run the Vista version of Nikonscan on a Win 7 64-bit system.  Works flawlessly with the Coolscan IV.  The hack and instructions are widely available on the web.  The jury's still out on whether the color cast I have seen on the Vs and 4000 I've tried is software related or hardware related, since I haven't yet been able to hook up a V or 4000 that self-tests OK.  But I suspect hardware because other people are using the hacked s/w on Win 7 systems with no problems as far as I can gather.

Thanks again for responding.

Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
Re: Question for people who have CoolScan 4000 or V
« Reply #3 on: May 09, 2012, 04:32:46 pm »

Alan, this is very interesting. If an amateur can get this software going on a 64-bit computer with a freely available "hack", one really wonders - just as a nice corporate gesture to its customers upgrading their computers but not their scanners - why Nikon could not have done this and issued it as a software upgrade. OK, they would need to spend resources testing and supporting it, so I suppose once they decided to retire that activity, retire is retire and they're out of it.

Turning to the colour cast, it could be hardware or software related. Perhaps the indicator lamp behaviour is symptomatic of a hardware issue affecting colour balance, or perhaps it is a profiling issue. Some questions that might help a diagnosis: Is the cast systematic on all scans of all kinds of media, or does its nature vary quite a bit? Did you at any time profile the scanner? If so, how and how long ago, and has anything about your operating environment (that could affect a profile's performance) changed since you last profiled it? Have you made intensive use of the device since you last profiled it? What software produced the scans with the cast? Does the software have colour management settings that may have changed somehow?
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

AFairley

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1486
Re: Question for people who have CoolScan 4000 or V
« Reply #4 on: May 09, 2012, 05:13:18 pm »

As far as I can tell (I am not that well versed in this stuff), all the .inf hack does is allow Win 7 to recognize the Vista .dll as the driver file.

I can't investigate the color cast issue further, because I don't have any of the scanners it came up on.  But it's way too severe to be a profiling issue, IMO, it's almost like one of the color channels dropped out.  On the only machine I had long enough to play with, both the preview and the scans were off using Nikonscan, but both the preview and the scans looked OK, color wise, using VueScan.  That I can't figure out unless VueScan was doing some kind of auto color balance (in addition to the media type selection) under the hood.  I haven't used VueScan enough to be really familiar with it.
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
Re: Question for people who have CoolScan 4000 or V
« Reply #5 on: May 09, 2012, 05:38:01 pm »

You can get very severe colour casts in both the preview and the scan with the wrong scanner profile - even between scanner models right within the Nikon family. I've constructed such things on purpose to demonstrate the importance of profiling a scanner and then using the correct profile.

This sounds very much to me like a software problem - specifically NikonScan was not properly set-up for colour management, while Vuescan was. Choice of media-type should not be the issue, because the difference between selecting Kodachrome versus another transparency film as the media type would cause some colour disconnect but not nearly as dramatic as you describe you encountered.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

AFairley

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1486
Re: Question for people who have CoolScan 4000 or V
« Reply #6 on: May 09, 2012, 05:53:50 pm »

Thanks, Mark.  I will keep that in mind.  Having said that, until I find come across a Coolscan V or 4000 that does not fail the boot self-test, the issue is moot.  Once I do, I'll know if it was/is a software issue or not.  I don't have any color management problems with the same hacked software and my Coolscan IV, so that's not an issue at present.
Logged

Geo Andy

  • Newbie
  • *
  • Offline Offline
  • Posts: 1
Re: Question for people who have CoolScan 4000 or V
« Reply #7 on: July 08, 2012, 08:50:09 pm »

Hi:

Just as an FYI, I have two Nikon Coolscan 4000 EDs. One is giving me the symptoms described above (color cast, rapid flashing LEDs on film insertion, and error msg). The other works just fine when swapped out on the same Mac. So, I going to open it up (already tried POST procedure) and clean it up and see if that helps. Fingers crossed, but a driver/software issue seems to have been ruled out.
Logged

AFairley

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1486
Re: Question for people who have CoolScan 4000 or V
« Reply #8 on: July 09, 2012, 01:12:44 pm »

I finally found a very clean working Coolscan V.  However, the Kodachrome scans exhibit bad highlight blooming when they are lightend significantly (either in the NikonScan software or in ACR) and some Googling and perusing extensive and well-informed discussions on the issue tells me that this present in many Coolscan models as a result of the design of the scanner (plastic rather than coated optical glass covering the CCD, if I recall correctly).  (I see the same thing to a lesser degree on my Coolscan IV -- and with both scanners, this is after cleaning the mirror and front surface of the lens.)  After doing testing, I have discovered that I can use my Olympus E-M5 and 50mm ZD macro lens to get files that with the proper care print almost the same as the Coolscan files (I mean really really really close at 17x22), and exhibit better color fidelity to start.  So I am bailing on the scanner route for now.  If I can get such good results with full 2:3 chromes usng a 4/3 sensor camera, I expect that one of the current APS sensor cameras (where you will not lose any sensor area because of the format mismatch) would smoke any of the 4000 dpi Coolscans.  Anyway, it's been quite a trip....
Logged
Pages: [1]   Go Up