Reflecta ProScan 7200 - Vertical Banding Issue

Fragen/Themen zu Reflecta-Filmscannern
Forumsregeln
In diesem Forum dürfen keine Beiträge mit gewerblichen Angeboten, privaten Verkaufsangeboten oder Links zu gewerblichen Seiten eingestellt werden.
D.Page
Beiträge: 11
Registriert: Mittwoch 12. Juni 2013, 11:03

Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von D.Page »

Is it normal for scans made on the Reflecta ProScan 7200 to have subtle vertical banding over plain areas, such as sky, particularly with denser negatives?

I have just bought this scanner, and have noticed that my (particularly denser) film negatives appear to have some vertical banding, mostly on the left-hand side of the frame. It is only noticable with plainer areas, such as sky. In order to try to establish whether it is the scanner or the negative causing this issue, I scanned one frame of film normally, and then scanned the same frame of film but inserted it into the film strip holder upside down, and then rotated this image 180 degrees in Photoshop so that it appears the correct way up. The same vertical banding now appears on the right-hand side of the frame.

What is slightly strange is that I cannot say that I have found any noticable vertical banding when scanning using the supplied Cyberview X5 software, only when using SilverFast Ai Studio 8 software (where I have the demo version installed, at present) can I notice any of this vertical banding. Having said that, the Cyberview scans are relatively pale and lack contrast, so such banding would be far less noticable.

By the way, I am using the latest 'revised' film strip holder which comes with this scanner, which now incorporates bevelled edges, and, I believe, a more matte-type finish, due to a well-documented problem of bad vertical streaks appearing on the extreme left and right hand sides of scans when using the original film strip holder. I do not believe that my particular issue is related to this. The film strip holder issue was caused by reflections bouncing off the shiny film strip holder (the revised holder now incorporating bevelled film frame edges, and also an overall matte-type finish, as far as I can see), and my issue, I would presume, is being caused by some sort of problem with calibration.

I am in the process of getting my scanner replaced, but I am unsure as to whether I am to expect a certain amount of this vertical banding with all ProScan 7200's, due to the calibration aspect of these scanners never performing 'perfectly', and certainly not to the same standard as, say, your average high street lab.

Let me be clear - this vertical banding is subtle, and you have to look fairly carefully to see it, but once you know it's there, it's hard not to notice it, if you know what I mean.

Can anybody please help?

I have SilverFast Ai Studio 8 and Cyberview X5 scanning software installed, all with the latest firmware and drivers installed.
I am using Windows XP.
Benutzeravatar
Jossie
Beiträge: 1948
Registriert: Sonntag 10. Februar 2013, 19:40
Wohnort: im Kraichgau
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Jossie »

Good evening,

although I do not have the same scanner as you do, I see a phenomenon, which might be related to yours. I see the effect in scans with both SilverFast and VueScan, so I think it is the scanner and not the software.

I have scanned the same slide twice, without touching slide nor scanner. Then I have subtracted the two images one from the other. I know that the hardware of the DD6000 is not accurate enough as to expect a zero image (modulo noise, of course). But there was not only the structure of the image visible, but in addition I see vertical stripes of varying noise: One stripe is of low noise, as if the 2 images were almost perfectly aligned. Then there is a stripe with considerably higher noise, as if the 2 images had a larger offset, etc. --- at least this is my current interpretation. I attach a part of the difference image for your information.

I have sent a copy of the difference image to PIE, the builder of the scanner, and have been told, that "the engineers have not performed such tests and will need to do so before providing any additional information." So I wait for their explanation.

I could imagine that the stripes you see and those in my difference image might have the same (hardware type of) origin. I understand, that this does not help you to solve your problem, but it may shed additional light on it.

Best wishes

Hermann-Josef
Dateianhänge
Difference image of 2 scans of same slide showing vertical stripes of varying noise level.
Difference image of 2 scans of same slide showing vertical stripes of varying noise level.
Streifen_Demo.jpg (195.66 KiB) 28287 mal betrachtet
DigitDia6000 (CyberView, SilverFast Archive Suite 8 & 9) / CanoScan9950F (ScanGear, VueScan Pro), Eizo CS240, xrite i1studio, Win11 (64bit), Intel i9 (3.4GHz), Speicher 64GB, Nvidia Quadro P2000
D.Page
Beiträge: 11
Registriert: Mittwoch 12. Juni 2013, 11:03

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von D.Page »

Interesting. Thanks for that, Jossie.

I wonder how much of a problem your issue is when scanning, generally, though, and how much it bothers and preoccupates you. I am a perfectionist and will notice things, maybe, many people will not be bothered about.

One thing I should have mentioned is that I conducted an experiment, whereby I scanned a completely black area of a negative (the black area within the couple of frame-widths of black at the very start of a roll of processed negative). I scanned this "frame" of black, which obviously scans as a completely white image. Then, in Photoshop, I played about with 'levels' - and there it was, I can make out a series of ever-so-slightly darker vertical bands, mostly on the left-hand side. They seemed to correspond with the bandings on the scanned images, when lining them up and looking carefully between the two scans, in Photoshop. Seems to me to be a calibration issue. The thing is, though - are they all the same, to some degree - and, in other words, am I just expecting too much of these scanners, or was there a problem with just the one I had?

I will be getting the replacement ProScan 7200 fairly soon.

I'll post on here my findings.
D.Page
Beiträge: 11
Registriert: Mittwoch 12. Juni 2013, 11:03

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von D.Page »

I received the replacement ProScan 7200, and immediately noticed that scans show one prominent vertical banding, on the right-hand side of scans. I am at a loss to know what to do, as this is the second ProScan 7200 to suffer from vertical banding issues.

Please see the two attachments - they both show cropped areas of my scans to show the problem -the banding appears in exactly the same place in the frame, each time.

Is it possibly a batch problem with the ProScan 7200?

This banding is so prominent that, unlike my first ProScan 7200, this banding can be seen through 'busy' areas of the film frame as well as plain areas such as sky. Also, it can clearly be seen without me needing to darken the image.

I really like the ProScan 7200, but I cannot tolerate this banding.

I don't know what to do, at the moment. Any help would be much appreciated.
Dateianhänge
2.jpg
2.jpg (111.6 KiB) 28210 mal betrachtet
1.jpg
1.jpg (84.05 KiB) 28210 mal betrachtet
Samoreen
Beiträge: 13
Registriert: Montag 27. August 2012, 15:54
Wohnort: Samoreau, Frankreich
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Samoreen »

Hi,

Same problem here. I have sent the negatives to a professional lab using a Kodak HR500, so I'll be able to compare the results. This is very annoying but I'm afraid this is a limitation of the ProScan.
Patrick
Benutzeravatar
Jossie
Beiträge: 1948
Registriert: Sonntag 10. Februar 2013, 19:40
Wohnort: im Kraichgau
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Jossie »

Hi,

the problem I see in my scans is different from yours. The stripes I see are along the CCD-direction and are due to a problem with the stepper motor. Reflecta confirmed this diagnosis and my scanner is now on the way to Rottenburg for repair.

Am I correct that the bands you see are perpendicular to the CCD, i.e. in the scan direction? If so, and as you say, the bands are at the same position and have the same pattern in all scans, I would assume that it is a flatfield problem. However, I wonder how this should come about. The scanner should be measuring its current flatfield when turned on :? ? In your two examples the bands are perpendicular to the horizon. If you turn the slide by 90° the bands should become parallel to the horizon if due to a flatfield problem.

If all this is true I would contact Reflecta and ask them if there could be a problem with the flatfield treatment in the ProScan 7200.

Best wishes

Hermann-Josef
DigitDia6000 (CyberView, SilverFast Archive Suite 8 & 9) / CanoScan9950F (ScanGear, VueScan Pro), Eizo CS240, xrite i1studio, Win11 (64bit), Intel i9 (3.4GHz), Speicher 64GB, Nvidia Quadro P2000
Samoreen
Beiträge: 13
Registriert: Montag 27. August 2012, 15:54
Wohnort: Samoreau, Frankreich
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Samoreen »

I know it's hard to believe but I just read somewhere that after a power off / power on of the Proscan, the problem might disappear. So I tried and indeed, the vertical banding was totally gone afterwards.

To be more accurate, I read the comment here : http://www.buzzillions.com/reviews/paci ... er-reviews (look for "banding" in that page.
Patrick
Benutzeravatar
Jossie
Beiträge: 1948
Registriert: Sonntag 10. Februar 2013, 19:40
Wohnort: im Kraichgau
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Jossie »

Hi Patrick,

how long did the scanner work without banding after the power reset? How long had it been turned on before the reset?

Hermann-Josef
DigitDia6000 (CyberView, SilverFast Archive Suite 8 & 9) / CanoScan9950F (ScanGear, VueScan Pro), Eizo CS240, xrite i1studio, Win11 (64bit), Intel i9 (3.4GHz), Speicher 64GB, Nvidia Quadro P2000
Samoreen
Beiträge: 13
Registriert: Montag 27. August 2012, 15:54
Wohnort: Samoreau, Frankreich
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Samoreen »

Jossie hat geschrieben: how long did the scanner work without banding after the power reset? How long had it been turned on before the reset?
Hi,

I had been working with the scanner during about an hour and all scans presented this vertical banding. Then I turned off / turned on without any delay and started to re-scan the same negatives (just a few). I didn't notice the return of the banding after half and hour. Then I had to do something else. I'll try to give more information the next time I have a longer scanning session.
Zuletzt geändert von Samoreen am Samstag 10. August 2013, 17:33, insgesamt 1-mal geändert.
Patrick
Samoreen
Beiträge: 13
Registriert: Montag 27. August 2012, 15:54
Wohnort: Samoreau, Frankreich
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Samoreen »

Samoreen hat geschrieben:Same problem here. I have sent the negatives to a professional lab using a Kodak HR500, so I'll be able to compare the results. This is very annoying but I'm afraid this is a limitation of the ProScan.
By the way, I have received the Kodak HR500 scans. Of course, they don't have the vertical banding issue but otherwise, the quality is not better than what Proscan 7200 can produce. In the contrary, the scans are significantly overexposed and also the advertised scanning resolution (5300 dpi) is greater than the resolution of the PS 7200, I get more details and better color accuracy with my Proscan.
Patrick
Benutzeravatar
Jossie
Beiträge: 1948
Registriert: Sonntag 10. Februar 2013, 19:40
Wohnort: im Kraichgau
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Jossie »

Hi Patrick,

are you sure, that also scans immediately after power-on (when the scanner had been off for a long time) show the banding? This would (almost) kill my theory :o !

Hermann-Josef
DigitDia6000 (CyberView, SilverFast Archive Suite 8 & 9) / CanoScan9950F (ScanGear, VueScan Pro), Eizo CS240, xrite i1studio, Win11 (64bit), Intel i9 (3.4GHz), Speicher 64GB, Nvidia Quadro P2000
Samoreen
Beiträge: 13
Registriert: Montag 27. August 2012, 15:54
Wohnort: Samoreau, Frankreich
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Samoreen »

Jossie hat geschrieben: are you sure, that also the scans immediately after power-on show the banding? This would (almost) kill my theory :o !
I can't say exactly. All the scans, even the very first, had this problem. I'll try to take care of this the next time I scan.

I also made a test with CyberView X5 (which I usually don't use). I scanned a test image almost immediately after turning the scanner on and launching the program. There was a single, dark vertical band on the left. I re-scanned and the band was away. However, the regular vertical band pattern over the whole width of the image was again there (in a very slight form but it was there). I closed CyberView X5, turned off and turned on immediately and re-scanned. Same result : pattern of very slight vertical bands (if you don't know it's there, you don't see it).

It should be noticed that CyberView "calibrates" (?) the scanner when it starts. VueScan doesn't do so. This may explain some differences.
Patrick
Benutzeravatar
Jossie
Beiträge: 1948
Registriert: Sonntag 10. Februar 2013, 19:40
Wohnort: im Kraichgau
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Jossie »

Hi Patrick,

to shed more light onto this issue, let me speculate and describe what I think is going on. Then you may perform tests to verify / falsify this.

As I have already said, this bands look to me like a flatfield problem (I have no other explanation). When the scanner is turned on, it calibrates -- I believe it is measuring its current flatfield. There may be pixels at the X-position of the bands, which are unstable or temperature sensitive, thus changing their sensitivity with time / temperature.

This would have the following implication: A flatfield is not valid long after the scanner is powered up after a longer rest period (temperature below operating temperature). Once it has warmed up, the flatfield changes. Thus if you turn power off and on again at this time, a new flatfield is measured, which is valid for the operating temperature.

Does this theory, please note it is a theory based on what has been described in this thread, reflect what you see?

Hermann-Josef
Zuletzt geändert von Jossie am Sonntag 11. August 2013, 10:14, insgesamt 1-mal geändert.
DigitDia6000 (CyberView, SilverFast Archive Suite 8 & 9) / CanoScan9950F (ScanGear, VueScan Pro), Eizo CS240, xrite i1studio, Win11 (64bit), Intel i9 (3.4GHz), Speicher 64GB, Nvidia Quadro P2000
Samoreen
Beiträge: 13
Registriert: Montag 27. August 2012, 15:54
Wohnort: Samoreau, Frankreich
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Samoreen »

Jossie hat geschrieben:A flatfield is not valid long after the scanner is powered up after a longer rest period (temperature below operating temperature). Once it has warmed up, the flatfield changes. Thus if you turn power off and on again at this time, a new flatfield is measured, which is valid for the operating temperature.
This makes sense. I'll try to adjust my use of the scanner in order to confirm or invalidate this idea.
Patrick
Benutzeravatar
Jossie
Beiträge: 1948
Registriert: Sonntag 10. Februar 2013, 19:40
Wohnort: im Kraichgau
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Jossie »

Good morning,

great, let's see what you find out!

By the way, I would think the calibration procedure is done by the scanner when you turn it on, independent of any software. Also the flatfield correction must be done by the firmware, not the application program. From what I observe, I would assume that SilverFast and CyberView inititate a new calibration when this software is launched.

Hermann-Josef
DigitDia6000 (CyberView, SilverFast Archive Suite 8 & 9) / CanoScan9950F (ScanGear, VueScan Pro), Eizo CS240, xrite i1studio, Win11 (64bit), Intel i9 (3.4GHz), Speicher 64GB, Nvidia Quadro P2000
Samoreen
Beiträge: 13
Registriert: Montag 27. August 2012, 15:54
Wohnort: Samoreau, Frankreich
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Samoreen »

Jossie hat geschrieben: great, let's see what you find out!
Hallo Jossie,

I made some additional tests in order to find a workflow that could avoid this banding problem.

First, I worked during about an hour with a single picture in order to fix some color issues and find satisfactory color settings (I used Silverfast this time). During this time, the banding was there with every scan (I had just powered on the device once).

Then I turned off / turned on and rescanned. Banding gone. I repeated the same scan at increasing time intervals (5, 10, 15, 30 minutes) so that the device be able to cool down more between each scan. After 5, 10 and 15 minutes, the scan was still clean. After 30 minutes, the device did something it didn't do after 5, 10 and 15 minutes : it obviously recalibrated before doing the prescan. The result was a single, strong dark band on the right of the image (see attached screen capture). Re-scanning didn't help. I had to turn off / turn on and the scan was clean again.
Banding1.JPG
Banding1.JPG (14 KiB) 27779 mal betrachtet
So I guess that I should power on the device about 20-30 minutes before starting any actual work and launch dummy scans every 5 minutes until the warm up is really complete. Then I should power off / power on and start the actual scans. I should also scan continuously without pausing the device too long between two consecutive scans.

Awkward but it's a better solution than purchasing a Plustek Opticfilm 120 or a used Nikon Coolscan 5000 ED. It seems that someone else came to the same conclusion and confirmed your thoughts. See http://www.scanhancer.com/index.php?art=2&men=3 .

MfG.
Patrick
Benutzeravatar
Jossie
Beiträge: 1948
Registriert: Sonntag 10. Februar 2013, 19:40
Wohnort: im Kraichgau
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Jossie »

Hi Patrick,

I am glad this helps. Although I would still contact Reflecta, describe what you see and ask for a solution.

The lamp, as suggested in the contribution mentioned above, will not cause the banding. The lamp might change in the global illumination pattern slightly but will never ever produce such a sharp (positive or negative) band exactly vertical (in the scan direction). Since the lamps are LEDs I don't think they need a warmup. I am quite sure it is the CCD with some unstable pixels. After they have reached their operating temperature their sensitivity settles and with power off/on you force the scanner to measure a new flatfield appropriate for the current sensitivity. That should be it.

However, as you already point out, one should not let the scanner be idle during the warmup phase but do dummy scans. Should one have to pause scanning for a longer period, it would be good to repeat this procedure. But some experimenting will be needed to find the appropriate time Intervalls.

Good luck and best wishes

Hermann-Josef
Zuletzt geändert von Jossie am Dienstag 13. August 2013, 16:50, insgesamt 1-mal geändert.
DigitDia6000 (CyberView, SilverFast Archive Suite 8 & 9) / CanoScan9950F (ScanGear, VueScan Pro), Eizo CS240, xrite i1studio, Win11 (64bit), Intel i9 (3.4GHz), Speicher 64GB, Nvidia Quadro P2000
Samoreen
Beiträge: 13
Registriert: Montag 27. August 2012, 15:54
Wohnort: Samoreau, Frankreich
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Samoreen »

Jossie hat geschrieben:The lamp might change in the global illumination pattern slightly but will never ever produce such a sharp (positive or negative) band exactly vertical (in the scan direction).
Agreed. This case is really exceptional. In my case, "usual banding" is much lighter. People not used to examine images with accuracy would not notice it, except in the most severe cases. It's more like that what D.Page posted in his samples above, except that I usually don't see a single band but multiple regularly spaced vertical bands. The intensity of the banding is in most cases lighter than what we see in D.Page's samples.

As to getting in touch with Reflecta, the device is no longer under warranty anyway. As long as I can get satisfactory results with the off/on procedure, I won't undergo the hassle of sending the scanner back to Germany.
Patrick
Samoreen
Beiträge: 13
Registriert: Montag 27. August 2012, 15:54
Wohnort: Samoreau, Frankreich
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Samoreen »

Samoreen hat geschrieben:As to getting in touch with Reflecta, the device is no longer under warranty anyway. As long as I can get satisfactory results with the off/on procedure, I won't undergo the hassle of sending the scanner back to Germany.
The day after writinng this, the scanner started to exhibit strong banding all the time, whatever I did. So I guess it's completely broken now (after 3 years of moderate use - did they say reliable?). I tried a cleanup but this didn't help.

I have found a used Coolscan V for a little more than the price of a new PS 7200. I know it's a risky choice but according to a maintenance center with which I recently got in touch in London, spare parts for the Coolscan V are still available from Nikon. Anyway, this scanner is supposed to produce better scans.
Patrick
Samoreen
Beiträge: 13
Registriert: Montag 27. August 2012, 15:54
Wohnort: Samoreau, Frankreich
Kontaktdaten:

Re: Reflecta ProScan 7200 - Vertical Banding Issue

Beitrag von Samoreen »

I have re-checked all the photos that were scanned with the PS 7200 since the date of purchase. The banding problem started to appear during the last 10 months, a few months after the end of the 2-years warranty period. Planned obsolescence ?

The problem was almost not detectable one year ago. However, darkening uniform surfaces (like skies) beyond acceptable limits makes the problem apparent. Then the banding became more and more apparent until the recent final breakdown.
Patrick
Antworten