Ad
Ad
Ad
Pages: [1]   Bottom of Page
Print
Author Topic: Silverfast HDR Studio 8 - Single Pixel Frame Issue  (Read 2269 times)
TimG
Jr. Member
**
Offline Offline

Posts: 97


« on: January 13, 2013, 03:50:10 PM »
ReplyReply

I'm hoping the resident Silverfast 8 expert (Mark D. Segal) can tell me what I am doing wrong here.

Scanning a 6x7 transparency on an Epson V750 using Silverfast 8 as 64-bit HDRi for processing in HDR Studio 8.  When I process the HDRi file in Silverfast HDR Studio 8, the oddest thing happens.  The resulting file has a single pixel wide "line" which runs the entire height of the image.  It's always on the right side of the image.  Looked at the original and it is not on the original (not a scratch, etc).  If I re-frame the HDRi file in HDR Studio 8, same thing. 

I have checked that I am framing within the film rebate, so that ain't it.  Have also tried re-scanning but no dice.  I cleaned the glass, the holder, and have made sure it isn't a calibration/alignment issue.  Interestingly enough, if I re-scan using Silverfast Ai Studio 6.6 and process in HDR Studio 6.6, I do not have this problem.

Is this a bug or a known issue?  Is HDR Studio 8 simply not ready for prime time?

Thanks!
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6903


WWW
« Reply #1 on: January 13, 2013, 08:51:57 PM »
ReplyReply

I haven't seen this problem so I tried to reproduce it. I don't have any 6*7 transparencies, but I do have some 6*9 colour negatives. First I updated my SilverFast 8 versions to the latest "dot releases" (r19) for both SilverFast 8 Ai Studio for Epson V750 and SilverFast HDR Studio. Using Negafix to get a positive image on display I scanned the negative at 64bit HDRi, opened it in SilverFast HDR, Processed it using Negafix to get a positive image, opened the Processed image in Photoshop CS5, blew it up to 100% and examined all four dimensions closely. I could not replicate your problem. There is no line on any of the four dimensions of the scan I made. I recommend you do a complete software reset for both applications by entering the Service dialog when you first click on the application icon (before you click Start). Check the option to Remove Scanner Preferences and click the Software Reset bar. Try rescanning your film and see whether you get the same outcome. If you do, I can only suggest you contact SilverFast tech support. I'd be interested to hear back whether the reset procedure helps.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
TimG
Jr. Member
**
Offline Offline

Posts: 97


« Reply #2 on: January 14, 2013, 01:16:52 PM »
ReplyReply

Thanks Mark.  I went ahead and did a reset; didn't work.  Next step is to contact Silverfast support.

Will let you know what happens.
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6903


WWW
« Reply #3 on: January 14, 2013, 01:19:14 PM »
ReplyReply

Hi Tim, thanks for informing me and sorry it didn't work. I'll be interested to hear of what happens.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
degrub
Sr. Member
****
Offline Offline

Posts: 274


« Reply #4 on: January 14, 2013, 08:22:33 PM »
ReplyReply

Hi Tim,

Just a couple ideas-
Are you using Digital Ice or SRD on the scan ?
There have been random reports that i can see in the forum about some occasional strange issue with Digital ICE both on 8 and 6 leaving streaks of pixels either dark or white. There were no solutions posted that i saw.

Also, is the problem there if you go ahead and process the file in AiStudio instead of saving for HDR ?

If you used ICE, try SRD and see if the problem goes away. If it does, you may consider cleaning out the Epson drivers and software and re-installing it and/or updating the drivers. The ICE software is part of that package. i suppose there is a possibility this could cause an issue with SF6, so have a good backup  just in case.

It might be an issue between the Epson driver and what SF is expecting or the twain interface is corrupted. Follow the Epson directions for removal exactly if this is a TWAIN interface device.

Frank
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6903


WWW
« Reply #5 on: January 14, 2013, 09:04:09 PM »
ReplyReply

Frank,

The OP is using SIlverFast 8; ICE is not provided in SilverFast 8 for Epson V750, and scanning in HDRi mode, iSRD, the superior SilverFast replacement, is grayed out because one does this in the HDR application using iSRD there. However the infrared channel is scanned at the scanning stage for it to be available in the HDR application. If perchance there is a misregistration between the infra-red channel and the RGB channels, perhaps this could be a cause of the problem, but if the application were doing this systematically and if we are both using the same dot-version (r19) I should have been able to reproduce it and I couldn't, so I am assuming the infrared channel is correctly registered in the SilverFast application. Their tech-support should be able to confirm this one way or another. As far as I know, SilverFast 8 does not use the Epson driver, but Tech Support could confirm that too.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
TimG
Jr. Member
**
Offline Offline

Posts: 97


« Reply #6 on: January 15, 2013, 01:33:14 PM »
ReplyReply

Frank & Mark,

No, I was using neither Digital ICE nor SRD (though I was using iSRD, if that matters).  The word from Lasersoft was it must be a hardware issue.  I called Epson tech support and they said it was a software issue.  Punt, Pass or Kick, I guess?

Anyways, here's what I tried next...which WORKED (!).  I deleted the preferences from Silverfast Ai Studio 6.6 and Silverfast HDR Studio 6.6 and guess what?  When I went to re-scan using Silverfast 8, the mysterious line was no longer there.

Now if somebody would like to tell me how two versions of an application, one being 32-bit and the other being 64-bit, residing in different directories, could somehow corrupt one another, well, I would love to hear the explanation.

Bizarre, but at least I got it solved.

Thanks again guys for helping me troubleshoot.  This forum is orders of magnitude better than Lasersoft's own support forum.
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6903


WWW
« Reply #7 on: January 15, 2013, 01:46:37 PM »
ReplyReply

Tim,

Firstly, congratulations on getting it solved. Who would ever have thought of the one application influencing the other that way. But now that I think of it, not impossible there are some shared components between the two application versions (e.g. preferences or application settings files that are purposely retained and get borrowed) such that resetting both cleaned up everything. I think you should get back to SilverFast and relate this experience to them; ask them whether there are any components in 6.6 that hold-over to 8 which may have caused this. They may learn something from your experience, or have something to contribute about it. Interestingly however, I also still have 6.6 installed on my computer, didn't reset anything there, and I wasn't able to replicate your problem. Well, what to say, all's well that ends well!
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
TimG
Jr. Member
**
Offline Offline

Posts: 97


« Reply #8 on: January 15, 2013, 03:20:07 PM »
ReplyReply

Uh oh.  The problem is back, and this time, I think Frank may be on to something with the suggestion of AACO and/or iSRD being at fault.

Here is the same slide scanned with AACO turned on and off.  Notice even with AACO turned off there is still a thin line on the right side of the scan frame.

Logged
degrub
Sr. Member
****
Offline Offline

Posts: 274


« Reply #9 on: January 15, 2013, 04:20:58 PM »
ReplyReply

Besides the possibility of SF components interacting between the versions, also ask if they are both using the TWAIN interface to talk with the scanner.

Also, run the scan without the infrared channel (just R-G-B) and see if you get the same issue. That should help narrow it down.

If the issue always goes away after clearing the SF 6 preferences, they should be able to find the issue.
Frank
« Last Edit: January 15, 2013, 04:25:16 PM by degrub » Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6903


WWW
« Reply #10 on: January 15, 2013, 04:54:26 PM »
ReplyReply

Ah shucks. That sucks!

I just spent about an hour trying to reproduce your problem based on your latest information, and the outcome is interesting. Firstly, I tried using another image altogether just to make sure the first image wasn't fooling me, but still a 6*9 cm colour negative which converts to positive with Negafix in the HDR preview. I made two 64 bit HDRi scans at 3200ppi. For the first one, I placed the negative in the holder with the top edge flush with the top-end stops in the plastic film frame holder. I blew it up to a 1:1 preview in the iSRD window in the HDR application. I scrolled around the edges looking carefully and found a discoloured band accross the top of the image. Nothing unusual on the sides and bottom however. I turned ACCO and iSRD on and off, but that hardly made a difference. So I thought I was in part but not altogether replicating your problem, but then I wondered about the scan frame. So I made a second scan, but this time with the film moved away from the top edge of the film holder. This time no such problem showed up, so I conclude that the lighting of the image during the scan was reacting against the top scan frame edge in a manner that produced reflectance into the scan, causing the appearance of this streak. The screen grabs show what I mean. This leads me to wonder whether your problem may be somehow related to the scan frame. Otherwise I cannot reproduce it whether AACO or iSRD individually or together are turned on or off and for any of the four edges.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6903


WWW
« Reply #11 on: January 15, 2013, 04:58:57 PM »
ReplyReply

I should add - placing the negative to the top of the frame holder is not a good idea - it cuts off a bit of the image edge.

Next, I suggest you try making a scan using the glass tray instead of the plastic frame holder. Don't worry about affixing the film to it. See if you get a different kind of outcome.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
TimG
Jr. Member
**
Offline Offline

Posts: 97


« Reply #12 on: January 29, 2013, 03:31:35 PM »
ReplyReply

Hi Mark,

I thought it might be useful to provide an update on my issue.  It had something to do with both AACO and ME being used on the scan frame.

Thanks again for your help!

Tim
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6903


WWW
« Reply #13 on: January 29, 2013, 03:36:18 PM »
ReplyReply

You are welcome.

Great that you seem to have gotten to the bottom of it. Now, did you find this out on your own, or through tech support? I ask, because if through tech support that would make it a "known issue" that could also affect others. Or if you have hit on something generic, it could also affect others, in which case I should experiment as soon as time permits and see what happens here.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
TimG
Jr. Member
**
Offline Offline

Posts: 97


« Reply #14 on: January 29, 2013, 05:00:32 PM »
ReplyReply

Unfortunately, I figured it out the way I always do - trial and error and conducting a series of tests using different settings combinations.  The response I got from Lasersoft was pretty funny.  They wanted me to send them full res files but their email system couldn't accept files larger than 5MB.  I could dropbox it to them, that was about all I could do.  They couldn't reproduce it, even after I spelled out the steps.  I should say, though, this was a real head scratcher and I wouldn't expect most users to encounter this issue.

In my experiments, I thought it might be registration issues caused by film popping due to excessive heat from the scanner ccd.  So I went into town and asked the local frame shop to cut two strips of conservation clear glass to fit in the film channels of the Epson medium format film holder.  I removed the lids to each channel and placed the film curled/bowed side up, then placed the glass on top.  What do you know?  $12 of glass keeps the film flat, with a visible increase in sharpness, especially in the corners.
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6903


WWW
« Reply #15 on: January 29, 2013, 06:45:30 PM »
ReplyReply

I don't see anything "unfortunate". What you did is fine - sometimes trial and error is the ONLY way we find out certain things, especially if we are not scanner engineers or software developers. The fact LSI couldn't reproduce it is interesting. This means when time permits I should do a test or two. Flattening the media with conservation clear glass would improve overall sharpness, but does carry a risk of producing rings.The other approach would be to use the flat glass frame and film cleaner to hold the film to the glass, but also risks showing rings unless it is well-bound. Not easy. :-)
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
TimG
Jr. Member
**
Offline Offline

Posts: 97


« Reply #16 on: January 30, 2013, 10:22:28 AM »
ReplyReply

As long as the glass is kept clean, newton rings shouldn't be an issue (should they?). I've now done some 50 scans using the glass setup and have not seen rings in any of the scans.  But I will keep an eye out for it. 

If I were to sandwich the film between the conservation glass and the scanner bed, wouldn't I lose the ability to do HDRi, since the function (as was explained to me a year or so ago by LSI) requires the film be in on of the dedicated film holders?
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6903


WWW
« Reply #17 on: January 30, 2013, 11:14:26 AM »
ReplyReply

Ouch - possibly. The issue could be the alignment of the infrared channel. Best thing - do a test. Use the glass the way you would like in HDRi mode, then in the HDR application activate iSRD and see whether the corrections of the spots end-up where the spots are. If they don't, you have an issue, if they do, you're good to go with that technique.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Pages: [1]   Top of Page
Print
Jump to:  

Ad
Ad
Ad