(Lack of) Frame Definition in Negatives - How to Scan?

[EthanFrank]

I'm using an Epson V500 with the included software.

I shot some Tri-x pushed to 1600 last month, and I'm having a hard time scanning it. The shots were in the dead of night, and the edges of your average frame are completely black - that is to say, there is no definition between frames.

This is causing the scanner to misread the frame edges, and scans my images in a very strange way, i.e: one 'frame' is really a combination of two.

Does anyone know of a solution?

 

[Keith]

You need to use the scanner in professional mode and create your scan frames manually. In auto mode the scanner is tryng to read the contrast at the edge of the negative to create the frames automatically ... if it can't detect sufficient contrast what you are describing will occur.

It's a little labour intensive but far superior in the long term no matter what the negatives are like. I used my V700 in auto mode once or twice before giving up and taking the manual route.

 

[EthanFrank]

Keith,

My first thought was to try and use professional mode - however, it's still autodetecting frame edges! I may have changed a setting a while back...any idea about how to remedy that?

Edit: Got it! I had the 'thumbnail' tick box...ticked.

 

[Keith]

I'm not that familiar with the V500 but maybe it's different. With my V700 after deleting any existing frames I do a preview scan which scans the entire bed area. I then manually draw a 36mm x 24mmm frame for the bottom left negative then duplicate it and move it to the next frame then repeat the process and so on until complete. I then go back and highlight each frame and hit the zoom feature which enlarges that particular frame to full screen and then move it around until I have it where I want it. I don't recall any setting in professional mode that allows the V700 to set the frame automatically?

Maybe someone else can chime in here?

 

[Paolo Bonello]

Yeah that feature that selects the 24 frames of the tray Automatically is OK but for night shots it's a pig.
I was just recently told to try scanning as a positive with Auto frame select then invert in post because it might find it easier to find the thumbnails as positives however I have not tried it and can't really understand how it would work better because it would still be looking for the contrast change between the framelines and the image which would still not be apparent. I had the same problems with dark shots and found Silverfast better for batch scanning night shots but then that too had it's own set of problems with the program having glitches and scanning stopping during a batch scan.
Keith I must try your method on epson scan. I could not work out how to select more than one frame in manual area selection mode. Must try it as you described.

 

[EthanFrank]

Paulo,

I'm trying Keith's method right now, and while it's certainly slow, it works!

The only thing I'm afraid of, Keith, is adjusting my target document size. Original, I assume, will create 35x24mm files. How do I ask for a larger size without it changing the crop lines I just so painstakingly created?

 

[Keith]

I go nowhere near that target size setting. Once I've created my first window to the exact dimensions I want manually there doesn't seem any need and it duplicates to the same dimensions each time. When I go to scan MF negs I loose my setting of course and have to re-create them to suit and the same when returning to 135mm but that doesn't bother me because I'm obsessed with accuracy.

Starting from scratch I can create twenty four separate frames then go back and zoom and fine tune them in around five minutes so it's no big deal IMO. Mind you I've had infinite practice at this! 😀

 

[Paolo Bonello]

Ethanfrank I think I understand your issue but as Keith said it's probably not necessary to worry about target size. My output size is always same as original size. I scan at 2400dpi. It saves me doing the math of trying to calculate the reduction in required DPI as you increase the output size. The only time I would bother changing output size is if I were printing straight from the scanned file. Most people don't do that but choose to run it through photoshop first or some other photo editor to add corrections to the image and to output the file as required for printing or web.
Disclaimer: I'm no pro at this so please don't take what I say as gospel. I'm sure there's others here with far more scanning and photoshop experience than I.

 

[kokoshawnuff]

I had the same problem with my v500 and the same problem would arise with my v700 (though now i use the film 'area guide' and no frame holder), Keith's answer is really the only possible one with the V500, as the flatbeds often has problems determining very light or dark frame boarders as a positive or a negative.

 

[haempe]

Similar experiences here (with V700).
Sometimes, if the complete roll are nightshots, I use the manual mode and scan the whole 6 frames strip in one and crop it later in 6 pieces in PS.

 

[shimokita]

OT: I had a roll break in camera (don't ask) and the film was split down the middle. I cut the strip into individual images and mounted it (two pieces per image) in a transparency slide frame and scanned it that way... a little work in post and good as new. Obviously you don't want to cut your film, but just to say...