Silverfast users?

[Alan Edward Klein]

Alan, I'm not sure which Epson you're talking about, but their V8xxx models have Dmax 4.0 without having to slow down, so why bother?

I agree with you. The V850 is the model SF will do the two scans on. So that makes the claim Silverfast can improve its dMax even less credible and just seems like a selling point. I've never seen anyone post a photo using the two scans approach that you couldn't increase shadow details using the shadow slider in a post scan editing program of a single scan.

 

[Les Sarile]

How about a three scan approach.

DMAX, but not specific to Silverfast and applies to general scanning tools. In this case I am using the Nikon Coolscan+Nikonscan and post process to demonstrate getting the max detail out of a scan.

I was curious if I could gain shadow detail without blowing out highlights. In this first example, I am scanning Velvia 100 with an Analog Gain of 0, +1 and +2.

Larger version -> http://www.fototime.com/893E0E37B4C4087/orig.jpg

As you can see, default exposure of '0' looks too dark, +1 is more appropriate and at +2 the horse looses texture.

I use HDR to merge the three exposure settings as one image. Floor texture from the shadows and horse texture is achieved and it has that HDR look/coloring

I also used ACDSEE shadows tool (or PS shadow tool) to bring out the shadow detail on the analog '0' and bring out some shadow detail without the HDR look. So in an otherwise deep shadow, you can bring out shadow detail achieved in the scan.

This next one is from Kodak Portra 400 which seems to have endless highlight range.

Full res -> http://www.fototime.com/DCE615918D77901/orig.jpg

Again, the single scan has achieved the full DMAX, but in order to show the shadow and highlight details, you will need to use post to bring those out if those are important.

I didn't really need the three scans . . . unless you want that HDR look.

 

[Alan Edward Klein]

Les, You proved multiple scans are overrated. Thanks.

 

[Ko.Fe.]

@Ko.Fe. I'm still trying to decide which scanning software to use. In what ways do you find that SF is better than VS?

A lot better on result, but some time is needed to find good settings for each roll.
If you are up to MF scanning, nothing is close to Epson with its software for simplicity and good results on all is in auto.

 

[138S]

How can multi exposure improve the dMax of the scanner?

Alan, it improves DMax by extending dynamic range in the same way HDR photography of a DSLR does it.

With M-E the scanner takes two scans of the same image, each at different exposure, later both images are combined.

the longer exposure image (probably) has the highlights blown but the shadows have better detail, so this image provides the detail for the shadows.

Multi-Exposure solves shorcomming in the electronics capability, when making the Analog to Digital conversion (this is taking the voltage accumulated in each pixel and converting it to a digital value) it is difficult to have top precission and top speed at the same time. Multi-Exposure solves that, allowing to get an optimal result close to what is possible in that scanner.

Probably some Pro scanners can take both shots for a row (the short and the long exposure) before advancing the carriage, in the Epson case it is necessary to make to passes to get two scans which is not as nice, but still M-E provides a powerful enhacement for deep Velvia shadows, DMax is increased and noise in the very dense areas is decreased impressively, many times avoiding to need for a drum scan or flextight.

The Epson V850 preforms impressively well with very dense Velvia, at least in (say) 95% of the times we don't need more, but there can be situations in what we may require a drum job to recover better ultra dark slides, mostly coming from a quite flawed exposure in the taking.

 

[Alan Edward Klein]

Alan, it improves DMax by extending dynamic range in the same way HDR photography of a DSLR does it.

With M-E the scanner takes two scans of the same image, each at different exposure, later both images are combined.

the longer exposure image (probably) has the highlights blown but the shadows have better detail, so this image provides the detail for the shadows.

Multi-Exposure solves shorcomming in the electronics capability, when making the Analog to Digital conversion (this is taking the voltage accumulated in each pixel and converting it to a digital value) it is difficult to have top precission and top speed at the same time. Multi-Exposure solves that, allowing to get an optimal result close to what is possible in that scanner.

Probably some Pro scanners can take both shots for a row (the short and the long exposure) before advancing the carriage, in the Epson case it is necessary to make to passes to get two scans which is not as nice, but still M-E provides a powerful enhacement for deep Velvia shadows, DMax is increased and noise in the very dense areas is decreased impressively, many times avoiding to need for a drum scan or flextight.

The Epson V850 preforms impressively well with very dense Velvia, at least in (say) 95% of the times we don't need more, but there can be situations in what we may require a drum job to recover better ultra dark slides, mostly coming from a quite flawed exposure in the taking.

Silverfast: What's the difference in the two scan times? Are they adjustable? Does SF combine the two scans into one resultant file or does it provide two separate files?

Epsonscan: Is there a way to adjust the scan times using Epsonscan?

 

[138S]

Epsonscan: Is there a way to adjust the scan times using Epsonscan?

Yes, you go to Proffesional mode, Configuration and Color, there you have the AutoExposure Level slider. It provides and indirect control on exposure, but no direct control.

See it here: https://www.alexburkephoto.com/blog/2013/06/02/scanning-and-editing-color-negative-film

Silverfast: What's the difference in the two scan times? Are they adjustable?

Until I know you cannot control the multi-expoure adjustment, you activate it or you disable it. It has an optimized hardcoded behaviour. I don't know what difference there is between the two scan times.

Does SF combine the two scans into one resultant file or does it provide two separate files?

You get a single regular file, but with better effective dynamic range with way lower noise in the shadows. But save it in TIFF 16bits/channel, if not much of the benefit from multi-exposure can be lost.

 

[Alan Edward Klein]

Yes, you go to Proffesional mode, Configuration and Color, there you have the AutoExposure Level slider. It provides and indirect control on exposure, but no direct control.

See it here: https://www.alexburkephoto.com/blog/2013/06/02/scanning-and-editing-color-negative-film




Until I know you cannot control the multi-expoure adjustment, you activate it or you disable it. It has an optimized hardcoded behaviour. I don't know what difference there is between the two scan times.




You get a single regular file, but with better effective dynamic range with way lower noise in the shadows. But save it in TIFF 16bits/channel, if not much of the benefit from multi-exposure can be lost.

Something doesn;t make sense. So by running the scan slower, the scanner performs better than the manufacturer's specs? That means for the V850, Epsons spec'd dMax of 4.0, or whatever it really is, is improved? Why didn't Epson just run the scan slower to begin with? It just doesn't seem likely that SF improved Epson's own design. It seems that slower speed would just distort the signal one way or the other. Or combining two scan like an HDR without any fine controls is just going to get you one mess of a picture.

Also, if SF can slow down the scan speed, why not just do one scan at slower speed? As long as you don't clip the highlights, you'll get a cleaner scan because you won't have to combine two scans. IS there a way of trying this with SF or creating separate files of the two scan method?

Frankly, I still think it's a lot of hype from SF. I've never seen a comparison posted in any forum that showed any difference or improvement that couldn;t be gotten using a regular shadow slider to bring out some more of the details in the shadows. The Epson scanner captures more than is apparent in the shadows than is at first obvious.

 

[138S]

Something doesn;t make sense. So by running the scan slower, the scanner performs better than the manufacturer's specs?

Alan, it makes sense. Imagine shot a very high contrast scene with a DSLR. You have highlights and deep shadows but the DSLR cannot take all well: if you expose to have good detail in the shadows then highlights are blowed, if you use a shorter exposure to conserve highlight texture then deep shadows are lost... but the HDR function of the DSLR takes two or three shots and combines both images in a single one. That image has texture in the highlights and detail in the shadows. The scanner does exactly the same.

The benefit does not comes from the scanner running slower, it comes from the second scan made with longer exposures for each row.

That means for the V850, Epsons spec'd DMax of 4.0, or whatever it really is, is improved?

the 4.0 DMax rating of the Epson is inflated as in many other scanners. First measuring DMax in scanner is complex, imagine you scan a single 35mm strip, it is not the same if you have the other receptacles dor strip covered with an opaque foil or if those windows are openned and generating flare from illumination.

Take a really dense Velvia and try Multi-Exposure, you will notice an impressive difference.

Why didn't Epson just run the scan slower to begin with?

It is not about running slower or faster, this is about having two photographs for each row, each made with different exposure times.

Frankly, I still think it's a lot of hype from SF.

Personally I use more Epson Scan than SIlverfast, I only use Silverfast for color negative film because Negafix allows to selecs the film type, and for the conversion an optimized color map is used for each film type, and als I use SF when I need to recover ultras deep shadows in slides.

For BW I don't much the advantage of SF, and I prefer Epson Scan for that.

 

[Bormental]

and for the conversion an optimized color map is used for each film type

BTW I'd love to understand how this feature works and why it's needed. I've seen the color profiles of negative stocks used for simulations, i.e. you take a photo with your digital camera and apply a preset to make it look like it was shot on film. The name "color map" makes sense, as most cameras distort color on purpose to make it more appealing, so a preset is needed to map from camera's own color profile to the film's.

Why does a good scanner need to know which film stock it's scanning? It should not need a "map", what is it mapping from and to? A scanner shouldn't have its own color "signature" and the original negative is right there, so just acquire and digitize the color as truthfully as possible and transfer it into a computer. I know next to nothing about wet printing from color negs, but an RA4 paper doesn't care which emulsion you're printing from. Why would a scanner care?

One possible explanation is that scanners have shitty sensors that introduce all kinds of color shifts, i.e. the hardware is basically incapable of scanning properly. Quite sad if true.

 

[138S]

BTW I'd love to understand how this feature works and why it's needed. I've seen the color profiles of negative stocks used for simulations, i.e.

There is very complex math about that...

For decades negative color film and color RA-4 photopapers evolved together, and the way RA-4 photopaper interpretes each negative type is particular. The scanner CCD has its own spectral sensitivity that's not the RA-4 one and it has to work well for slides and for color negative film at the same time, so the single possible way is to provide optimized color maps for each negative type.

Note that color inversions from DSLR scans are challenging, and DSLR sensors are not "shitty".

 

[MattKing]

Why does a good scanner need to know which film stock it's scanning? It should not need a "map", what is it mapping from and to? A scanner shouldn't have its own color "signature" and the original negative is right there, so just acquire and digitize the color as truthfully as possible and transfer it into a computer. I know next to nothing about wet printing from color negs, but an RA4 paper doesn't care which emulsion you're printing from. Why would a scanner care?

Different films have different characteristics respecting contrast and basic hue (starting filter settings).
In addition, I don't know that there are any scanners that use a light source that matches the tungsten or halogen light sources that colour negative films and RA4 papers are designed around.
I expect that all scanning systems - both dedicated scanners and camera based digitization would work better if the light sources used were the same continuous spectrum sources used in colour printers and enlargers.

 

[Bormental]

@138S Thanks, that makes perfect sense. In this case though, why not have a single "C41 profile" and apply it to sensor data? Interestingly, @Adrian Bacon also went similar route with his Simple Image Tools, IIRC he ended up with "profiles" for different emulsions, hopefully he'll be able to contribute.

 

[PhilBurton]

Different films have different characteristics respecting contrast and basic hue (starting filter settings).
In addition, I don't know that there are any scanners that use a light source that matches the tungsten or halogen light sources that colour negative films and RA4 papers are designed around.
I expect that all scanning systems - both dedicated scanners and camera based digitization would work better if the light sources used were the same continuous spectrum sources used in colour printers and enlargers.

Or if the electronics of the scanner uses a continuous spectrum light source and the electronics compensate for the difference from say a tungsten or cold light source used for printing.

 

[138S]

@138S Thanks, that makes perfect sense. In this case though, why not have a single "C41 profile"

It is not possible... because the color dyes in each film reacts different to the illumination light in the scanner and to the R-G-B color filters that pixels have over them.

 

[Bormental]

It is not possible... because the color dyes in each film reacts different to the illumination light in the scanner and to the R-G-B color filters that pixels have over them.

... and to make it possible one needs to construct a film scanner with a specialized light source (similar to a C41 enlarger?) which makes the color dyes react similarly? I'm quite certain we don't need a special "C41 sensor" in our hypothetical perfect scanner, but the negative+light source combination is puzzling.

The reason I am bringing this up is that I actually want to to build a film scanner that requires no software, i.e. you plug it into a USB, it shows up as a new drive, with DNG files for each slot on the film tray. Basically, an electronic version of RA-4 process which "prints" into DNG files.

[EDIT] The reason I ordered Plustek 120 Pro is to have some kind of baseline performance, to add to my Epson V600 and my two DSLRs that I scan negatives with.

 

[MattKing]

... and to make it possible one needs to construct a film scanner with a specialized light source (similar to a C41 enlarger?) which makes the color dyes react similarly? I'm quite certain we don't need a special "C41 sensor" in our hypothetical perfect scanner, but the negative+light source combination is puzzling.

You need a continuous spectrum light source - preferably with the same colour temperature as the light sources used in optical enlargers and printers.
You need some means of adjusting the colour of that light source in a way that mimics the effect of the colour printing filters used in optical enlargers and printers. It is those colour printing filters that allow adjustments for different negative film stocks. They also allow adjustment for different color temperatures of ambient light at the scene, but that can be done in post processing as well.
You don't need a special C-41 sensor, but you need some combination of sensor hardware, firmware and software to take the non-linear, inverted colours from the masked C-41 negative - which are designed to be printed on the non-linear RA-4 colour paper - and to invert them and render them as natural looking positive colours in the manner that RA-4 paper does.
The package of sensor hardware, firmware and software will need to be modified extensively if you want to use it for scanning transparency film. You may also want to modify the light source, but you can probably do that with filtration if it is continuous spectrum.

 

[PhilBurton]

You need a continuous spectrum light source - preferably with the same colour temperature as the light sources used in optical enlargers and printers.
You need some means of adjusting the colour of that light source in a way that mimics the effect of the colour printing filters used in optical enlargers and printers. It is those colour printing filters that allow adjustments for different negative film stocks. They also allow adjustment for different color temperatures of ambient light at the scene, but that can be done in post processing as well.
You don't need a special C-41 sensor, but you need some combination of sensor hardware, firmware and software to take the non-linear, inverted colours from the masked C-41 negative - which are designed to be printed on the non-linear RA-4 colour paper - and to invert them and render them as natural looking positive colours in the manner that RA-4 paper does.
The package of sensor hardware, firmware and software will need to be modified extensively if you want to use it for scanning transparency film. You may also want to modify the light source, but you can probably do that with filtration if it is continuous spectrum.

+1 to everything @MattKing said.

A proper scanner + good scanning software should address all these issues. If you are doing a DSLR scanning setup, then it's, "Good luck, Charlie" getting everything right. I also read recently (or maybe it was a YouTube video) that it's harder to control dust with a DSLR scanning setup than with a proper scanner.

 

[Bormental]

A proper scanner + good scanning software should address all these issues.

Yes. But unfortunately nobody built a proper scanner yet. And, by the way, a proper scanner doesn't need any software.

 

[Les Sarile]

Yes. But unfortunately nobody built a proper scanner yet. And, by the way, a proper scanner doesn't need any software.

Tens of thousands of scans later and I beg to differ - Nikon Coolscan V, 5000 & 9000 + Nikonscan. But hopefully the Opticfilm - or other, proves a viable replacement . . .

 

[Bormental]

@Les Sarile Hehe, I owned one many years ago, IIRC it was 5000 ED. Using it was like pulling teeth. I'll see how this OpticFilm 120 Pro fares, and if it's anything resembling my Coolscan experience, I'll build a proper scanner for myself. @PhilBurton is right, DSLR is not it.

 

[Les Sarile]

@Les Sarile Hehe, I owned one many years ago, IIRC it was 5000 ED. Using it was like pulling teeth. I'll see how this OpticFilm 120 Pro fares, and if it's anything resembling my Coolscan experience, I'll build a proper scanner for myself. @PhilBurton is right, DSLR is not it.

The majority of my scans came from the 5000+Nikonscan combination which I can only describe as uneventful. What about it was like pulling teeth?

 

[Bormental]

Speed. Digitizing a roll of film shouldn't take longer than a minute. 2-3 seconds per frame sounds about right. For B&W my existing DSLR-based setup is already miles ahead of that Nikon. I can do about 4 rolls per hour. Scanning must take less time than developing. Call it the Bormental Rule

 

[Les Sarile]

To answer your question: I absolutely love my DSLR workflow for 35mm B&W. But medium format is another matter: I shoot color in MF. Stitching is annoying, and I only get 6000x6000 image out of it (I know Portra 160 is capable of more). Also, color inversion and orange mask removal for color negatives is very time consuming. I have not figured out how to automate color inversion, for some reason every shot requires separate level adjustment for each channel.

Speed. Digitizing a roll of film shouldn't take longer than a minute. 2-3 seconds per frame sounds about right. For B&W my existing DSLR-based setup is already miles ahead of that Nikon. I can do about 4 rolls per hour. Scanning must take less time than developing. Call it the Bormental Rule

Methinks there may be disappointment in your scanning future as for one not only is the Opticfilm already slower then the Coolscan, ICE is not as good and maybe hundreds of color negative scans later before we really know just how good it is.

Have you heard about the Nikon D850's built-in color negative conversion? That maybe a solution to your color negative DSLR scanning problem although I have yet to try it myself. BTW, what DSLR do you use?

As far as speed and quality on a scanner, back many many years ago, they used to hold a camera convention in Las Vegas and I got to try a Durst scanner system which gave similar output as a Coolscan at about 5 seconds per frame. I suppose it was an optimized motorized feed DSLR scanning system. I say it provided Coolscan results because at that show I took a strip of color negatives with me and I went around the different booths to have them scan a frame for me and the Durst and Coolscan results looked identical in resolution, color/contrast but of course no ICE. I remember it was about $35K then so . . .

Of course with the 5000+whole roll feeder, you can just feed the whole uncut roll of film and the Coolscan+Nikonscan will batch scan it all.

Also, I understand that Noritsu scanners are now getting to be more available at reasonable prices, I believe depending on the resolution setting, it can pump out scans at a very high rate.

 

[MattKing]

There is a reason that the scanners used by the motion picture industry cost such huge amounts of money.
As did the video assisted colour printers that the top line pro labs used.
I expect you could design a large sensor single shot scanner to replace the Coolscan 9000.
I would expect it to cost a lot more than a Nikon D850.