Ektachrome ........ needing to increase 1.7 stops in Lightroom .... normal?

[Dazzer123]

Hi folks,

I've recently started to shoot 4x5 Ektachrome.

Although the slides look pretty ok when held up to the light, after i scan them, i'm needing to add about 1.7 stops of exposure in Lightroom to get the exposure i want. And also contrast.

I also need to adjust a quite extreme color cast.

Is this normal(ish)?

The end pic is acceptable, but i'm curious if a better result is possible without such extreme LR tweaks?

I'm scanning using a v850 Pro, Silverfast 9, flat settings (no tweaks) in Silverfast, scanning from the glass.

So when i scan in Silverfast, the scan looks pretty dark. Below is the pre & post Lightroom version.
(this is a test shot, it's not supposed to be award-winning photography!)

I'm using Cinestill CS6 "CREATIVE SLIDE" DYNAMICCHROME KIT (E-6) to develop. I read a lot of negative things about this process on the Interwebs, perhaps herein lies the problem?

Thanks in advance!

 

[koraks]

What does the slide look like in real life? If you hold it to the light (e.g. overcast sky), does it look right to you?

Scanning is a can of worms.

Also, it's the same issue you've brought up here: https://www.photrio.com/forum/threads/e6-lab-versus-cinestill-kit.203056/unread
And you've been asking about metering a lot lately. These are all sensible issues you bring up, and it may be time to start excluding some variables. I think you may get stuck if you try to interpret the same slide in terms of metering, processing and scanning. You'll never find the cause that way; you'll have to play with parameters.

Learn to meter and judge slides by shooting a 35mm or 120 roll of slide film and get a reputable lab to process it for you. This will tell you what a slide is supposed to look like, and how metering works. You can then scan that film to figure out how your scanner interprets the film with your preferred scanning software and settings.

 

[Dazzer123]

When held up to a very bright light (cloud with sun behind), it looks pretty good.

On my Kaiser Slimlite Plano light tablet, it looks pretty underwhelming .......

 

[koraks]

When viewing a slide, the black margins should be pitch black. If they allow light to pass, your light source is too bright. If you adjust the lighting conditions to that the rebate around the edges is absolutely dark and the image then looks underexposed, there's something wrong with the slide. It's either underexposed, or underdeveloped. Given your recent questions about metering, I would start by working on the exposure part first. The suggestion to have a roll of film processed by a lab will eliminate that aspect of uncertainty, assuming the lab does a reasonable job.

On my Kaiser Slimlite Plano light tablet, it looks pretty underwhelming .......

Yeah, that doesn't sound good. Generally, slides look pretty magnificent when viewed on a light table of any kind.

 

[Dazzer123]

OK!

So you're saying that with a well exposed & developed slide, after scanning you should have (more or less) the final exposure and color balance that you want without any added digital tweaking?

 

[Alan Edward Klein]

Scans with a V850 will be dark if scanning "flat" with Epsonscan with no scan exposure adjustments. (I use Epsonscan when I scan on my Epson V600 and V890. I can't advise on Silverfast as I never used it). If you want to get a more normal exposed scan, set the black and white points to just beyond the edges of the data of the photo shown in the histogram. That's what I do. Then I tweak the rest of the exposure in Lightroom.

Of course, if you set the scan on auto exposure, then the scan program will compensate and brighten the picture automatically. I stopped using this auto mode because it tends to clip highlights. Follow the first paragraph.

Frankly, you could scan flat and do all the adjustments in Lightroom on the darker picture scan and get basically the same results.

 

[miha]

I would guess the Cinestill CS6 "CREATIVE SLIDE" DYNAMICCHROME KIT (E-6) is the culprit. Ektachrome needs proper E6 processing.

 

[Dazzer123]

Scans with a V850 will be dark if scanning "flat" with Epsonscan .......etc

Thanks, i need to try the Epsonscan software, Silverfast is a little idiosyncratic. Although scanning with "multi-exposure" on is quite a dramatic difference, curious if Epsonscan has a similar feature.

re: flat scanning: So that is actually what the pics are in the OP. The first pic is flat outta Silverfast, the second is that same pic tweaked in Lightroom.

The end result is quite acceptable, it's just for my own learning that i want to know how good a slide can actually be before scanning.

 

[Dazzer123]

I would guess the Cinestill CS6 "CREATIVE SLIDE" DYNAMICCHROME KIT (E-6) is the culprit. Ektachrome needs proper E6 processing.

Right, i'm going to finish off the Cinestill chemicals that i have left over, and then move over to Fuji Hunt.

I think the results with Cinestill are ok, especially for the modern user that'll probably photoshop the heck out the pic anyway, but i'm hoping proper E6 processing will give me a more vibrant and contrasty image straight out of the box.

 

[koraks]

OK!

So you're saying that with a well exposed & developed slide, after scanning you should have (more or less) the final exposure and color balance that you want without any added digital tweaking?

Not exactly; what I said is that a well-exposed and well-processed slide will look great when viewed with the human eye on a light box. How it scans, really depends on the scanner and the software. However, having said that, my experience is that slides generally scan pretty easily as long as they're properly exposed and processed. There are some caveats to this, but overall, scanning slides is fairly straightforward.

 

[Dazzer123]

EDIT: never mind, will carry on with my experimentation, getting closer each time!

 

[Alan Edward Klein]

Thanks, i need to try the Epsonscan software, Silverfast is a little idiosyncratic. Although scanning with "multi-exposure" on is quite a dramatic difference, curious if Epsonscan has a similar feature.

re: flat scanning: So that is actually what the pics are in the OP. The first pic is flat outta Silverfast, the second is that same pic tweaked in Lightroom.

The end result is quite acceptable, it's just for my own learning that i want to know how good a slide can actually be before scanning.

I always look at the slides on a light panel to see if they're exposed correctly. Then scan. You can't get blood from a turnip.

As far as I know, Epsonscan does not do multiscan or muti exposure. I've investigated this before and never found anyone proving this actually works as claimed with Silverfast. Sure, you can get two scans. But it doesn't help, just wastes time.

The scanner is designed for maximum dMax I believe it's 4.0 for the V850. So you're already getting max in the shadow areas. Two scans don't make a difference, at least I've never seen anyone provide proof, only claims that Silverfast does this and makes a difference. All you do is add noise and artifacts with two scans. Just raise the shadow slider in Lightroom a little with one scan, but don't overdo it. Maybe some curves. Good luck.

 

[runswithsizzers]

I would never use the results of a scan to judge the adequacy of the film exposure / development unless it was a film+camera+developer+ scanner+software combination which I had a LOT of experience with.

Back when I was scanning slides using a film scanner + Vuescan, underexposed slides were a problem. I could see details in the shadows when my dense slides were projected, but my scanner was not really capturing them.

For me, the answer was to switch to using a digital camera + macro lens to copy my slides. That gave me a lot more control over how much exposure the slide got at the time it was digitized. Watching the camera's histogram allowed me to "scan to the right" -- that is, I was able to give the slide as much exposure as possible without clipping any highlights. When processing slide scans from the old film scanner, I almost always needed to boost the exposure in Lightroom. Now, when processing RAW files from my digital camera, I usually need to reduce exposure in Lightroom by 1/2-1-stop.

It looks to me like your setup is capturing adequate shadow detail. In fact, your results look pretty good to me. I believe it is fairly common for any and all scans to need a certain amount of cleaning up in something like Lightroom.

OK, let's put it another way, should a well-exposed and well-processed slide look roughly as bright on a light box as on a computer monitor (all other things being equal)?

But all other things are never equal, so it's impossible to generalize. I have not used a Kaiser Slimlight, but I had a very similar looking light table. When copying b&w negative film on that light table my shutter speeds were typically about 1/4th - 1/10th second. After I switched to a different brand light table, I needed shutter speeds of around 1/125th - 1/250th sec. So light tables vary in brightness. And I'm sure people have their computer monitors set to all kinds of different brightness levels, too.

 

[Ivo Stunga]

I put my money on their CS6. Have put 2 E-6 films through it, both came out dark.

 

[Dazzer123]

Alright!

Here's the difference i'm seeing with multi-scan (unsharpened crop of the full scan).

Not sure if there is actually more information being retrieved with multiscan, but even without the crop / pixel peeping, i have the feeling the multi is a little nicer!

 

[runswithsizzers]

When using Vuescan's multiscan option with my old Minolta film scanner, I saw a similar improvement in noise reduction in the darkest areas of the slides. Vuescan let me set the number of samples, and of course the more times it samples, the longer it takes. For me, 2x was worth it, and probably 3x if I was trying to recover shadow detail from an underexposed slide. But any improvements from 4x were questionable, at best, so not worth the extra time.

 

[runswithsizzers]

The scanner is designed for maximum dMax I believe it's 4.0 for the V850. So you're already getting max in the shadow areas. Two scans don't make a difference, at least I've never seen anyone provide proof, only claims that Silverfast does this and makes a difference. All you do is add noise and artifacts with two scans. Just raise the shadow slider in Lightroom a little with one scan, but don't overdo it. Maybe some curves. Good luck.

My experience using Vuescan with a Minolta film scanner is different.

I believe the problem with noise in near-black areas is because the sensor/software is having trouble differentiating between black and some very small amount of light. By taking multiple samples and averaging them, random variations are minimized and noise is reduced. It is not a huge, OMG!, knocks-your-socks-off level of improvement, but I can definitely see slightly smoother tones in very dark areas of my slides when using Vuescan's multiscan. And I think @Dazzer123's examples in post #15 demonstrate this, as well.

If there is a lot of noise in the near-black areas, then raising the shadows in post processing tends to make the noise look worse. So in my experience, files which were multisampled can tolerate a bit more shadow enhancement than those which were not. YMMV

 

[bfilm]

One needs to distinguish between multi-exposure and multi-scanning (aka multi-sampling, if it is done in one pass). They are very different.

Multi-exposure takes two scans with different exposure times.

Multi-scanning takes two or more scans (either in one pass ('number of samples' in VueScan) or multiple passes ('number of passes' in VueScan) depending upon the scanner) to try and cancel out noise, but using the same exposure. Also known as single-pass multi-scanning and multi-pass multi-scanning.

In regard to viewing transparencies, the modern LED transparency viewers are very inconsistent. The better choice is to use the high-grade transparency viewers using standardized 5000K fluorescent lamps from GTI and JUST Normlicht. These are not inexpensive, but they are probably the best.

Profiling the scanner with an appropriate IT8 target for Ektachrome film, like those from Wolf Faust, might also help get accurate scans.

And you might consider using an output or woking space profile like Ekta Space PS 5 or DonRGB4, both of which were designed specifically for scanning film and covering the colors of color reversal film.

 

[Alan Edward Klein]

My experience using Vuescan with a Minolta film scanner is different.

I believe the problem with noise in near-black areas is because the sensor/software is having trouble differentiating between black and some very small amount of light. By taking multiple samples and averaging them, random variations are minimized and noise is reduced. It is not a huge, OMG!, knocks-your-socks-off level of improvement, but I can definitely see slightly smoother tones in very dark areas of my slides when using Vuescan's multiscan. And I think @Dazzer123's examples in post #15 demonstrate this, as well.

If there is a lot of noise in the near-black areas, then raising the shadows in post processing tends to make the noise look worse. So in my experience, files which were multisampled can tolerate a bit more shadow enhancement than those which were not. YMMV

I agree that two scans tend to smooth out the results, something that you can do with the noise slider and a single scan. But you're not getting more details worth anything. Normal scans get the most dmax out of the machine. They're design to get the max. You can't get blood from a turnip. Doing two scans isn;t going to raise the exposure that show more details. At least I haven;t ever seen anyone present samples that actually prove the difference. Just a lot of claims from the software manufacturers to sell their products. Use Epsonscan and save time and money.

 

[Dazzer123]

I tried to denoise in Lightroom to achieve the same effect in my example above and it wasn't as nice as the multi-scan.

I do have the feeling there is more true detail (for example in the brushed metal) in the multi-scan ...... and there was no sharpening used on either example.

I also notice the multi-scan is slightly moved, even though the two scans were done right after one another (is that part of the multi-scan, a slight movement shift?).

Of course this is all extreme and irrelevent pixel peeping, but interesting none the less.

 

[bfilm]

I also notice the multi-scan is slightly moved, even though the two scans were done right after one another (is that part of the multi-scan, a slight movement shift?).

If it is multi-pass multi-scanning, then there is a chance that the CCD sensor is not in the exact same place on each pass. But I think this would show up as reduced sharpness and not the whole image being moved.

 

[PhilBurton]

I'm scanning using a v850 Pro, Silverfast 9, flat settings (no tweaks) in Silverfast, scanning from the glass.

What are all the different settings in Silverfast? Which version?

So when i scan in Silverfast, the scan looks pretty dark.

Doesn't seem right.

 

[Dazzer123]

Silverfast 9, no settings changed, everything as flat as possible, glass scan.

Ooops, looks like Fuji Hunt E-6 for consumers is no more, perhaps this?:

Watkins Factor

 

[bfilm]

You could try the Bellini Foto E-6 chemicals.

 

[Dazzer123]

You could try the Bellini Foto E-6 chemicals.

I was only joking, i'll order some Bellini.

I'm going to take the same photo twice and develop with both Bellini and Cinestill to compare, as soon as my new (old) Apo Ronar 480mm f11 arrives!