How do you scan color?

[Photo_Smith]

I must be about the only person in the thread that gets good or at least what i feel satisfactory colours on both the V500 and V750. The resolution is not good for 35mm but I shoot 120 and 4x5 mainly for web view so there isn't an issue regarding quality especially when so few viewers will have CMM in place and probably view images on glossy, too bright TN displays.

Viewed on a decent screen with CMM I can get accurate colour:

At lest the values measured on screen match the values of the input....

 

[PatrickT]

I must be about the only person in the thread that gets good or at least what i feel satisfactory colours on both the V500 and V750. The resolution is not good for 35mm but I shoot 120 and 4x5 mainly for web view so there isn't an issue regarding quality especially when so few viewers will have CMM in place and probably view images on glossy too bright TN displays.

Nope! As you can see waaaaay back in page 3, I was able to get good color from my V500 as well (at least, what I consider to be good color). That's the route I'm going to take from now on, Epson Scan, a few tweaks, and then into Lightroom. I tested my "workflow" on 6 separate negatives and it seems to work pretty well!

 

[Pete B]

I must be about the only person in the thread that gets good or at least what i feel satisfactory colours on both the V500 and V750. The resolution is not good for 35mm but I shoot 120 and 4x5 mainly for web view so there isn't an issue regarding quality especially when so few viewers will have CMM in place and probably view images on glossy, too bright TN displays.

Viewed on a decent screen with CMM I can get accurate colour:

At lest the values measured on screen match the values of the input....

Is this with EpsonScan or Silverfast? My problems were with Vuescan +/- ColourPerfect.
Pete

 

[Photo_Smith]

I agree with the standard being arbitrary. But for the purposes of achieving relatively uniform results for the consumer across a multitude of labs, that standard becomes the bull's eye in the target. When those labs consistently achieve those standards within a small tolerance (I assume the tolerance would mean differences are not readily detectable by the human eye) then those labs are both precise and accurate. So I disagree that there is no accuracy. The accuracy is based on the standard even if the standard is arbitrarily designated the manufacturer.
.

The manufacturer didn't design the standard, the one they use is called the Munsell colour system– mid grey aim being Munsell N 5.
The Munsell system is derived from the mapping of the human visual response. In a sense Munsell using the number 5 as half of decimal base ten is arbitrary, the mid point measure of stimuli of the rods in your retina isn't.

Perhaps you think 18% figure used in grey cards are arbitrary? Or is there a link between 18% and Munsell N 5 ?? (the measured mid point stmuli )

 

[Pete B]

Was your colour card scanned with Epsonscan or Silverfast?
Pete

 

[Photo_Smith]

Is this with EpsonScan or Silverfast? My problems were with Vuescan +/- ColourPerfect.
Pete

I can't remember exactly, I use Vuescan, Epson and Silverfast-my guess is Vuescan (if you're dripping water on my forehead)

Point being I can do that with any film/scanner/software combination.
Vuescan and Silverfast are the hardest to use with marginally better results but a lot more control over the *th degree of accuracy and more control (which can be a double edged sword for some)

I would say Epson scan gets you 90% of the way, the other 10% will require learning the way the more expensive packages work it's up to you to decide if the pain is worth the gain.

My advice to beginners is to keep it as simple as possible until you're confident, then slowly get to use the other packages jumping straight in might make some feel overwhelmed.

 

[Pete B]

I can't remember exactly, I use Vuescan, Epson and Silverfast-my guess is Vuescan (if you're dripping water on my forehead)

Point being I can do that with any film/scanner/software combination.
Vuescan and Silverfast are the hardest to use with marginally better results but a lot more control over the *th degree of accuracy and more control (which can be a double edged sword for some)

I would say Epson scan gets you 90% of the way, the other 10% will require learning the way the more expensive packages work it's up to you to decide if the pain is worth the gain.

My advice to beginners is to keep it as simple as possible until you're confident, then slowly get to use the other packages jumping straight in might make some feel overwhelmed.

So, what do you do differently? This is what I was never able to get to the bottom of because no one seemed able or willing to explain. People would talk of LAB and other stuff, the usual RGB clipping, using a whitebal dropper,...go and research etc (I don't know what they thought I was doing) but ultimately, I wasn't able to achieve it with the Epson, but the Plustek just seems to work as people say it should, no fuss. I'm really interested to hear what it takes to get good colour from the Epson V700 using Vuescan. Who ever writes that will be the first person to do so on the net, just as Edge100 is the first person to think of showing their routine from scan to ColourPerfect to Lightroom/photoshop.
In the end I was so exasperated I resolved to send it out to a lab, until the Plustek came along.
Pete 🙁

 

[TXForester]

The manufacturer didn't design the standard, the one they use is called the Munsell colour system– mid grey aim being Munsell N 5.

When I said standard, I was referring to how a negative film with specific development and printing on a specific paper would look. Ektar and Portra developed in the same chemicals, with recommended temps and times and printed on the same paper would have different looks. This is designed by the manufacturer, so the user can achieve different looks (and the manufacturer can sell more film).

If I went to different labs with rolls of Ektar and Portra and the shots are the same subject with the same light, I'd expect my Portra to prints to look similar (allowing for some tolerance) from all the labs and the same goes for my Ektar. That is the standard or bull's eye for the labs to aim at. Assuming here the labs use the same machines, chemistry and paper. However I would not expect my Ektar prints to look like my Portra, otherwise I'd only need one film.

 

[edge100]

I must be about the only person in the thread that gets good or at least what i feel satisfactory colours on both the V500 and V750. The resolution is not good for 35mm but I shoot 120 and 4x5 mainly for web view so there isn't an issue regarding quality especially when so few viewers will have CMM in place and probably view images on glossy, too bright TN displays.

Viewed on a decent screen with CMM I can get accurate colour:

At lest the values measured on screen match the values of the input....

We agree on something! The V500 is a pretty crappy film scanner (though I'm sure it's great to scan prints of your Aunt Mable's 80th birthday), but if you can't get good colour out of it, you're doing something wrong.

One of the things I love most about Internet photographers is their propensity to always and everywhere blame their tools first. Yes, the V500 is crap (and the V700 is slightly less stinky crap), but both are better than you are.

 

[Pete B]

but if you can't get good colour out of it, you're doing something wrong.

How can I learn how to do something if no one tells me how?
Pete

 

[edge100]

How can I learn how to do something if no one tells me how?
Pete

I have told you how.

 

[Pete B]

But I'm refering to the Epson, not the Plustek. This is what we're talking about at the moment, the difference between the Plustek and the Epson raw files. Remember when earlier you were saying "Derrrr, of course there's a difference, the Epson isn't a dedicated film scanner" shortly before your present stance (based on no experience with the Epson scans) that the Epsonscan files process just as readily.
Pete

 

[edge100]

But I'm refering to the Epson, not the Plustek. This is what we're talking about at the moment, the difference between the Plustek and the Epson raw files. Remember when earlier you were saying "Derrrr, of course there's a difference, the Epson isn't a dedicated film scanner" shortly before your present stance (based on no experience with the Epson scans) that the Epsonscan files process just as readily.
Pete

Who said I have no experience with Epson scans? You assumed that.

The Epson isn't a dedicated film scanner. I'm sure that it doesn't produce scans as nice as the Plustek. I'm also sure that you can get something useable from it, with minimal effort.

Send me an Epson linear raw from Vuescan. Make sure it's of a well-exposed (but not horridly over-exposed) negative.

 

[Pete B]

Who said I have no experience with Epson scans?

Sorry, I thought you said earlier you hadn't used them.

Here's a jpeg of a Plustek Vuescan raw:

Here's a V700 Vuescan raw:

There's an obvious difference in colour. Not what I was expecting from Linear raw files.
I'm about to process them so hold on.
Pete

 

[edge100]

Sorry, I thought you said earlier you hadn't used them.

Here's a jpeg of a Plustek Vuescan raw:
View attachment 98793
Here's a V700 Vuescan raw:
View attachment 98794
There's an obvious difference in colour. Not what I was expecting from Linear raw files.
I'm about to process them so hold on.
Pete

What film??

 

[brbo]

It's clear that one is scanned as slide film (Plustek) and the other (V700) as negative. That's why they look so different. Properly inverted files are much closer than RAWs. My CanoScan 9900f RAW scans are green. Like green green. Doesn't make a difference.

I've already explained in this thread the difference in RAW files when different media input types are selected (or assumed by scanning software).

2s per image inverts in ColorPerfect...

 

[edge100]

It's clear that one is scanned as slide film (Plustek) and the other (V700) as negative. That's why they look so different. Properly inverted files are much closer than RAWs. My CanoScan 9900f RAW scans are green. Like green green. Doesn't make a difference.

I've already explained in this thread the difference in RAW files when different media input types are selected (or assumed by scanning software).

While I agree that linear raw files don't always look identical, ALL of my negative scans (scanned as negative, not slide) from VueScan (SprintScan 120) look like the Plustek here.

 

[Pete B]

It's clear that one is scanned as slide film (Plustek) and the other (V700) as negative. That's why they look so different. Properly inverted files are much closer than RAWs. My CanoScan 9900f RAW scans are green. Like green green. Doesn't make a difference.

I've already explained in this thread the difference in RAW files when different media input types are selected (or assumed by scanning software).

I believe they're both scanned with the same Vuescan settings:
Plustek

V700


And this shows the problem I have. Whereas the Plustek scan gives normal reds, the V700 scans give magenta reds.

 

[brbo]

While I agree that linear raw files don't always look identical, ALL of my negative scans (scanned as negative, not slide) from VueScan (SprintScan 120) look like the Plustek here.

That is possible. AFAIK Plustek 120 doesn't even have exposure control, let alone individual channel gains (in Vuescan at least). Probably every other semi-serious scanner exposes negative film differently than slide film (longer G and B channel exposures to counter the negative orange mask). I'm not saying Plustek 120 is bad because of this (quite possibly that sensor in Plustek 120 doesn't need this and is able to record all B and G channel info with single exposure). So, every RAW scan in Plustek 120 with Vuescan would be the same, regardess of input media type (negative or slide).

Im 100% sure that V700 was scanned as negative, though.

 

[edge100]

I believe they're both scanned with the same Vuescan settings:
Plustek
View attachment 98795
V700
View attachment 98796

And this shows the problem I have. Whereas the Plustek scan gives normal reds, the V700 scans give magenta reds.

Is your screen calibrated? On my calibrated 15" MBP, the V700 is very slightly more magenta, but I have to switch back and forth between the images to really see it.