First C41 roll how do I know if it's success?

keyofnight · Feb 20, 2013

So I developed my first roll (Ultramax 400) in Kodak Flexicolor SM chemistryusing those weird F1 and F2 processing units. To my surprise: images came out without too many noticeable problems. I inspected the negatives for streaks and ripples, scanned a few images to get used to the workflow, but honestly I can't tell if I did a good job or not.

Why? I have to edit these pictures a bit to get them usable. I end up: reversing colors (of course), doing a curves adjustment to set black point (because the film base is so dark/orange), setting white balance, setting levels. How would I know if I were having color shifts? What are some problems I should keep an eye out for?

Here's a sample from the roll. Let me know what you guys think.

By the way: I got a lot of help from you guys, so thanks a lot!

bvy · Feb 20, 2013

Congrats. It's a nice shot. Probably not the best subject matter for anyone to judge your processing (beyond confirming that there are no streaks, spots, etc. -- at least in this one frame). Also, the fact that you adjusted contrast and color balance after scanning doesn't help. You've introduced too many variables.

One thing I found helpful was to scan my own C-41 processed film alongside a lab processed film of the same type and subject matter. And I mean scan the whole negative -- sprockets and all. I placed them side by side on the scanner -- three frames from one film on top, and three more on the bottom. Then I took a high-res scan of the six frames with the "no color correction" option checked and compared. It's hardly scientific, but it might help.

polyglot · Feb 20, 2013

Neon light has a pretty narrow and spiky spectrum so it's basically impossible to tell from that shot if your process is OK because the green- and blue-sensitive dyes were probably not exposed at all. You certainly couldn't tell if there was crossover or anything subtle like that.

If you want to evaluate whether your process is working, you need a range of identifiable colours in the test scene and a fairly full range of greys from very-dark to nearly-white. The usual answer is a Gretag-Macbeth chart, or you can just use your hand (for skintones) and a handful of primary-coloured and white plastic things. I went with my hand, a blue plastic bottle, can of coke and a box of HP5+ to get the primary colours, range of greys and skintones. Light it with a flash if you have one, otherwise direct sunlight.

There's a C41-scan article in my FAQ (link in signature), which may or may not help you...

keyofnight · Feb 21, 2013

polyglot said:
Neon light has a pretty narrow and spiky spectrum so it's basically impossible to tell from that shot if your process is OK because the green- and blue-sensitive dyes were probably not exposed at all. You certainly couldn't tell if there was crossover or anything subtle like that.

I honestly didn't post that image as an exampleI was just proud of it.

polyglot said:
If you want to evaluate whether your process is working, you need a range of identifiable colours in the test scene and a fairly full range of greys from very-dark to nearly-white. The usual answer is a Gretag-Macbeth chart, or you can just use your hand (for skintones) and a handful of primary-coloured and white plastic things. I went with my hand, a blue plastic bottle, can of coke and a box of HP5+ to get the primary colours, range of greys and skintones. Light it with a flash if you have one, otherwise direct sunlight.

I think I'll go the route you did, or just look through the roll for better test shots. I had a picture of different fruits that I wanted to post, but I scanned it at such a low resolution by accident. Whoops. I think I'll post it anyway. Here goes:

While we're talkingI have to say that your FAQ has been a huge help over the last year, so thanks a lot.

polyglot · Feb 21, 2013

Glad to be of service!

There's something wonky with your scan though as the highlights are clipped to pink (it's clearly a digital artefact, not chemical). My guess is you adjusted the colour balance after setting the black & white points, which means clipping the image and then modifying the colours so the net result is the white-point is no longer white. You need to get the colours right during the inversion before clipping the dynamic range to an 8-bit file.

wblynch · Feb 21, 2013

I think your first C-41 was a success! The fruit looks great.

keyofnight · Feb 21, 2013

wblynch said:
I think your first C-41 was a success! The fruit looks great.

Thanks! ;D

polyglot said:
There's something wonky with your scan though as the highlights are clipped to pink (it's clearly a digital artefact, not chemical). My guess is you adjusted the colour balance after setting the black & white points, which means clipping the image and then modifying the colours so the net result is the white-point is no longer white. You need to get the colours right during the inversion before clipping the dynamic range to an 8-bit file.

Just for the record, I use a Hasselblad Flextight X1 and it's wonky Flexcolor software. I don't make adjustments to scans while I'm scanning beside the automatic inversion it does (I should turn that off too, I think, but I imagine it saves me a ton of time). I do all of my adjustments in Aperture 3. With that said… let me think about what might've gone wrong. OH! If my settings reset to 300dpi, they certainly reset the bit depth to 8-bit too. Cool! When I do a real scan on the whole roll—at 16-bit—I probably won't have that problem, right?

I'll post again with more pictures.

polyglot · Feb 21, 2013

Maybe. The problem is order of operations, not the absolute bit depth.

chuck94022 · Feb 21, 2013

From a home processing and home post processing perspective, I think any result that gives you an acceptable image is by definition success. One way to check your results is to take your negatives down to the local mini lab (assuming this is 35mm) and ask them to make you a print. If *they* can print it, and the result looks good, I'd say your negs are just fine. Only time will tell if you got the "permanence" part right.

In terms of hybrid post processing, Polyglot's recommendations are fine. There are also software packages that attempt to automate the orange mask removal, with varying levels of quality. That's probably more of a DPUG discussion, but Polyglot's info is very good for doing it yourself.

chuck94022 · Feb 21, 2013

keyofnight said:
Thanks! ;D

Just for the record, I use a Hasselblad Flextight X1 and it's wonky Flexcolor software. I don't make adjustments to scans while I'm scanning beside the automatic inversion it does (I should turn that off too, I think, but I imagine it saves me a ton of time). I do all of my adjustments in Aperture 3. With that said let me think about what might've gone wrong. OH! If my settings reset to 300dpi, they certainly reset the bit depth to 8-bit too. Cool! When I do a real scan on the whole rollat 16-bitI probably won't have that problem, right?

I'll post again with more pictures.

(Apologies in advance to the APUG community for the following...)

One thing to note: Aperture 3 is one of the worst tools you can be using to do these adjustments, because it does not have a simple Invert control. Instead, you have to invert using the Levels tool, when ends up reversing every other control you will be using. It absolutely sucks. Photoshop is really the best tool I've found when trying to work with adjusting color negatives, then once you've got the inversion done, you can use Aperture for the rest of your work. (I know, Photoshop is an expensive tool... maybe you can get away with using GIMP, but it has been a while since I've looked at it.) On the Mac, you might be tempted to try Pixelmator, since it is much cheaper than Photoshop and seems to be powerful. Don't. It doesn't have individual channel levels, and it only does 8 bits per channel. Worthless for this.

keyofnight · Feb 24, 2013

chuck94022 said:
One thing to note: Aperture 3 is one of the worst tools you can be using to do these adjustments, because it does not have a simple Invert control. Instead, you have to invert using the Levels tool, when ends up reversing every other control you will be using. It absolutely sucks. Photoshop is really the best tool I've found when trying to work with adjusting color negatives, then once you've got the inversion done, you can use Aperture for the rest of your work. (I know, Photoshop is an expensive tool... maybe you can get away with using GIMP, but it has been a while since I've looked at it.) On the Mac, you might be tempted to try Pixelmator, since it is much cheaper than Photoshop and seems to be powerful. Don't. It doesn't have individual channel levels, and it only does 8 bits per channel. Worthless for this.

Yeahit's a little dumb how there's no "invert" tool in Aperture 3, but oh well. Up until now, I gotten around Aperture 3's limitation here by using Hasselblad's Flexcolor scanning software. I don't adjust levels, add sharpening, add curves, remove casts; I turn all of the correction tools off. Then I set the film type to "color negative," and it does the inversion step. I'm not sure if it does anything else.

I hope it doesn't, because I'm pretty close to doing it the way you're suggesting to do it. I have a copy of Photoshop (and if I didn't, my university has site licenses that I'm probably entitled to as a staff member/student) and I'm not afraid to use it. After all, it shouldn't be too difficult to just make an action to invert and (maybe) remove the orange castthen I can use that action to bulk edit. Eh. I might as well. Once thing are set up, it's just a few clicks worth of extra work. I'll try it out on my next few batches.

MattKing · Feb 24, 2013

Try FastStone Image viewer - it is available on the internet for free (donation requested), it has an inversion ("Negative") function, an excellent resizing tool, a very functional clone and heal function and lots more - and it is quick!

Now back to our regularly scheduled programme .

EDIT: oops - just realized you are on a Mac

First C41 roll how do I know if it's success?

Moderator

Privacy & Transparency

Privacy & Transparency