Colorperfect Plugin
Theodor Urena
This happens due to one of two things:
a) Color space conversion before using ColorNeg
b) Black Point adjustment by Scanner software or user before using ColorNeg
For EpsonScan do follow these instructions precisely:
www.colorperfect.com/scanning-slides-and-negatives/scans/...
After Scanning that way do use our auxiliary program MakeTiff to assign sRGB to the Tiff.
www.colorperfect.com/MakeTiff/
Section: "Stripping profiles from linear scans or batch assigning the desired one"
Make sure in PS / PSE that you do not convert to any other color space upon opening the file, that may be specified on your color preferences.
Just try ColorPerfect's ColorNeg if the problem is gone, it's gone if not work your way through this:
www.colorperfect.com/working_spaces.html?lang=en
With Epsonscan bear in mind that the result is not the linear file ColorPerfect expects. On the otions select "dual Gamma C", revert (
Sorry, you are having problems, you do have to be careful that everything is set up right on your scanner and in ColorPerfect. But when you get it down it will be a most useful tool. I've made a few personal tweaks but I think you will like it when you learn how to use it.
Also, If you follow the the method, which is easy for RAW files, you will never use anything else.
69 months ago(permalink)
I tried to get it work with the Hasselblad/Imacon-Scanners for about half a year but with no success. We (Christoph - the developer of Colorperfect - and I) tried everything that came to our mind (to be fair, mostly his mind).
Also contacted Hasselblad but they weren't very helpful. Maybe I'll contact them again in the near future.
For now there is no possibility to get "correct" results with these scanners and Colorperfect.
If you only have an Imacon-Scanner the results you get might look very reasonable but only until you scan the same negative with any other scanner - then you can see the difference and this is sometimes a huge difference (only in very rare cases the results looked very similar).
there is also one other problem: 3F-Scans are pre-sharpened and there is no way to turn this off. If you develop your negative in Flexcolor and then export it as Tif with USM set to -60 it blurs the image in an "OK way" that this isn't of much concern, but it is a "problem" if you work with the 3F-files directly in Photoshop (at least for me).
I also asked Hasselblad about this but they didn't want to help me on this either.
ages ago(permalink)
Daniel,
thank you for your update.
By using my workflow and comparing the histogram with the one from the same image scanned on an Epson V750 + Vuescan (RAW) + ColorPerfect I can't notice big differences. But I haven't instruments and knowledge other than the histogram to do comparisons.
Moreover, if it's true that there is no way to get correct results from an Imacon scan with ColorPerfect I strongly suggest you (and Christof) to remove the incorrect informations and workflow about Imacon scans from all the ColorPerfect sites.
I bought ColorPerfect just because on the site it is indicated as a working solution also with Imacon scanners.
ages ago(permalink)
At the moment I can't access a secondary scanner (I sold my V750) but I'll try to post an example as soon as possible.
I saw your sample, did you ASSIGN 'Flextight Input'" first and then CONVERT to your working profile before opening the RAW scan with the ColorPerfect plugin?
(Ok, sorry. About the wrong informations on the site I'll send a message to Christian...)
Originally posted ages ago. (permalink)
mristuccia edited this topic ages ago.
This is a damm shame as I'm tired of being stuck in photoshop trying to get accurate colours out of imacon scans.
My question is, does using the ColorPerfect plugin with the above or any other work flow, offer any aesthetic advantage over just sticking with Flexcolor?
ages ago(permalink)
To be honest with you I can't get any good color rendition with FlexColor whatever setting I try. I played for days with histogram, auto/manual settings, gray point, etc... and can't get the same good results I get with 3f+ColorPerfect and my workflow.
Maybe we get different results with different versions of FlexColor or with different OS Platform. I indeed use Mountain Lion 10.8.2 and the last FlexColor, the 4.8.13.
ages ago(permalink)
It's strange because sometimes images processed with ColorPerfect are very similar and sometimes you can see a huge difference.
There does not seem to be any logic in it and that's why we never found out what might be causing this and we tried many things (e.g. measuring the values of color negative grayscales and comparing the values in Excel).
ages ago(permalink)
No argument here. :)
And as I already said, I'm a huge supporter of ColorPerfect and if you are happy with the results you get then nothing should stop you to do so.
What bothers me most about the 3F-scans is the pre-sharpening issue. The basic sharpening of these files is just too much for me. I already asked Hasselblad what's that all about but they weren't very willing to change this in a future software update.
ages ago(permalink)
: )
You are right about the sharpening. But I always do a post-scan sharpening to images, maybe less than this, but I do.
Moreover, if you're right about the fact that in Flexcolor the negative sharpen settings corresponds to "blur", we can also blur the opened 3F file in PS, just before or after applying the ColorPerfect plugin. (but indeed I would not have heart to do this : ).
I'm proceeding with the analysis, by comparing your two RAW negatives (Coolscan and Flextight) and I made a big discovery:
==> it seems that in 3F files only the red channel is gamma encoded and it is encoded to 1.8.
The other channels, the G and B, seems having only a different contrast and brightness, or maybe are linear.
By using ImageMagick I was able to convert the 3F file to be very similar to the Coolscan RAW, by linearising only the red channel gamma.
And indeed by applying ColorPerfect on that I obtained an image identical to the Coolscan one.
Unfortunately I no longer have a different scanner other than my 848 so I can't produce by myself other test scans to proceed with my analysis and to find the correct conversion formula.
ages ago(permalink)
Didn't have time yet. I worked on it only a little bit.
Only to say that the "plain" conversion of ColorPerfect on both the linear and the normal 3f files produce an identical final result.
And that again, with the usual white balance correction, it is beautiful and natural in respect to the Coolscan one (white balanced), which comes out more saturated and less balanced. But I was not there in front of the real subject so don't know which image is the more realistic.
If I have the time will work more on this the next week, by seing the numbers...
Thanks again for sharing this.
ages ago(permalink)
Hi Gusy, I have read with great interest your dsicussion. I'm trying to experiemnt with Colorperfect which I still find is the better conversione mode for 3F simil-linear scan. I would wonder to understand how to linearize 3F scan with Imagemagic. Any suggestion? thanks, Giuseppe.
100 months ago(permalink)
Hi Duncan,
unfortunately I don't think so.
This documentation is there since the beginning of this discussion but it seems it does not tell the true. And I still don't understand the reason why they don't remove that page. I wasted my money by believing through that page that ColofPerfect could have been used with the Imacon FlexTight scanners.
83 months ago(permalink)
Why bother with C41 Colour Negative Scanning? I think Andy Barton put it perfectly when he wrote C41 film, when exposed correctly and processed properly, will always give you highlights that are under control and shadows that are detailed. It has a dynamic range way beyond E6 slide film and has a beautiful, slow tail in the highlight end of the histogram when compared to digital shots, which might as well have an on/off switch when it comes to blown highlights. No digital capture can compete to a properly exposed and processed C41 negative, scanned by someone who knows how to scan, with a suitable scanner.
The following isn't intended to imply in any way that I'm an expert on scanning. I've been doing it for about a week! There is surprisingly little on the net on the subject of colour negative scanning and I hope that posts following in this thread will only provide further tips to successful scanning with further insights into the software mentioned.
The overview of the process is this: Vuescan gets all the information from the negative which is passed to Photoshop still in negative form. We would convert to a positive in Photoshop but it isn't very good at it so we convert to a positive with the Photoshop plug-in ColorPerfect.
There is an excellent article here Creating linear scans / VueScan Professional - ColorNeg - Your RAW converter for negative scans which you should read. I haven't tried the colour brightness stuff at the bottom yet but you should see if it makes a difference to your scans especially if you have an old scanner.
AT THIS STAGE ONLY CHECK THE MULTIEXPOSURE BOX. Your scanner will probably do a couple of scans later in the scanning stage one at a higher intensity to bring out the detail in the dark bits of the negative.
Curve low and Curve High This relates to an S curve of the levels/contrast curve. I've zeroed these as I believe ColorPerfect wants the info as untouched as possible. I must admit I don't know if it's better to do this. I think Julian Thompson leaves his at the default value of 0.25 and 0.75 respectively. See which you prefer.
c80f0f1006