ColorNeg: NegPos successor for colour neg scanning

[Leigh Perry]

There's been quite a bit of discussion in the past on this forum about NegPos, David Dunthorn's Photoshop plug-in for accurate handling of scanned colour negatives (and b&w).

David has now released a next-generation product called ColorNeg. You can find it here: http://www.c-f-systems.com/Plug-ins.html. As a committed NegPos user, I've only spent an hour or so playing with it, but if anything, I would say it gives even better colour conversions than NegPos did. The user interface is simplified, while keeping all the advanced functions still within reach.

If you are a registered NegPos user, ColorNeg will share your existing licence key. If you are not a NegPos user, and you shoot colour negative film, now might be a good time to check it out.

[Leigh Perry]

BTW: the essential step to getting a good conversion (inversion) is getting a linear 16 bit scan. There are some hints to getting a linear scan for some scanners on this page (http://www.c-f-systems.com/Scanners.html), including any scanner supported by VueScan. (Silverfast can also be persuaded to yield a linear scan, but I don't know the details.)

[Al Seyle]

Too bad it's PC only.

[Tim Lookingbill]

Are there any side by side comparison images using NegPos or ColorNeg as apposed to the scanner's software including time it takes for each to arrive at the results?

[Robert Budding]

a color conversion example is posted here:

http://www.daltonrooney.com/weblog/2008/04/16/taking-a-picture-part-four-analog-to-digital/

(not my work, but I'll have something soon).

[Kirk Gittings]

If you already own Silverfast with Negafix, after allot of side by side testing there is no advantage that I could ever find with NegPos. Unless I am missing something.

[Robert Budding]

I just bought a Nikon Coolscan 9000, and I'd rather not spend an additional $400 on Silverfast (particularly since I already paid for an Epson version of Silverfast). So the $67 for ColorNeg and ColorPos seemed reasonable. I'll run tests soon - as soon as my travel schedule lightens a bit.

I'll be happy to share a few 'RAW' scans with you, Kirk, if you'd like to compare your workflow results to my results. I am fairly new to scanning, though, so I do have a lot to learn (I've spent considerable time in darkrooms, mostly B&W, but some color, too).

[daverich4]

Just to throw this into the discussion. Nikon is not going to update the Nikon Scan software to be compatible with Mac OS 10.5. Silverfast has a version in the works, don't know about Vuescan.

http://nikoneurope-en.custhelp.com/cgi-bin/nikoneurope_en.cfg/php/enduser/std_adp.php?p_faqid=23460&p_created=1193919402&p_sid=oz3K-s1j&p_accessibility=&p_lva=&p_sp=cF9zcmNoPTEmcF9zb3J0X2J5PSZwX2dyaWR

[Robert Budding]

Ouch!

I do have Vuescan, so that may be my future choice (I plan to move to a Mac in the coming year).

[Christopher Hansen]

I run Nikon scan with my 9000 on a Mac with OS 10.5.1. It may not be optimized for the Intel processors or Leopard, but it does run.

[Ken Allen]

I'm always hoping that some tool like "ColorNeg", "NegFix", or others will meet my needs, but IMO nothing works better than making a linear scan (RAW scan) and manually adjusting the image yourself in photoshop using curves and Hue/sat. I know a few have posted their techniques, and I've posted a rough "How To" some time ago. So far the manual approaches seem to be similar. I'll post my approach in more detail on my site soon.

Some may find the manual approach difficult, but the more you do the easier it gets. We do all our client scans that way, then deliver both the Raw file and the color corrected. Since we do this often, it's gotten very easy, quick and the results just can't be beat.

Best,

[Tim Lookingbill]

Robert,

Thanks for posting the Dalton Rooney link to the ColorNeg comparison samples. I'ld plumb forgot about this thread since no one seemed to be able to post one sample comparison.

I'm assuming the comparison shots are the pic of the field of cattails at the bottom of the blog.

The appearance of the ColorNeg version looks as if it's still in RAW color space. I assigned sRGB when opened in PS 7 because it was left untagged. Is this greenish brown dull look typical of ColorNeg output?

I mean really. It's quite dull looking just like the one on the C-F system site. Couldn't make the color pop even applying Hue/Sat. That's when you know it's in RAW machine color.

However all I had to do to fix it was assign Joseph Holmes Ektaspace RGB, convert back to sRGB, neutralize in Levels using the highlite area of the branch as sampling point, apply 50% fade to the Levels neutralizer and the color just popped without appearing overly saturated and contrasty. Better than what could be achieved using the Epson software.

Since negs can't be color managed or profiled, I'm assuming ColorNeg must be CM'ed after the conversion to positive like I did by assigning JoRGB. If it works this good on the dull cattail image after what I did then it's at least a better starting point over any other method. I've tried Silverfast's NegaFix and all I got was even more dull brownish looking results on Kodak UC 400.

[dalton]

Hi Tim,
That's my photo with the cattails. I will say a few things in defense of the images posted:


I am not a color photographer by trade, I work in black and white mostly and some of the remarks you made below about color space and levels are a bit over my head.

The images posted were certainly not meant to stand up to any kind of scrutiny, they were produced months apart with absolutely no concern for a true comparison between the two and you will notice one of the links was actually to an image hosted on Flickr. I have absolutely no idea what kind of re-processing they do on images.

When I do work in color I prefer very low color saturation, and the image was processed to support that preference. I would be curious to see your version, I wonder if I would care for it?


But like I said, since I am not primarily a color photographer (I use ColorNeg almost exclusively for black and white, and I like it quite a lot for that purpose), my images are probably not the best test subject. The few times I have used it for color film, I have much preferred the ColorNeg version to a version that has been inverted by VueScan or the Epson software. I have tried to do a manual inversion a few times in PhotoShop with absolutely no luck.

I would be glad to answer any questions about the software I can if you guys are still curious, and maybe even learn a thing or two about color while I'm here. You won't change my mind about my preferences for low saturation, though! ;)

Dalton



Robert,

Thanks for posting the Dalton Rooney link to the ColorNeg comparison samples. I'ld plumb forgot about this thread since no one seemed to be able to post one sample comparison.

I'm assuming the comparison shots are the pic of the field of cattails at the bottom of the blog.

The appearance of the ColorNeg version looks as if it's still in RAW color space. I assigned sRGB when opened in PS 7 because it was left untagged. Is this greenish brown dull look typical of ColorNeg output?

I mean really. It's quite dull looking just like the one on the C-F system site. Couldn't make the color pop even applying Hue/Sat. That's when you know it's in RAW machine color.

However all I had to do to fix it was assign Joseph Holmes Ektaspace RGB, convert back to sRGB, neutralize in Levels using the highlite area of the branch as sampling point, apply 50% fade to the Levels neutralizer and the color just popped without appearing overly saturated and contrasty. Better than what could be achieved using the Epson software.

Since negs can't be color managed or profiled, I'm assuming ColorNeg must be CM'ed after the conversion to positive like I did by assigning JoRGB. If it works this good on the dull cattail image after what I did then it's at least a better starting point over any other method. I've tried Silverfast's NegaFix and all I got was even more dull brownish looking results on Kodak UC 400.

[Tim Lookingbill]

Hi Dalton,

I've posted the edited version to your original cattail shot. Since you don't like saturation, I didn't apply it. I still assigned JoRGB, converted back to sRGB and did a levels neutral eyedropper adjust but reset the blue gamma slider back to 1.00 and increased the blue highlite to bring out the blue sky, Faded Levels set to Color Blend at 50% opacity.

I can't believe after two years this is the only ColorNeg sample anyone has posted to this thread. What's up with that?

[dalton]

Thanks, Tim, for posting that. I do prefer the colors in your version, it makes mine look quite dull. I am going to experiment a bit with your workflow the next time I scan color film and see if I can get it to work for me.

Regards,
Dalton

[Robert Budding]

I'll try to get few scans done and processed with ColorNeg in the coming weeks - my travel schedule is pretty difficult theses days. A warning, though. I am not that experienced with scanning.

[Tim Lookingbill]

Dalton,

Thanks for the kind remarks.

I'ld like to add that if you do apply my workflow, it might not give consistant results scene to scene. Your cattail image has specific memory colors that JoeRGB just happen to bring out. Getting accurate looking to the scene color temp is the most important. You may have to eyeball it from memory, because you may not have a truly neutral object to sample from in the scene. I barely had one in the cattail image if it weren't for that one branch.

If you apply this workflow to more colorful scenery with a lot of primary colors, you may get unexpected hues and saturation levels requiring selective hue/sat adjusts by color which really doesn't take that long to do. It's good that ColorNeg renders this way as long as it does it the same way to all captured scenes.

This kind of hopefully achievable consistancy is the equivalant of linearizing your color negatives through ColorNeg and the scanner so they can be color managed and edited in a much wider color space in 16bit requiring the least amount of work.

The key here is to get the same dull looking conversions as you did on the cattail image and assign JoeRGB to those other scans. You might try assigning other wider than sRGB gamut profiles like CIERGB, NTSC, etc. to other frames with a combo of selective hue/sat.

You'll have to rely on your memory of the scene to guide you because even though JoeRGB made the cattail image look good it doesn't mean it's meant for it or is accurate. It's a colorspace designed for color slide film gamut as the Ektaspace name in the profile suggests.

Anyway, hope this helps.

Robert,

I haven't scanned for over a year so I can understand. I've been currently researching the bewildering world of Raw converters with my recently purchased Pentax K100D DSLR.

I pretty much gave up scanning my 35mm negatives which became a lot of work with less than desireable results in regards to color on my Epson 4870. Just wanting to see if this new ColorNeg software would make me want to return to scanning my negatives. It's just FYI for me because I don't think they have a ColorNeg version for Mac OS anyway.

[dalton]

Dalton,
Just wanting to see if this new ColorNeg software would make me want to return to scanning my negatives. It's just FYI for me because I don't think they have a ColorNeg version for Mac OS anyway.

Tim,
I'm using ColorNeg on Mac. The author even made a custom version for me with a large image window because I have a very big screen. I have tried it with 10.4 and 10.5. It works perfectly with Vuescan and my Epson 4990.

Regards,
Dalton

[Tim Lookingbill]

Thanks, Dalton.

I'll check it out and see how it works when I get some time.

[Tim Lookingbill]

BTW, have you tried getting a 16bit linear scan using the Epson's scan driver that produces tiffs ColorNeg does well with?

I'ld like to not have to purchase an additional piece of software like Vuescan on top of purchasing ColorNeg just to get a linear scan if I can avoid it.

[dalton]

BTW, have you tried getting a 16bit linear scan using the Epson's scan driver that produces tiffs ColorNeg does well with?

I'ld like to not have to purchase an additional piece of software like Vuescan on top of purchasing ColorNeg just to get a linear scan if I can avoid it.

I have not been able to find any information on getting a linear scan from Epson software. I don't think it's possible... Epson always manipulates the image as far as I can tell. There is more information about using Vuescan on the CF website

http://www.c-f-systems.com/Scanners.html#Epson

with absolutely no mention of the Epson software.

[Robert Budding]

I finally did a few scans and posted them here:

http://photo.net/digital-darkroom-forum/00MRuZ

ColorNeg adjustments are a lot better than what I can do in Nikon Scan. I don't have Silverfast and, at $400 for the Nikon 9000 version, I will not be buying it in the future. Over all, the $67 I spent for the C-F Systems plug-ins was a fantastic deal.

I also tried the invert filter in Photoshop. The results were dreadful!

[Robert Budding]

I've had good success with the CF System COlorNeg plug-in. But I also have color darkroom experience, so using CC filters is second nature.

I've started a Flickr user group to share tips and tricks here:

http://flickr.com/groups/977912@N23/

[dalton]

Since this thread still seems to get a fair amount of traffic, (noted in the referral logs on my website) I thought I would post a follow-up on my use of ColorNeg. I have been shooting color film for the last six months or so (I previously used ColorNeg for B&W only, as odd as that sounds), and I have gotten pretty familiar with the software. Here are some of my latest color photographs:

http://www.flickr.com/photos/daltonrooney/tags/colorneg/

It took me a little while to really figure out what the software does, but I feel like I have a pretty solid workflow going now. It beats the pants off the results I have been able to get straight from Vuescan, NikonScan, or Silverfast.

I use Fuji Pro 160S, but I have found that the best preset to start from is Ektacolor. Sometimes if I am not happy with any of the presets, I go into the Film Data slider and find something close. It is usually a good start.

I then manipulate the lightness, color adjust, and shadow settings independently to get exactly what I want. Shadows seems to be hugely important - all of my photos were coming out with a magenta cast until I learned how to fine-tune the shadows, which erased it completely.

The image I get from of ColorNeg is then almost completely neutral. I use some basic curves, hue/saturation, color balance, and occasionally photo filters in Photoshop to get exactly what I want.

This image in particular is an example that was giving me a lot of trouble until I learned to use ColorNeg properly.

http://www.flickr.com/photos/daltonrooney/3286128383/sizes/o/

Previous versions of the image had a strong magenta/purple cast that I could not get rid of. It's still challenging, it's not perfect, but it's a huge improvement over previous versions.

Regards,
Dalton

[Peter De Smidt]

Dalton, thanks for the update.