PDA

View Full Version : Make C41 look like E6? A challenge!



timparkin
29-Jun-2011, 06:29
I've read quite a few long threads on this forum about whether or not you can make C41 films look like E6 and they have gone backwards and forum ad infinitum. I thought it a good idea to challenge the 'yes you can' camp to put there money where there mouth is..

The following article shows normalised scans of all the available large format colour films used to take the same shot.

http://www.landscapegb.com/2011/06/colour-film-comparison-pt-3/

The challenge is to take the C41 photos and convert them to look like the E6 ones. That's a lot of combinations so I thought I'd make some specific challanges

1) Take the Portra 160, Pro160S and Ektar and make them look like the velvia

The real part of the challenge is to provide the photoshop file that contains the layers used to do so!

http://static.timparkin.co.uk/static/landscapegb/17/film-comparison/all/Velvia%2050.jpg
http://static.timparkin.co.uk/static/landscapegb/17/film-comparison/all/Ektar%20default.jpg
http://static.timparkin.co.uk/static/landscapegb/17/film-comparison/all/Pro160S%20default.jpg
http://static.timparkin.co.uk/static/landscapegb/17/film-comparison/all/Portra%20160%20default.jpg

These were all scanned on a Howtek 4500 using DPL Pro to set neutral black and white points. If anybody would like straight scans just pm me.

Look forward to seeing the results!

BennehBoy
29-Jun-2011, 06:34
Tim, I'm really not convinced by the C41 samples that you have on that site, they look _very_ flat to me. Could you describe your scanning AND post process workflow for us?

I abandoned the built in inversion and color correction of my scanning software (vuescan) some time ago and now use the very latest techniques available withing ColorPerfect (which has come on leaps and bounds since it's ColorNeg days).

timparkin
29-Jun-2011, 08:21
Tim, I'm really not convinced by the C41 samples that you have on that site, they look _very_ flat to me. Could you describe your scanning AND post process workflow for us?

I abandoned the built in inversion and color correction of my scanning software (vuescan) some time ago and now use the very latest techniques available withing ColorPerfect (which has come on leaps and bounds since it's ColorNeg days).

As mentioned, scanned using a linear scan and then set the black and white points. I have refrained from adding contrast to this. C41 is naturally very low contrast (you can't get nearly 19 stops in a picture without it looking low contrast).

If you look at the source article you can see where I've added basic post processing for the whole picture and then a 'graduated' contrast in the sky.

Larry Gebhardt
29-Jun-2011, 09:24
You really should provide us with the full 16bit files if you want to see what can be done. Also I wish you had included the film frames around these so we can really see what was in the shadows relative to the film base.

Also, how were the slides and negatives metered? I meter totally differently for each.

Edit: I missed your note saying to email you for straight scans. I'll do that.

timparkin
29-Jun-2011, 10:06
http://static.timparkin.co.uk/static/tmp/lfpi/Ektar%20100%20DPL.jpg- (916kb)
http://static.timparkin.co.uk/static/tmp/lfpi/Ektar%20100%20pp.jpg- (346kb)
http://static.timparkin.co.uk/static/tmp/lfpi/Ektar%20100.tif- (27.82Mb)
http://static.timparkin.co.uk/static/tmp/lfpi/Portra%20160%20DPL.jpg- (826kb)
http://static.timparkin.co.uk/static/tmp/lfpi/Portra%20160%20pp.jpg- (172kb)
http://static.timparkin.co.uk/static/tmp/lfpi/Portra%20160.tif- (27.95Mb)
http://static.timparkin.co.uk/static/tmp/lfpi/Velvia%2050.tif- (27.96Mb)

The DPL files are scanned using Aztek DPL Pro's ability to set curves in hardware. Black and white points were set but no gamma was applied.
The pp files are my quick conversions (not intended to look like chromes but intended to be more 'photographic').
The main tif files are scanned using the wide gamut negative uncorrected setting in Aztek DPL which gives the best open scan (and most close to linear) possible.

Be aware I have added a dash to the end of each url so that crawlers and pre-page fetchers don't end up eating my bandwidth. Just click on the file and then remove the trailing - when you get a 404 page.

I've put the images into Ekta Space where I have remembered. jpgs are 8 bit, tiffs are 16 bit.

I'll scan the Pro160S if anybody expresses an interest

The slides were spot metered to give a reasonable highlight at +1.5 stops on the poppies and then gradded to place the highlight in the sky about +2 to 2 1/3 (with Velvia rated at 40)

The negs were metered to give a mid tone on the brighter foliage/poppy heads and were rated at box speed and not gradded.

Tim

timparkin
29-Jun-2011, 10:32
Tim, I'm really not convinced by the C41 samples that you have on that site, they look _very_ flat to me. Could you describe your scanning AND post process workflow for us?

I abandoned the built in inversion and color correction of my scanning software (vuescan) some time ago and now use the very latest techniques available withing ColorPerfect (which has come on leaps and bounds since it's ColorNeg days).

Hi Ben, I've added a couple of post processed versions above to show very roughly how I'd start processing these. - Tim

Kirk Gittings
29-Jun-2011, 10:47
I have sold allot of stock files from scanned transparencies to magazines for articles and then had to supplement them with fresh photos. If that was at a time when I was just shooting color negs for scanning or digital I would have to match the "look" of those files to to the look of the transparencies. I didn't find it all that hard, When I had the scans of the transparencies on hand to work towards. Curves, hue, saturation, clipping -its fairly straight forward.

timparkin
29-Jun-2011, 10:58
I have sold allot of stock files from scanned transparencies to magazines for articles and then had to supplement them with fresh photos. If that was at a time when I was just shooting color negs for scanning or digital I would have to match the "look" of those files to to the look of the transparencies. I didn't find it all that hard, When I had the scans of the transparencies on hand to work towards. Curves, hue, saturation, clipping -its fairly straight forward.

Would love to see how you do ..

Kirk Gittings
29-Jun-2011, 11:06
Sorry, It is a rare proposition these days. It has probably been a couple of years since I last was asked to do such a match.

engl
29-Jun-2011, 11:10
I've been thinking about a challenge very much like this, but I think the challenge would be more relevant for real shooting if there was no Velvia reference frame to compare to. Perhaps that could be the next challenge :) ?

But anyway, it will be very interesting to see how this turns out, good luck to all who participate!

timparkin
29-Jun-2011, 11:13
I've been thinking about a challenge very much like this, but I think the challenge would be more relevant for real shooting if there was no Velvia reference frame to compare to. Perhaps that could be the next challenge :) ?

But anyway, it will be very interesting to see how this turns out, good luck to all who participate!

Well I have another frame I can supply with no reference for the next challenge.. :-)

I agree that this is more realistic though.. I would hope that the 'best' conversion includes very few local colour or contrast changes (apart from maybe a grad) but we'll see what results we get.

Tim

timparkin
29-Jun-2011, 11:14
Sorry, It is a rare proposition these days. It has probably been a couple of years since I last was asked to do such a match.

Oh go on!! After all it's "fairly straightforward" ;-)

timparkin
29-Jun-2011, 14:27
Here is the Pro160S version..

http://static.timparkin.co.uk/static/tmp/lfpi/Pro160S.tif- (29.3Mb)

And here are the tests that you don't get to see the Velvia 50 for... :-)

http://static.timparkin.co.uk/static/tmp/lfpi/test1-pro160s.tif- (29.4Mb)
http://static.timparkin.co.uk/static/tmp/lfpi/test2-160nc.tif- (29.2Mb)
http://static.timparkin.co.uk/static/tmp/lfpi/test3-pro160s.tif- (29.3Mb)

Once I've had a two or three trys, I'll publish the actual chrome version (I'll publish test 1 in a couple of weeks, test 2 a couple of weeks later -- etc.. give people a chance to play..)

Tim

Struan Gray
29-Jun-2011, 14:46
Couldn't resist. Done on an eyeball-calibrated laptop, so just an indication of which buttons I'd push.


http://struangray.com/miscpics/test2-160nc.jpg
Taken from the 160NC 16-bit file

.

timparkin
29-Jun-2011, 14:57
Couldn't resist. Done on an eyeball-calibrated laptop, so just an indication of which buttons I'd push.


Taken from the 160NC 16-bit file

.

Not a bad start! Thanks for taking part..

Bruce Watson
29-Jun-2011, 15:15
Why would anyone want to make C41 look like E6? The C41 process generates considerably more accurate color -- that's what that orange mask is for. C41 is also capable of much greater dynamic range; you can get great shadow detail without blowing out highlights.

There are only two advantages to E6 that I can come up with. One is WYSIWYG which is largely an advantage if what you're doing is time sensitive -- but if it's really time sensitive you'd probably be doing digital capture and cutting out the time and expense of film processing completely. The other advantage is that with E6, the graininess is in the shadows where it's less visible. But since this is a LF group, by the time you hit 5x4 this advantage is largely meaningless.

So the *real* question to me is: Can you make E6 look like C41? And the answer is: no, you can't.

Larry Gebhardt
29-Jun-2011, 15:16
Here's my attempt with only across the board changes, except for one layer to simulate the graduated ND filter. Converted to sRGB for web display. The PSD file is tiny, but you can simply drag the layers over to the full sized image to see the effect.

I know it's not an exact match, but it's close. The part that's the farthest off is the sky, and I wonder how much better that would have been if the neg was shot with the same grad.

Taken from the Portra 160 file.

Larry Gebhardt
29-Jun-2011, 15:21
So the *real* question to me is: Can you make E6 look like C41? And the answer is: no, you can't.

That's why I usually shoot C41 now. But I do like the look of Velvia, when the contrast range isn't too great, so I do shoot it on occasion.

timparkin
29-Jun-2011, 15:39
Why would anyone want to make C41 look like E6? The C41 process generates considerably more accurate color -- that's what that orange mask is for. C41 is also capable of much greater dynamic range; you can get great shadow detail without blowing out highlights.

There are only two advantages to E6 that I can come up with. One is WYSIWYG which is largely an advantage if what you're doing is time sensitive -- but if it's really time sensitive you'd probably be doing digital capture and cutting out the time and expense of film processing completely. The other advantage is that with E6, the graininess is in the shadows where it's less visible. But since this is a LF group, by the time you hit 5x4 this advantage is largely meaningless.

So the *real* question to me is: Can you make E6 look like C41? And the answer is: no, you can't.

Completely agree personally but some people don't, hence the challenge. As for why you would want to do so - ask the thousands of people addicted to Velvia 50..

Velvia 50 in particular has an amazing ability to purify colour, perhaps inaccurately but in a lot of peoples eyes, very aesthetically. Also, slide film has an different highlight handling than neg - it's highlights very often hold colour - adding to the rich tones..

In other words there are lots of reasons to want to have a film that is so tolerant of exposure and that handles large dynamic ranges but with the tonality of the velvias family for instance.

Accurate colour is definitely *not* what a lot of people want ;-)

Tim

Drew Wiley
29-Jun-2011, 15:43
Nothing beats a chrome for just plopping down on a lightbox and seeing exactly what you got. Hues tend to be a lot cleaner within the dynamic range of the film. And there
are certain instances when higher contrast is a distinct advantage. With color negs you've got a few more hoops to jump through, but also some new opportunities.

timparkin
29-Jun-2011, 15:44
Nothing beats a chrome for just plopping down on a lightbox and seeing exactly what you got. Hues tend to be a lot cleaner within the dynamic range of the film. And there
are certain instances when higher contrast is a distinct advantage. With color negs you've got a few more hoops to jump through, but also some new opportunities.

Oh they definitely have new opportunities - I'm loving learning to use neg films and they've made me reassess the post processing of my E6 films too...

Tim

timparkin
29-Jun-2011, 15:49
Here's my attempt with only across the board changes, except for one layer to simulate the graduated ND filter. Converted to sRGB for web display. The PSD file is tiny, but you can simply drag the layers over to the full sized image to see the effect.

I know it's not an exact match, but it's close. The part that's the farthest off is the sky, and I wonder how much better that would have been if the neg was shot with the same grad.

Taken from the Portra 160 file.

Interesting - nice tonality in the greens. The reds are a *lot* more muted. The sky is always an issue but I would have thought that if the colours aren't changing depending on exposure, just reducing the exposure should produce something similar. However I really think Velvia does something extraordinary in skies (not sure why or how).

Thanks!

Tim

Drew Wiley
29-Jun-2011, 16:06
Me too, Tim. But that wasn't the case a few years back. Color neg films have come a
long way. My main motivation for shooting chromes now is for Cibachrome and dye transfer printing; but I have plenty of older shots on hand to keep me printing these
media as long as I wish, or as long as the supplies can still be scraped together. So I still keep a little 8x10 E100G loaded in the holders just in case the subject matter warrants it. But mainly, it looks like the new Kodak neg films are going to be more versatile and much more affordable to print; and the less types of film I have to lug around, the better. Strictly speaking, one cannot simply interchange one type for
another in just any situation; but neg films have become versatile enough to cover
most of the bases, and with better contrast range. Yet there's nothing like Vevia in
a deep fog when the entire scale is only two or three stops. I suspect even Ektar
would be a bit disappointing under those circumstances.

timparkin
29-Jun-2011, 16:12
Yet there's nothing like Vevia in
a deep fog when the entire scale is only two or three stops. I suspect even Ektar
would be a bit disappointing under those circumstances.

This is one of the critical things for me with landscape photography - beautiful colours often rely on a low dynamic range film to be able to capture all of the tonality.

Combined with some slide films colour couplers ability to push colours out to the primaries, it makes shooting in overcast conditions a treat. choice between e6 and c41 is like N-1 and N+1 ... maybe ... :-)

Tim

Drew Wiley
29-Jun-2011, 16:24
Tim - I just don't know how far Ektar can have its contrast boosted before something
goes awry. I do have some interesting 8x10 low-contrast fog shots to try, but just
haven't got to them yet. With the moderate amount of magnification generally involved
with large format negs, I suspect I can control the issue with basic contrast increase
masking. I'll let folks like you experiment with the Photoshop options. With small format,
however, major contrast changes become a lot more difficult because of the way the
dye curve geometry tends to change. With chromes you have relatively high contrast
to begin with and work backwards, which is sometimes easier, sometimes not. Anyway,
it's all a lot of fun, and I certainly enjoy the opportunity to learn some new tricks.

Struan Gray
30-Jun-2011, 00:29
The sky is always an issue but I would have thought that if the colours aren't changing depending on exposure, just reducing the exposure should produce something similar. However I really think Velvia does something extraordinary in skies (not sure why or how).

Portra (at least, the recently replaced NC Portra) goes cyan when overexposed. It shows up most clearly in indoor photos, where a brightly-lit scene outside can turn quite ugly, or cool, depending on your taste. Daytime skies don't suffer too much, but at twilight if you expose to get detail and good colour in the ground the sky can end up fairly anaemic. I quite like the look, but it certainly doesn't have the deep blue Caribbean beach look of Velvia.

The one thing I couldn't do in my quick jiggle of the contrast in your image was to simulate the way Velvia and other saturated films push all colours towards the primary hues. When I switched from small format slides to rollfilm and LF negative I was initially disappointed by the yellowy-greens I was getting in a lot of grass and foliage. Then I looked again, and realised that I was simply used to the way most slide films turned all greens into a single undifferentiated colour. One of my favourite uses of neg film is to (try to) capture the subtle minimalist variations in colour across salt marsh grasses and moorland.

But those deep blues are nice.

BennehBoy
30-Jun-2011, 01:55
Hey tim, thanks for the scans, will have a play about tonight - I don't expect to ever match the E6, but it ought to be simple enough to breath some life into the C41 images.

Drew Wiley
30-Jun-2011, 08:41
Traditionally color neg films have been engineered to dump the neturals into a pleasing
skintone range, and then try to sort out any saturated primaries on the side if possible.
To some extent 160VC and now especially Ektar do a much better job of rendering saturated hues cleanly, but you really need to find the sweet spot in the exposure,
color temp balance, and finally printing contrast. But once you do that, the results are
pretty remarkable and don't look stereotypically color neg at all. A wide variety of greens with reproduce vividly and cleanly. Blues and reds can be done nicely too. I can't speak for inkjet output, but with the newest Type II Crystal Archive paper, even yellows will come out cleanly and vibrant. Not exactly a substitute for the gleaming primaries and deep blacks of Cibachrome, but overall, even better with the wide range of subtle hues. You just have to more finicky with your procedure if you want those especially saturated color to land squarely. Every week lately I'm putting difficult
color negs in the enlarger just to learn some new idiosyncasie of the system. It's like
learning to ride a bicycle all over again. You get a few scrapes and bruises, but that's
just part of the learning curve.

JimL
1-Jul-2011, 01:43
Here's my attempt at Portra 160NC > Velvia 50 - maybe a touch overcooked :) . I added a high radius USM layer to get more "pop". Interesting exercise. I was working with the 3rd Velvia image on your comparison page: Whole Photograph (added grad to neg and shadow/highlight to slides).