You Can Stop Missing Kodachrome Now

[Guest malland]

Advertisement (gone after registration)

Steve, I have used Color Efex extensively. Generally, I found that it is best to pull back on saturation on files that you sent through the Film Effects filter, especially for the Kodachrome effect. Also, the Kodachrome effect exaggerates reds and yellows. While I had some interesting results, it is a very different tool than RPP, which, of course is a raw processor. As stated in the original posting, I find RPP unique because it "developed" files with greater resolution, apparently owing to its use of floating point calculations, and has better color rendition than other raw developers I have tried, which include Aperture, ACR, Irident, Lightzone (now defunct, Silkypix and Capture One.

 

—Mitch/Chiang Mai

Days and Nights in the Forest (WIP)

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[stunsworth]

An interesting thread, but to stop people running out to buy a Mac can I just mention that Color Efex Pro has a film preset for Kodachrome and can be run on Windows as a Photoshop plugin

 

The same applies to Alien Skin Exposure.

[250swb]

Steve, I have used Color Efex extensively. Generally, I found that it is best to pull back on saturation on files that you sent through the Film Effects filter, especially for the Kodachrome effect. Also, the Kodachrome effect exaggerates reds and yellows. While I had some interesting results, it is a very different tool than RPP, which, of course is a raw processor. As stated in the original posting, I find RPP unique because it "developed" files with greater resolution, apparently owing to its use of floating point calculations, and has better color rendition than other raw developers I have tried, which include Aperture, ACR, Irident, Lightzone (now defunct, Silkypix and Capture One.

 

—Mitch/Chiang Mai

Days and Nights in the Forest (WIP)

 

 

Yes, the success of the film presets does depend on a pretty neutral image as the starting point. Otherwise all sorts of odd things can happen. As I tend to use both Silver Efex and Color Efex a lot I don't try to generate anything special during RAW processing, a neutral or low saturation and low contrast image has all the tonal information and offers more scope for further post processing.

 

Steve

[Guest malland]

Yes, the success of the film presets does depend on a pretty neutral image as the starting point. Otherwise all sorts of odd things can happen. As I tend to use both Silver Efex and Color Efex a lot I don't try to generate anything special during RAW processing, a neutral or low saturation and low contrast image has all the tonal information and offers more scope for further post processing.

 

Steve

Steve, I also use Silver Efex for B&W, but generally find the results better doing the raw conversion in RPP: better resolution and much lighter adjustments in Silver Efex, just some burning and dodging — and often I don't use any of the film presets. Often, but not always, I end up with a better look than using Aperture as the raw developer. For B&W, I haven't yet figured out how to judge whether I'll get better results using RPP; for color most of the time the RPP-developed file will look better.

 

—Mitch/Chiang Mai

Days and Nights in the Forest (WIP)

[wildlightphoto]

I downloaded RPP and have started working with it. Looks promising.

[plasticman]

Mitch - your enthusiasm is infectious. I'm gonna give RPP another chance.

 

Thanks for sharing your experience, anyway - even if I ultimately stick with RawDev, it's always interesting to give these smaller, low-profile applications a try. In general, they're often produced by 'geeky' enthusiasts who have a genuine interest in the science of imaging and less so in squeezing money out of their client-base in endless rounds of meaningless upgrades.

[Guest malland]

Advertisement (gone after registration)

I downloaded RPP and have started working with it. Looks promising.

Mitch - your enthusiasm is infectious. I'm gonna give RPP another chance...

You would do well to keep in mind that the learning curve is steep and that the way of working with RPP is different, and that you may initially be discouraged. In the last month, I tried it initially and was discouraged. Then, I started to use it just to check how my exposure was when I was starting "exposing to the right", and at one point I figured out how to use it and started to like it.

 

Exposing to the right is another major issue. Eventually, I concluded that optimum exposure for digital is not the same as for film. I used to think that it was necessary to expose digital for the highlights, like transparency film; but eventually figured out that this was try only for high contrast situations, when, there is no headroom to expose to the right. I don't want to start the ETTR argument here, but only would like to mention that it does give me better gradation than I was getting before. Essentially, in exposing to the right you want to get the best possible file, without worrying how the file looks when you open: in a "standard" raw developer, i.e. not in RPP. you simply pull back on the exposure slider until the image looks like you want to. I only mention ETTR here because it is a useful method to keep in mind when you use RPP, about which you can read in the RPP documentation.

 

—Mitch/Chiang Mai

Days and Nights in the Forest (WIP)

[kevinparis]

Steve, I have used Color Efex extensively. Generally, I found that it is best to pull back on saturation on files that you sent through the Film Effects filter, especially for the Kodachrome effect. Also, the Kodachrome effect exaggerates reds and yellows. While I had some interesting results, it is a very different tool than RPP, which, of course is a raw processor. As stated in the original posting, I find RPP unique because it "developed" files with greater resolution, apparently owing to its use of floating point calculations, and has better color rendition than other raw developers I have tried, which include Aperture, ACR, Irident, Lightzone (now defunct, Silkypix and Capture One.

 

—Mitch/Chiang Mai

Days and Nights in the Forest (WIP)

 

Mitch... I gave RPP a try yesterday.... and in my experience I am not seeing any improvement in resolution over Aperture... if you are seeing this then it won't be attributed to Floating point calculations as Aperture, as I suspected also uses floating point calculations as outlined at the end of this document

 

http://manuals.info.apple.com/en/aperture_photography_fundamentals.pdf

 

cheers

 

K

[Guest malland]

Mitch... I gave RPP a try yesterday.... and in my experience I am not seeing any improvement in resolution over Aperture... if you are seeing this then it won't be attributed to Floating point calculations as Aperture, as I suspected also uses floating point calculations as outlined at the end of this document

 

http://manuals.info.apple.com/en/aperture_photography_fundamentals.pdf

 

cheers

 

K

Kevin, I am not an expert on floating point calculations, and when I first tried RPP and simply looked at two files, I also didn't at first think that there was an increase in resolution. It was only when I looked at both an Aperture-developed and an RPP-develooed files that I saw a clear difference. Also, I don't know what camera you were using, but with the Ricoh GXR M-Module and the Elamarit-21 ASPH I found a substantial resolution improvement over Aperture. I don't know whether these differences are as compelling with files from sensors that require sharpening. I tend to think that Doug Herr will see this difference with his DMR, which I suppose is what he is saying when he writes above that he finds RPP promising. The other thing is that I found that I was also getting better color developing files with RPP. Keep in mind that RPP is only a raw developer and that I do further processing in Aperture — it's not a replacement for Aperture.

 

—Mitch/Chiang Mai

Days and Nights in the Forest (WIP)

[ismon]

All very well and good, but how does it look when projected 50X through an Elmaron?

[kevinparis]

Mitch... I am no expert on any kind of calculations - floating point or not... but something in the deep recesses of my brain from my Apple days reminded that when we were talking about Shake (Apples effects and compositing software), that a big thing was made of the floating point stuff.

 

I believe that some of the Shake technology was probably used in the design of Aperture - namely the floating point calculations and the concept of concatenation to prevent unnecessary calculations. The latter is why you can't change the order of the adjustments in Aperture.

 

My brief tests were done with E-P1 files - for which there is a specific raw conversion profile - I am guessing that your Ricoh files will use DNG, and perhaps the generic DNG convertor within Aperture isn't fine tuned to the M-Module files.

 

anyway hope you are well and with dry feet in Bangkok

 

K

 

Kevin, I am not an expert on floating point calculations, and when I first tried RPP and simply looked at two files, I also didn't at first think that there was an increase in resolution. It was only when I looked at both an Aperture-developed and an RPP-develooed files that I saw a clear difference. Also, I don't know what camera you were using, but with the Ricoh GXR M-Module and the Elamarit-21 ASPH I found a substantial resolution improvement over Aperture. I don't know whether these differences are as compelling with files from sensors that require sharpening. I tend to think that Doug Herr will see this difference with his DMR, which I suppose is what he is saying when he writes above that he finds RPP promising. The other thing is that I found that I was also getting better color developing files with RPP. Keep in mind that RPP is only a raw developer and that I do further processing in Aperture — it's not a replacement for Aperture.

 

—Mitch/Chiang Mai

Days and Nights in the Forest (WIP)

[andybarton]

How can software improve resolution?

[kevinparis]

In absolute terms I am guessing not... I think what mitch may be seeing in in RPP as opposed to Aperture is differences in perceived resolution due to sharpening applied at the Raw decoding stage.

 

I am no expert in image processing... but I personally believe that the 'differences' between raw convertors are minimal - they all start with the same data and can only do so much with interpreting that data.

 

There also is the issue of where the raw conversion process stops and the image manipulation process starts.

 

kevin

 

 

 

 

How can software improve resolution?

[250swb]

How can software improve resolution?

 

Probably in terms of the micro contrast, which isn't real resolution, but it looks like a sharper image. I found this in a brief experiment with RawTherapee that produced wonderful looking images straight out of the bag. But delve deeper and all sorts of artifacts seemed to be generated in doing so. I didn't go much further, but I admit I didn't try to correct the initial defaults.

 

Steve

[Guest malland]

...My brief tests were done with E-P1 files - for which there is a specific raw conversion profile - I am guessing that your Ricoh files will use DNG, and perhaps the generic DNG convertor within Aperture isn't fine tuned to the M-Module files...

How can software improve resolution?

Probably in terms of the micro contrast, which isn't real resolution, but it looks like a sharper image. I found this in a brief experiment with RawTherapee that produced wonderful looking images straight out of the bag. But delve deeper and all sorts of artifacts seemed to be generated in doing so. I didn't go much further, but I admit I didn't try to correct the initial defaults...

Kevin, Andy, and Steve, clearly a raw processor cannot bring out resolution that is not there, but I understand from the RPP documentation that they are not doing it through sharpening or micro contrast. However, the RPP files have this higher resolution and have the what I would call the greater color clarity, that is even evident in the JPGs that I have linked above — that is a look that I have not been able to get from any other raw developer. That is certainly there, as I see with my own eyes and with how positively people have reacted to the pictures. In comparison, when I try to get the look I want, other raw processors produce some weird looking pictures which look overprocessed and don't maintain visual integrity. Look at a photograph in which I was trying to get a certain look and compare the Apertrure and RPP versions by clicking here. That is the best I could do with Aperture, but it pushed me past where I wanted to be. BTW, RPP produces much better reds.

 

I would be interested in knowing what Doug Herr finds because he works carefully and has a good eye, Most people who react negatively to RPP simply don't try to get over the learning hump and haven't tried it enough.

 

—Mitch/Chiang Mai

Days and Nights in the Forest (WIP)

[wildlightphoto]

I would be interested in knowing what Doug Herr finds because he works carefully and has a good eye, Most people who react negatively to RPP simply don't try to get over the learning hump and haven't tried it enough.

 

I wouldn't try to read too much into my observations because I haven't used Lightroom or Aperture or any version of Photoshop beyond 6 (which is OS 9 software ). I'm also an engineer by training and in my day job I write engineering software so RPP's reliance on numbers (vs. sliders) doesn't intimidate me.

 

For me the most important criteria for choosing a raw converter is 1) image quality, followed closely by 2) does it run on my crusty old G4 and 3) can I afford it. At a suggested donation cost of $25 the 32-bit version of RPP easily meets all three of these criteria.

 

I'm comparing RPP to Imacon Flexcolor which in my experience has produced excellent 16-bit TIFF files from the DMR's DNGs. After about an hour of playing with RPP I've been able to match or exceed the detail and color I've been getting with Flexcolor. In addition, RPP doesn't crash the computer, has more predictable file handling and is being updated.

 

I really really like the highlight compression exposure adjustment. It's something I could do by adjusting curves in Flexcolor or Photoshop but it's a whole lot more predictable with RPP. I'm still looking for a curves-like control but I won't dump the program if it doesn't have it. I'll be re-processing a number of raw files over the next several weeks and forming my impressions and opinions not with rigorous tests but by simply comparing the best I can do with RPP against the best I can do with Flexcolor. The goal of my test is to get the best quality TIFF files given my functional requirements, my skills and my obsolete computing environment (unfortunately the computing environment isn't going to change any time soon). My impression right now is that the space Flexcolor is occupying on my hard drive will soon be put to more productive use.

[Guest malland]

Doug,

 

Thanks for the information. I know where you're coming from because I have a Flextight Precision 3 scanner, so I know the FlexColor softwaro. As far as I remember FlexColor. RPP should provide better control and results. Do you get the fantastic color you get in your DMR wildlife shots by using only FlexColor, or do you go into Photoshop under OS9 after that? Incidentally, I see that the DMR Is one of the cameras that supported through an RPP camera profile. so that you should have a good starting point.

 

As for curves In RPP, my Impression in that the closest you'll come is what is covered in the following Q&A from the RPP manual:

I’ve applied Compressed Exposure correction and it helped, but my images still lack the “punch” I’m used to with other converters. Why?

Because RPP doesn’t do any automatic corrections. If you want more contrast use the Black Point field; try 0.1-0.3. The Contrast control is another way to do this. Try 4 and see if you want more or less.

For some files, I have gone as far as 0 6 In the Black Point field. The RPP manual states the following on the Black Point control:

The Black-Point adjustment in RPP alters the black point with a film-like roll-off to avoid rough shadow clipping. When the black point is raised, shadows are compressed instead of being cut off. This allows fairly large corrections to be made without ruining an image. The scale is in percentages; zero means no correction and 99 means an almost entirely black image. In most cases even a small black-point correction does much to raise contrast. Values in the range of 0.1%-1% are usually enough. Negative values, which will lift shadows, are also allowed.

Most people will want to use RPP for raw processing only, and there is provision for using it as a Lightroom plugin. But is there any good shareware image editor that you can use for additional adjustments after RPP?

 

In any case, it would be great if you could report back on your learning experience with RPP because I have only been using it for a week and am a novice with it.

 

—Mitch/Chiang Mai

Days and Nights in the Forest (WIP)

[wildlightphoto]

Do you get the fantastic color you get in your DMR wildlife shots by using only FlexColor, or do you go into Photoshop under OS9 after that?

 

Primarily Flexcolor. I always go into Photoshop afterward even if only to clean up dust spots, but Photoshop gives me more opportunities to fine-tune the black point, white point, color balance and contrast. My primary tool in Photoshop is Curves. Another good image editing program is Gimp (free, my kind of price) but IIRC Gimp doesn't support layers which I'm finding to be a very handy feature for the more difficult photos.

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[Guest malland]

In post #9 above, I gave a suggested workflow for using RPP, as follows:

 

1. Set the B&W balance if there is neutral gray in the picture (Command-click on the neutral area);

2. Select the KR64 color film curve;

3. Set the exposure using Compressed Exposure;

4. Set the Black Point (try 0.3 or 0.6); and

5. Set the Saturation (try 30-50).

 

However, if RPP has a camera profile for your camera, the Saturation adjustment may not be necessary, or may need to be much less than in step 6. Also, initially I started using a Local Contrast setting of 1-5, but now find that I need 10-30.

 

—Mitch/Chiang Mai

Days and Nights in the Forest (WIP)

[Guest malland]

I've now added another step to the ones in the post above:

 

6. Set Local Contrast. Try 10-30.

 

—Mitch/Chiang Mai

Days and Nights in the Forest (WIP)