Filmomat SmartConvert. (Negative conversion software)

This will sound like an advertisement, and it kind of is, but I'm in no way affiliated with the company. I just think good products deserve some recognition.

One of the reasons I stopped shooting digital and went back to film is that I don't like the process of digitally editing photos, and I never had a digital camera that produced jpg's to my liking. Scanning and converting negatives yourself still involves quite a bit of digital post processing, especially color negatives. I've tried Negative Lab Pro, but I don't use Adobe products anymore, so that's off the table. For a while now I've been using Affinity photo, and it works quite well. Still, some color negatives are troublesome and require a lot of work. Out of desperation I did a google search about a week ago, and came across Filmomat SmartConvert. My expectations were low but I downloaded the demo and was pleasantly surprised. The demo lets you try unlimited conversion, but present them in lower resolution, and with a water stamp. The full version lets you export in TIFF and JPG, and has taken every camera raw format I've thrown at it this far. I paid the 99 euros for the full version the same day, and have been very happy this far.

The number of adjustments in the software are quite limited, but it does a really good job out of the box, plus you can point it to a folder and tell it to convert every image that pops into that folder, which is great when you scan or tether your digital camera. You don't even have to include any film base in your scans for setting white balance, although I have no idea how the software manages it.

That's enough from me. Again, not affiliated with Filmomat. But hey, Filmomat! Now you know where to find me, if you should be so inclined.

These came straight out of the software. Some of you may remember the Harley from another thread. The Harley took 5 seconds to convert and has had no adjustments whatsoever done to it. The boat was cropped in SmartConvert and then had the whitebalance automatically recalculated to exclude the cropped out negative holder. The Harley was shot on Portra 800 and the boat on Harman Phoenix 200.

I just saw this post and have to give another recommendation for Filmomat Smart Convert, i heard about it by chance on Nico’s photo news some time ago. I also do not use Adobe any longer and Smart Convert works very quickly to process a number of scans of color negatives. I also use it to convert black and white negative scans if they are in the same folder. I think it’s well worth the $100 I paid for it.

How are you scanning the negatives?

mshchem said:

How are you scanning the negatives?

Click to expand...

DSLR. Well, mirrorless to be precise, but by camera.

I’m camera scanning with a Sony A7Riii and Samsung 70mm macro lens. I find the process goes quickly though what I did last time was use auto crop and then usually adjust each crop to my liking as I went along. For the black and white negatives I changed them to that setting as I went along also. I do find it much simpler to use than Negative Lab Pro as well.

I have Nikon Coolscan machines but so freaking slow.

It's been discussed a few times here recently. Some good info in these threads:



I have used it quite a bit and while I was originally very impressed (especially with the workflow) a few deal breakers became apparent after some time. White point handling and clipping is especially problematic. I do really like the controls though. YMMV

radialMelt said:

It's been discussed a few times here recently. Some good info in these threads:

Nikonscan 4.0.3 C41 inversions and RA4 paper colour signature question

This question straddles the two realms of fully analogue and hybrid film photography and I'm not sure where to post it - so please mods move as appropriate. I have been experimenting a lot with medium format colour negative film as of late. I am mostly a black and white photographer and I...

www.photrio.com

Filmomat Smart Convert

Has anyone already tested the newly released Filmomat Smart Convert for converting negatives? Information according to the manufacturer: „CONVERTING NEGATIVES MADE EASY. FOR COLOR, B/W, NEGATIVES AND SLIDES. A STANDALONE TOOL - NO ADDITIONAL SOFTWARE REQUIRED. NO SUBSCRIPTION. BUY ONCE, USE...

www.photrio.com

I have used it quite a bit and while I was originally very impressed (especially with the workflow) a few deal breakers became apparent after some time. White point handling and clipping is especially problematic. I do really like the controls though. YMMV

Click to expand...

I haven’t noticed any issues. What problems do you have with white point and clipping?

It's discussed in depth in the threads that I've shared.

radialMelt said:

It's discussed in depth in the threads that I've shared.

Click to expand...

I haven’t noticed any of the issues, but haven’t needed to manipulate the tiffs yet. I’ll go have a try.

I've done some tests and you definitely have a point @radialMelt

This was just a quick test where I used SmartConvert to convert both files. One is a raw-file straight from camera, the other is the same raw-file processed to show a more balanced histogram and exported as a tiff. Given how much of a difference I see here I might try exposing differently in the future.

If they upgraded it to include a histogram I might consider trying it out again - at least then it would be possible to determine if data is being clipped or not. In my case I was feeding it raw DNGs from VueScan (as I use a Coolscan 9000). Perhaps the app is built specifically for DSLR scans.

For now I've decided to stick with ColorPerfect. The UI is nightmare but I have a simple workflow going that gives me the most flexible possible files to edit within LR.

cerber0s said:

a raw-file straight from camera

Click to expand...

But it's converted to a viewable format for display. How did you do this in this case? Information like that is essential in interpreting the comparison.

koraks said:

But it's converted to a viewable format for display. How did you do this in this case? Information like that is essential in interpreting the comparison.

Click to expand...

I might be misunderstandbg your question, so if my answer makes no sense at all you know why

SmartConvert handles raw files directly. So in one case I just dragged the raw file into SC, in the other I first tweaked it in Affinity Photo, exported it as a tiff (still a negative), and then dragged the tiff into SC.

Ah, right; sorry, I misunderstood!
The question then still remains what tweaks you did in Affinity Photo to make the TIFF file. That SmartConvert gives a different output for a different input in itself isn't too surprising. In fact, that's what you want it to do, I'd say.
Keep in mind that a RAW file is fundamentally different from a TIFF file, JPG, PNG or any other bitmap file format. In a way, a RAW file is not even an image (yet). It's still open for interpretation, whereas a bitmap format is already interpreted.

koraks said:

Ah, right; sorry, I misunderstood!
The question then still remains what tweaks you did in Affinity Photo to make the TIFF file. That SmartConvert gives a different output for a different input in itself isn't too surprising. In fact, that's what you want it to do, I'd say.
Keep in mind that a RAW file is fundamentally different from a TIFF file, JPG, PNG or any other bitmap file format. In a way, a RAW file is not even an image (yet). It's still open for interpretation, whereas a bitmap format is already interpreted.

Click to expand...

Ah right! The histogram in the raw file was pushed to the left, resulting in a conversion with highlights blown out more than necessary. I adjusted exposure and shadow/highlights to “center” the histogram.

Nothing of that is strange or unexpected, but what I take from the test is that SmartConvert doesn’t have a function for bringing “the best” out of the negatives, at least when it comes to exposure compensation. Nor does it include enough information in the resulting tiff to do it afterwards. Next I will see what happens if I overexpose when I digitize the negative.

Having to manipulate the raw files before bringing them into SC ruins an otherwise wonderfully streamlined workflow.

I also did the test above using a color negative, but I couldn’t make it look better by editing it first, here SC did a pretty good job.

Maybe @Filmomat can shine some light here?

cerber0s said:

SmartConvert doesn’t have a function for bringing “the best” out of the negatives

Click to expand...

I don't think that's a realistic expectation of any tool, since it's so subjective. A tool can never know what you want with a particular image.
Which of these two is 'best'?

They're two different interpretations of the same image (and not a particularly strong one at that, but that's a different matter). Neither has clipped highlights or shadows, so 'technically speaking' they're both 'good' (apart from the posterization because I did a quick & dirty edit of an 8-bit original). Neither version is what I'd consider a desirable end result, btw, but again, that's not really the point of the illustration.

cerber0s said:

Nor does it include enough information in the resulting tiff to do it afterwards

Click to expand...

IDK; I can't tell that from the example you posted. Both versions have some clipping in the shadows, although that's kind of natural given the subject matter. They're not very different w.r.t. the shadow detail they contain and what limitations I see in recovering it, stem mostly from the 8-bit web versions I can work with. Neither has any substantial highlight clipping. They're really just the same image with a different (non-linear) compensation curve applied to it. Which version you prefer is a matter of taste and intent.
The only key difference between the images is that one seems to be corrected for barrel distortion while the other isn't.

Maybe I'm just misunderstanding what you're trying to establish here, and/or I'm overlooking something obvious w.r.t. technical quality. I do admit that I have very little confidence that any tool will magically guess consistently what you're trying to achieve as a photographer, and as a result the only really sensible thing to do is to take control over curve adjustments and to apply whatever technique required to get to the desired end result. I don't really 'believe in' tools that perform black-box adjustments (e.g. selective shadow or highlight enhancements) based on some kind of algorithm that decides without knowledge of the image content, photographer's intent etc. what would be a proper end result. It's very well possible that a given tool may give an outcome that's generally pleasing to a particular photographer and that may be good enough if they don't want to bother doing any manual work on the images. Again, that's something I consider so far removed from what B&W film-based photography actually has to offer that I can't really wrap my head around why someone would want to adopt such an approach. Given this skepticism, ignorance and inflexibility on my part, I think it's best if I try not to interfere any further. Sorry for barging in in the first place, in fact...

koraks said:

I do admit that I have very little confidence that any tool will magically guess consistently what you're trying to achieve as a photographer, and as a result the only really sensible thing to do is to take control over curve adjustments and to apply whatever technique required to get to the desired end result.

Click to expand...

I totally agree with everything you say. I don’t expect it to guess what I want, the “problem” is that it doesn’t let me do what I want. There is no curve adjustment and no exposure compensation in SC, which means that it’s not possible to get the photo to look like the topmost car photo. To acheive that I first have to take it through some other software.

This is not a dealbreaker but I think it’s a shame since the SC workflow is so streamlined otherwise. That’s all I’m saying

cerber0s said:

I haven’t noticed any issues. What problems do you have with white point and clipping?

Click to expand...

I'm the author of one of those threads. I purchased Filmomat and used it to invert raw 16bit/channel data produced by my Coolscan ED via Vuescan. Filmomat did an extremely poor job at that, with highlight clipping being only one of its issues.

Perhaps it's not fully compatible with Vuescan RAW files, in spite of what the manual says.

Those of us with a Coolscan ED scanner IME are much better served by Nikonscan, Colorperfect or even vanilla Vuescan inversions.

Their sfuff looks very good.

albireo said:

Those of us with a Coolscan ED scanner IME are much better served by Nikonscan, Colorperfect or even vanilla Vuescan inversions.

Click to expand...

- this has been my experience as documented on Photrio.

Tom Kershaw said:

- this has been my experience as documented on Photrio.

Click to expand...

Affinity Photo does not handle DNG’s from Plustek Opticfilm well. There are probably (evidently) poor matches.

cerber0s said:

Nor does it include enough information in the resulting tiff to do it afterwards.

Click to expand...

I have to revise this statement. The problem, as it turned out, is not with the tiff files from Smartconvert, it’s with Affinity Photo. AP just seems unable to bring any information out of the seemingly clipped highlights in the files. Bringing them into Darktable proved to be a completely different story.

I haven't really liked the results I've gotten from SmartConvert. The tiff files from it don't seem to have the same amount of leeway as files converted using NLP. They also sharpen poorly. I love the way the app works, but I can't use it because of the lower quality output.

cobb said:

I haven't really liked the results I've gotten from SmartConvert. The tiff files from it don't seem to have the same amount of leeway as files converted using NLP. They also sharpen poorly. I love the way the app works, but I can't use it because of the lower quality output.

Click to expand...

Interesting! I had problems with the tiff files in Affinity, but the files seem to have plenty of information in them when I edit them in Darkrable.

I can’t compare them to NLP as I no longer use Adobe products, but I can convert a raw file in Darktable for comparison.