NegFix8 workflow

broncc · Aug 5, 2015

I just got an Epson 4490 and decided to try something more advanced than the default Epson scanning software hoping to get better results. My current workflow is: using VueScan with all enhancements turned off and a gamma of 1, I scan a raw negative. Then with the help of NegFix8 I invert the image to a positive and set the colors. The problem is that I'm not very happy with the end product. I'm not sure whether I'm not doing something correct, the photo itself is bad or it needs post processing.. I'm using a pretty cheap Kodak ColorPlus 200 film. Maybe it just can't do any better?

The 1st photo is the negative, the 2nd is the positive converted with NegFix8, and the 3rd one is a scan with the auto settings in Epson Scan (just ICE was on). What do you think?

- Album on Imgur

I just want to be sure that what I'm doing with it is okay.. What I find confusing in the HowTo for NegFix8 is this:

The negatives are scanned as positives. Scanned frame must not contain any holder fragments (it means no pixels with value of 0) or script will fail. The scanned frame should, however, contains orange mask pixels (the area between frames, at least 20 pixels width).

Does this mean I have to scan them as negatives or slide film? And I'm not sure about what exactly is orange mask pixels.. Can you please explain?

RalphLambrecht · Aug 7, 2015

bronc said:
I just got an Epson 4490 and decided to try something more advanced than the default Epson scanning software hoping to get better results. My current workflow is: using VueScan with all enhancements turned off and a gamma of 1, I scan a raw negative. Then with the help of NegFix8 I invert the image to a positive and set the colors. The problem is that I'm not very happy with the end product. I'm not sure whether I'm not doing something correct, the photo itself is bad or it needs post processing.. I'm using a pretty cheap Kodak ColorPlus 200 film. Maybe it just can't do any better?

The 1st photo is the negative, the 2nd is the positive converted with NegFix8, and the 3rd one is a scan with the auto settings in Epson Scan (just ICE was on). What do you think?

- Album on Imgur

I just want to be sure that what I'm doing with it is okay.. What I find confusing in the HowTo for NegFix8 is this:

Does this mean I have to scan them as negatives or slide film? And I'm not sure about what exactly is orange mask pixels.. Can you please explain?

I've been very vocal in the past on this forum about my dislike of scanning negatives and have been critized heavily for itby (PhilS.) to name jut obne;Maybe the critics can speak up now and help you.I plan to try an experimrnt in which the negative is projected and then photographed with a DSLR;want to see if that has aby merrit.

lenny · Aug 8, 2015

ralphlambrecht said:
I've been very vocal in the past on this forum about my dislike of scanning negatives and have been critized heavily for itby (PhilS.) to name jut obne;Maybe the critics can speak up now and help you.I plan to try an experimrnt in which the negative is projected and then photographed with a DSLR;want to see if that has aby merrit.

For crying out loud, Ralph, what is wrong with you? Why don't you go somewhere where someone who knows what they're doing with a drum scanner and see what is possible? The issue with these systems is often the addition of another step thru a lens, and often a bad one, as in the case of Epson's plastic lenses. You are suggesting projecting a negative, thru another plastic lens? You haven't a friggin' clue. Go somewhere where you can see this done properly. It happens every day.... and you should shut up about scanning until you do. I know better than to get on a forum about different types of rocket fuel and start making comments. You don't know anything about the process and yet, somehow you feel like you should contribute.

Now, to the OP. I haven't used NegFix, I don't need it in my workflow. However, if they are asking for 20 pixels of the mask, they are likely talking about the rebate at the edge. The place at the edge of the image where there is no image, but there is still film that has the orange mask on it.

When I correct my negs I select an area inside the image. When dealing with certain types of film, there is a black edge around the image. If I include that in my selected area, it is going to give me a value that doesn't actually exist in the image. When the software analyzes it, it throws it off. It appears from your description that NegFix needs some of the clear edge, but doesn't want a holder, which would supply too dark of a value, thus throwing off what it thinks it sees.

Hope this is helpful...

RalphLambrecht · Aug 8, 2015

lenny said:
For crying out loud, Ralph, what is wrong with you? Why don't you go somewhere where someone who knows what they're doing with a drum scanner and see what is possible? The issue with these systems is often the addition of another step thru a lens, and often a bad one, as in the case of Epson's plastic lenses. You are suggesting projecting a negative, thru another plastic lens? You haven't a friggin' clue. Go somewhere where you can see this done properly. It happens every day.... and you should shut up about scanning until you do. I know better than to get on a forum about different types of rocket fuel and start making comments. You don't know anything about the process and yet, somehow you feel like you should contribute.

Now, to the OP. I haven't used NegFix, I don't need it in my workflow. However, if they are asking for 20 pixels of the mask, they are likely talking about the rebate at the edge. The place at the edge of the image where there is no image, but there is still film that has the orange mask on it.

When I correct my negs I select an area inside the image. When dealing with certain types of film, there is a black edge around the image. If I include that in my selected area, it is going to give me a value that doesn't actually exist in the image. When the software analyzes it, it throws it off. It appears from your description that NegFix needs some of the clear edge, but doesn't want a holder, which would supply too dark of a value, thus throwing off what it thinks it sees.

Hope this is helpful...

Lenny,therein no need to get so agressive,BTW,my enlarging lenses are not made of plastic. If yours are,I can see why you prefer negative scanning.Now, calm down and No, it was not helpful;Try againif you have something productive to add

OzJohn · Aug 8, 2015

Ralph, you're out of line this time. The OP posted a perfectly reasonable question that only someone with broad knowledge of scanning techniques and software could attempt to answer in a helpful way. I know a bit about scanning but more to the point I know what a really good scan looks like and where to get one if I need it and can't do it myself - I go to someone like Lenny. You seem to be suggesting that scanning does not have a place in modern photographic practice - if so a photograph of a projected image is most unlikely to be its replacement.

lenny · Aug 8, 2015

ralphlambrecht said:
Lenny,therein no need to get so agressive,BTW,my enlarging lenses are not made of plastic. If yours are,I can see why you prefer negative scanning.Now, calm down and No, it was not helpful;Try againif you have something productive to add

Ralph, Apparently there is a need. You keep beating this drum, people keep telling you the be quiet about something until you understand it, and you were also called a "dolt" (by someone else) just last week. You apparently are still not getting the picture.

It makes no sense whatsoever to project a negative, even using an enlarger lens, on a surface and attempt to photograph it. It's a patently ridiculous idea. One would be much better using a lightbox, or even illumination of some sort. There are a number of options, from inexpensive to very expensive.

My enlarger lenses were Companion and Rodenstock, thank you very much. I don't need them anymore, I am not doing darkroom printing, only inkjet and alt process, because I like those better. My scanner doesn't use a lens at all, it has a sensor called a photo multiplier tube. I run an 8,000 ppi Aztek Premier drum scanner, arguably the best scanner ever made. (There are a couple of others in its class.) Cost new is about 40K. So no, it isn't a piece of crap with plastic lenses.

I've made mistakes before posting in forums. I have misunderstood something and reported something which wasn't actually correct. One has to apologize, own up to it. You can't keep trying to be right or everyone will begin to dismiss you. I don't post unless I know what I am talking about, there are a large number of threads where you won't see my input...

My "hope this is helpful" was directed at the OP... The only advice I can give you is to wake up and smell the coffee... if enough people are complaining about your behavior you might consider modifying it.

RalphLambrecht · Aug 9, 2015

Lenny,I'm glad you know so much about scanning and that after spending $40k it works well for you.I get better results with digital cameras and will stick to that.You can stick to scanningbut,for $40k I 'd grather go with a digital Hasselblad set up,which would beat the pants off your scanned negatives for sure.It seems to me that you are the one who has to do some research and his homework before riding the high horse on DPUG.Have fun anda nice life.I'll see you in the 21st century when you get thereLenny

lenny · Aug 9, 2015

Ralph, I have tried to keep this on a civil level. I find this morning that I fail. I'm with MartinCrabtree now, you're a dolt, and further, a total idiot. You have no idea what you are talking about, you admit that you have failed to understand the basics of this field and yet you bluster along like you are some kind of expert. You should be tossed from the forum.

RalphLambrecht · Aug 9, 2015

lenny said:
Ralph, I have tried to keep this on a civil level. I find this morning that I fail. I'm with MartinCrabtree now, you're a dolt, and further, a total idiot. You have no idea what you are talking about, you admit that you have failed to understand the basics of this field and yet you bluster along like you are some kind of expert. You should be tossed from the forum.

If that is civil,I hate to see your true cgaracter or more likely the lack thereof.Why don't you show us your expert results and achievementsLenny?And yes, happen to be an expert in photographic matters but nobody knows it all, nor do I act like it and I also don't try to make up for a lack of knowledge by name-calling.That is not very convincing.Show us the goods or hush up.

RalphLambrecht · Aug 9, 2015

Oh now I see. You are using this thread to promote your wn business;very ethical, indeed.Shame on you

RalphLambrecht · Aug 9, 2015

ralphlambrecht said:
Oh now I see. You are using this thread to promote your wn business;very ethical, indeed.Shame on you

I wish I 'd known that scanning is a money-making business for you.That would have put many of your comments in the right context for me

lenny · Aug 9, 2015

Stating that I do this professionally doesn't mean I am trying to convince people of something that isn't true. I've not gotten one client from here, for all the time I've been on... and that's been fine with me...

RalphLambrecht · Aug 9, 2015

lenny said:
Stating that I do this professionally doesn't mean I am trying to convince people of something that isn't true. I've not gotten one client from here, for all the time I've been on... and that's been fine with me...

but it explains why you defend scanning so fiercely and gotten my case for challenging its quality.Your opinion was not objective if you run a scanning business.

pdeeh · Aug 10, 2015

It's not clear to me how it is helping answer the OP's question for Messrs. Lambrecht and Eiger to engage in this fierce and personal argument in the thread.

Perhaps better to take it offline to PMs or elsewhere and ask the mods to clean up the thread?

pdeeh · Aug 10, 2015

I wrote a short "primer" on negfix8 whichI posted in the "Hybrid" group at APUG.

The OP PM'd me there directing me to this thread and asking if I could offer any help, so I PM'd him back with some more details.

I'll reproduce my side of the PM exchange here, just in case it help someone else seeking info on negfix8.

It's really important to note that I am no scanning guru, and I scan only for indexing and websharing purposes.

So my approach is limited by that, and is not intended as a basis for producing the finest quality scans in a full hybrid process that ends up with an inkjet print. (in short, I am not Mr.Eiger )

me said:
I've read your post and had a look at your 3 example pictures.

The very first thing to say is that I am NOT an expert on negfix8 And I also very rarely use colour film at all, so when I wrote my "HowTo" it was really aimed at black and white shooters. I'm sorry if I didn't make that clear, so maybe I had better revise it.

I've kind of given up on negfix8 for the small amount of colour I do shoot and just use Silverfast in its own regular mode, because I've had variable results.

The guy who wrote it is probably the best possible person to advise you, but I will do my best. I have seen many of his colour photos for which he used negfix8 and they seem beautiful to me, but as I say I've had problems sometimes.

So far as I can tell, there are three things to comment on:

1. Is there a problem with Kodak ColorPlus 200 ?

It isn't the best colour film in the world, so it won't be as sharp, fine-grained, saturated or contrasty as something like Ektar. But it isn't the worst either, and I'm sure it's fine for general photography so long as you accept its limitations. I suppose it was designed as a "snapshot" film.

Also remember that film is not designed to be scanned. It is designed to be printed in a darkroom on paper. So the results of scanning film will always be a bit of a compromise - especially because of something called GRAIN ALIASING. I'll leave you to google that and read up on it!

2. Is there a problem with Vuescan plus negfix8 vs Epson scanning software?

Well it's really hard to say. I don't know what it is you don't like about the negfix8 picture vs the Epson picture!

Obviously the colour balance is a bit different, and there seems to be a bit less contrast, but both versions seem OK to me -- because of course I was not there when you took the picture, so I don't know which one is the more accurate version.

Also I don't know what the Epson software is doing as far as sharpening, curves and so on is concerned, and I don't know what editing (if any at all) you did to the negfix8 version afterwards either.

I think it's important to make sure that Vuescan is not applying ANYTHING to the scan and is just producing raw scans if you are using negfix8. I don't like Vuescan, I think the interface is ugly and confusing and it's really easy to leave some settings switched on when you think they are turned off. So be careful with that.

3. Orange mask pixels & negative vs positive

I think you are right, the way it is written is a bit confusing.

The latest version of the HowTo for negfix8 is here https://sites.google.com/site/negfix/howto

You scan them as NEGATIVES - don't scan them as slides/transparencies.

OK a bit of education (sorry if you know this already) -

When you look at a colour negative, all the unexposed areas, and some of the shadow areas in the actual frame, are coloured orange, and this is what is called the "mask". Obviously this is pretty different to normal black & white negative film.

The orange mask is there to help make the colours print correctly on darkroom colour printing paper. The details of how this works are too complicated to explain here and I'm not sure I properly understand how it actually works myself, but just hold onto that for a moment.

The trouble is, if you just scan a colour negative and then "invert" the colours (you can do this in PhotoShop or whatever), you get very weird results (try it!). This is because the orange mask interferes with how the colours are rendered, unless you do something in the software to allow for the orange mask, and allow the colours to render correctly.

In order to do this, negfix8 needs to "see" some of the orange mask without any other colours, and best way to do this is to include part of what is called the "rebate" - the unexposed areas of the film around the actual frame.

So here's a scan that includes a few pixels of rebate at the left hand side showing the orange mask :

20150807-Test1-Mask-001-2 by _loupe, on Flickr

I'll leave you to try and see what difference this makes (as Flickr uploading has just died on me otherwise I would post a couple of examples myself)

Also remember that the inverted scan from negfix8 is not "the finished article" - it will always need some adjustments, almost always some extra contrast and some exposure adjustments, sometimes some adjustments to white balance. This is perfectly normal.

Again, if you really need better details on how to scan colour film and process it in negfix8, I do urge you to contact the gentleman who wrote it https://www.flickr.com/people/jaz99/

RalphLambrecht · Aug 10, 2015

pdeeh said:
It's not clear to me how it is helping answer the OP's question for Messrs. Lambrecht and Eiger to engage in this fierce and personal argument in the thread.

Perhaps better to take it offline to PMs or elsewhere and ask the mods to clean up the thread?

I agree and hope to do better next time

MartinCrabtree · Aug 10, 2015

ralphlambrecht said:
I agree and hope to do better next time

In my personal experience that is not the first time you posted that very thing word for word. As yet the improvement remains elusive. Perhaps shutting the exit and opening the entrance would aid the process. If it is done with sarcasm simply remember this............sarcasm is the refuge of a man with nothing meaningful to say.

And welcome to my ignore list. Unfortunate because in the past your experience has been beneficial. That seems to be only historical in nature. Sad.

maxmars · Aug 12, 2015

Adding my two cents

After having tried a few things, currently I'm relying solely on Vuescan to produce a positive image out of my negatives, then I edit the final image to my heart's content.

The procedure is, in principle, somewhat similar to the one described here for negfix8:

- Enable "Restore colors" (filters tab). I could do the same with photo editing software but this puts the image in the right ballpark and I only need to adjust from there.

- feed vuescan a part of the roll where there is only the orange mask (e.g. end of the roll, beginning, space between two frames)
- do a preview
- select a small area containing only the orange mask
- click on "lock exposure" (input tab)
- preview again
- click on "lock film base color"

After that, the whole roll will scan fine without too many corrections (and potential differences between shots). I usually have low contrast, pastel look scanned images but that's fine, photo editing takes care of it when necessary.

MartinCrabtree · Aug 12, 2015

Sometimes that works for me,sometimes not. Don't understand what makes it work now and not later.

maxmars · Aug 12, 2015

Do you have an untouched example of an image that didn't come out as expected? Just to see what the software did.

NegFix8 workflow

Privacy & Transparency

Privacy & Transparency