MFDB and DNG

[brumbaer]

Do you see any functional difference between tags and input profiles when shooting RAWs? Seems to me like these two terms should be consolidated.
[a href=\"index.php?act=findpost&pid=87220\"][{POST_SNAPBACK}][/a]

The input profile tells the application how to convert the camera color information to a given color space. Very often this Input profile is stored as an ICC profile converting camera RGB values to XYZ or Lab. The profile data consists usually from one or more color lookup tables and a matrix, but there are differnt ways to handle this.

The term TAG is much broader.

An image  file has an internal structure.

Most raw formats are derivates from TIFF.

A TIFF file has a number of IFDs. An IFD is a kind of directory that holds information about an image. Usually there is an IFD for the raw Image and at least one IFD for thumbnail representations. Instead of making the information in each IFD self contained, the designers decided to have information in some IFDs which is valid for other IFDs as well.

To find a certain piece of information, each IFD has a number of TAGs. Each TAG has an ID and the corresponding information. There is a TAG for the width of the image (thumbnail or raw) and the corresponding information is the width in pixels. So if you need the width of an image , you look for a TAG with the corresponding id or name (i.e. "imagewidth").

There could be a TAG that holds an input profile in ICC profile format, but the same information could be spread over different TAGs. What TAGs exist in file is defined by the one who writes the file format specification. But as I wrote before many formats a based on TIFF so many TAG names are standardized.

So the term Input Profile describes a structure of data which has a well defined function. And usually this structure follows the blueprint of an ICC profile.

A TAG is just a piece of information with a meaning.

I hope the description was not too confused.

Regards
SH

[CliffSamys]

So the term Input Profile describes a structure of data which has a well defined function. And usually this structure follows the blueprint of an ICC profile.

A TAG is just a piece of information with a meaning.

I hope the description was not too confused.

Regards
SH
[a href=\"index.php?act=findpost&pid=87249\"][{POST_SNAPBACK}][/a]

Hardly confused. Quite well articulated actually.
But functionally, are they not really the same? Couldn't an input profile be considered a complex tag? I mean to me when faced with a pile of RAW files, I don't care if the default is +4 Y or a more tuned profile. If it isn't right (which it most certainly won't be) I'm changin it.
This is really a semantic debate. I do appreciate your detailed comparison though.

[James Russell]

How does the majority here use their raw development software, it could tell us something about what MFDB owners needs from their raw converters...?
[{POST_SNAPBACK}][/a]

I work for a lot of different clients that all have different needs and workflows and I can tell you there are no absolutes.

Early on in digital capture I would always produce a flat "negative" type of file, to build in layers for tonal values and color in photoshop.

that way I had complete control, but mostly did this because the software convertors were so simple in their ability to produce anything but a canned profile or a slightly moveable wb setting.

I think those days are becoming old think, especially after working with lightroom.

Lightroom allows me to change densities in shadows, highlights and midtones, effect split toning, adjust each color channel for density, saturation, and hue, even goes as far as to add a fill light funciton and burn or dodge the edges of a frame.

Prior to this the raw convertor made the negative, photoshop made the "transparency", but now with lightroom I work at getting as close as possible to the final transparency.

I've compared Phase and Leaf files in lightroom and it is the only software that allows me to produce almost identical looks from the two seperate manufacturers.

It really is a great equalizer and I hope becomes a great stand alone piece of software.

I have absolutley no interest in Aperture.  It doesn't work with medium format files and I have this feeling that apple is motivated to sell hardware and software so every upgrade to aperture will probably require a new computer to reach full functionality.

That is a merry go round that is never ending.

I would think that Adobe would have a higher agenda, or at least a smart enough agenda to make light room fast and fucntional in as many computers as possible for  many files as possible.

I hope adobe's agenda is to take it further, into the tehtered mode caastagory, because whether any of us like it or not, the standard for many commercial projects is tethering to a large monitor.

If I was Hasselblad and Leaf I would have the same hope.

JR
[a href=\"http://www.russellrutherford.com/]http://www.russellrutherford.com/[/url]

[godtfred]

I think those days are becoming old think, especially after working with lightroom.
[a href=\"index.php?act=findpost&pid=87301\"][{POST_SNAPBACK}][/a]

I agree on this and see my own workflow is moving towards preparing more and more in the raw converter (i use bridge and ACR now.) Yet the images always end up in photoshop for some small detail that only PS can handle, and have actions to take care of particular tasks so its not completely there.

I still miss good file handling on the part of the RAW converters... Lightroom seems to promise a solution to this.

[abiggs]

This has been an enlightening discussion, to say the least. I took delivery of a borrowed H2 / CFH39 system a month ago, and took it out on safari with me in Africa. Now I have a ton of proprietary raw images that I have to use Flexcolor with, which is frustrating to say the least. I am going to convert all images to DNG format, using Flexcolor, and will do some comparisons to see if I am losing any noticable quality in the process.

Looks like my day is shot. Grrrrr.

[godtfred]

Forget about the web galleries as they are lame anyway.
[a href=\"index.php?act=findpost&pid=86858\"][{POST_SNAPBACK}][/a]

I agree that the web galleries are lame, yet they are eminently useful as "contact sheets" for the less discriminating customer that resides far away from the studio. I sometimes use this function (sometimes, or even less is the operative word    ) when I need to show some relative all of the pictures I took at his or her wedding  (   )

No lightroom does not make sidecar files and I assume because it uses a different processing engine that pscs 1 and 2, (I'm just guessing).
[a href=\"index.php?act=findpost&pid=86880\"][{POST_SNAPBACK}][/a]

There is a setting in the lightroom preferences to make it create xmp sidecar files, see the attatched .pdf. I have not tested this, but at least it states that they are xmp...

Leaf and hasselblad desperatly need lightroom to come out of the box working, because their software's are just too far behind C-1 in stability and batch functions.
[a href=\"index.php?act=findpost&pid=86880\"][{POST_SNAPBACK}][/a]

I disagree with this (have I said this somewhere else on the forum?) FlexColor is really good at batching, but does have a steeper learning curve. It is also very stable, and the rare occasions it has crashed on me, it never hurts/corrupts the files it is working on. I used C1 during my testing of the P25/20 (loaners) for about a month, on both canon and phase files, and having used Flexcolor for a year now, my impression is that they do the same, in totally different ways. And I'm not sure I would go the C1 option today. (But I would welcome P-backs build quality and fan less operation on my HB back    )

Regardless of the platform, regardless of how pretty any of us think a certain file is, in the end we must have fast stable software for an easier workflow.  Leaf and Hasselblad need to work on thier canned white balance settings so everything out of the camera is close.
[a href=\"index.php?act=findpost&pid=86880\"][{POST_SNAPBACK}][/a]

I totally agree with this, and have to add that we need future safe filetypes that are standardized, not neccesarily .dng's, but some format or other. And the white balance settings from my Hasselblad back is not great, but it exels in some areas like skintones (I find them better than Phase One's, but not in any other area like product, landscape, architecture, etc.)

-axel

[attachment=1851:attachment]

[alexjones]

So far I have not found any reason to use DNG since FlexColor does such a great job with the Hasselblad/Imacon files.  I am a Digital Tech in Pittsburgh and handle a wide variety of projects for a variety of clients.  So far I have not found a reason to use the DNG method.  LightRoom seems to be an interesting possibility for the future and I am looking forward to trying it.  I think its usability may be the greatest for the 35mm work, but we will see.

a

http://www.alexrjones.com

Digital Tech Pittsburgh

[marc gerritsen]

I usually work with flexcolor, because of how it works with colors but if there is a file that has some blown highlight, they are so much easier to recoup with an exported dng file. Still do this in CS2 bridge
Drawback of that is that dng files generate maybe 10% more noise and strange but true, a tiff generated from dng is 2 pixels smaller in the width than when it is generated straight from flexcolor.

I just checked again what I was writing to see if it is actually true and was amazed at the difference of highlight management with a dng file and am now wondering who of you are using an alternative to flexcolor.

thanks
Marc

www.marcgerritsen.com

[godtfred]

I usually work with flexcolor, because of how it works with colors but if there is a file that has some blown highlight, they are so much easier to recoup with an exported dng file. Still do this in CS2 bridge
Drawback of that is that dng files generate maybe 10% more noise and strange but true, a tiff generated from dng is 2 pixels smaller in the width than when it is generated straight from flexcolor.

I just checked again what I was writing to see if it is actually true and was amazed at the difference of highlight management with a dng file and am now wondering who of you are using an alternative to flexcolor.
[a href=\"index.php?act=findpost&pid=101080\"][{POST_SNAPBACK}][/a]

I did not know about the two pixels, but the noise and highlights is a know issue/benefit with exported .dng's from flexcolor. I dont think it is just from the dng conversion though, for me ACR has better highlight recovery than any other raw converter I have tested. Phil Askey over at DPReview has even integrated the highlight recovery of ACR into his test parameters for raw files to judge the dynamic range of cameras.

Its weird about the noise though, goes to show that a lot goes on in software, and maybe the rumored iso increase on HB backs has something in it. I wonder if it will make cleaner files out of the old files as well, or if it will only apply to new images captured?

-axel

[Wim van Velzen]

It could be that I just haven´t worked enough with Flexcolour, but most of the time I just export my files as DNG to be viewed in Bridge. I like the new ACR convertor in the CS3 beta a lot.

But I hope to make some prints of files worked on in FC and in ACR. See what it does to colours, highlights and noise.

[marc gerritsen]

Hoi Wim
Good idea
I would be very interested to hear the result of that.
No CS3 here yet, but I will try myself with CS2.
bedankt
Marc

www.marcgerritsen.com