Can you explain why TIFF?

I'm aware of what it is. But as I move more and more into raw shooting, I'm

becoming more aware of people who shoot raw and then convert to tiff.

 

If you are such a person, can you tell me what this does that you wouldn't have

by storing the native raw image or storing a DNG image? And don't you lose the

'non destructive' nature of RAW by 'fixing' the development when you convert to

tiff? I can see the image quality value over jpg for a serious shooter, but at

this point, I don't see how the benefits of tiff could outweigh the benfits of raw.

 

I'm not trying to stir things up, just trying to make sure I don't miss

something important.

My practical application of this is that the RAW file remains forever unaltered. I don't convert to tiff as an alternative to RAW, but rather keep both an unaltered RAW file (with all the information available forever) and a tiff that can be manipulated in PS (or whatever). This gives the best of both worlds at the cost of a bit more disc space.

RAW files are comparable to negatives, which look different anytime somebody prints them. During a Tiff conversion you put them into shape and hone them later by postprocessing. I'd store RAWs and would convert finished TIFFs to JPGs when I'm sure I did aeverything right and need diskspace.

 

AFAIK you can't alter the contrast, WB ISO of a RAW and save this as RAW again, so TIFF is the best way to save all your effort. I know nothing about DNG yet.

<I>AFAIK you can't alter the contrast, WB ISO of a RAW and save this as RAW again, so TIFF is

the best way to save all your effort. I know nothing about DNG yet.</I><P>All of this and

more gets saved as an instruction set that resides with the "raw' file.

Glen,

 

In workflow, I make it a point to save a file as a tiff immediately after making adjustments

in raw. I can always go back to the raw CR2 (or whatever it is) and change those settings.

You can't really screw up a raw file, unless you delete it. I always save a native raw file on a

separate hard drive in case the computer or my beer-addled brain does just that. I haven't

taken a raw picture yet that can't benefit from a few nondestructive adjustment layers,

levels, curves, dodging, burning, etc.

 

The tiff eventually becomes a master file that you can purpose different sizes of prints and

web images from. If your PS skills progress beyond earlier versions of tiffs, which happens

to everybody, you can always go back to the raw conversion, and work them up again.

 

Hope that helps a little.

.

 

This is an eldless topic, and we'll revisit it with a new thread every day, eh?

 

One point, then I've GOT to make a cut-n-paste FAQ ...

 

Earlier: "... RAW files are comparable to negatives, which look different anytime somebody prints them ..."

 

No, try this:

 

RAW files are comparable to UNDEVELOPED negatives, which look different anytime somebody DEVELOPS them.

 

... and that's a GOOD thing, as it allows the photographer the greatest resource for reinterpretation of the original resource at any later time -- FAR better than developed-one-time negatives.

 

I like TIFs because ... oh, LATER!

 

Click! Love and hugs, Peter Blaise peterblaise@yahoo.com Konica Minolta Sony Photographer http://www.peterblaisephotography.com/

Since I began using Lightroom for my personal photograhy I don't save tifs, but I would (or a psd) if I did any selective editing in PS. Lightroom preserves its edits in a separate file.

 

Raw data is analogous to exposed but undeveloped film.

 

--

 

Don E

Here's my 2 cents on what I do:

 

RAW: Original 'bits'. Not sharpened, color corrected*, cropped to form, etc. Max detail.

 

TIF: Working image from Raw. Mass conversion (using Bibble), to provide sharpening, color correction, crop to my working standard size (4x5 multiplier, so 8x10, etc.) High res, uses lots of disk space. Where necessary, I custom edit the TIF to 'tweak' images as necessary. Try to minimize this. Save it as a replacement TIF.

 

JPEG: Final delivery form to customer.Smaller files. Cannot be edited without affecting quality (since it gets lossy recompressed).

 

If you're going to send a file to someone, you need to send it in a corrected, public form, so it's TIF, JPEG, BMP, or something. TIF has max detail but big files. JPEG uses lots of compression to make it smaller.

 

At least, that why I think/do, fwiw. I at all costs do not / will not change the NEF (RAW) form. That's my original...

 

pat

It would seem that people (at least many people) who DO use Tiff's from Raw aren't converting to Tiff and deleting the raw. Instead, they are saving BOTH!

 

This makes more sense to me than what I though was occurring. But I am still not sure what storing a Tiff alongside the raw really buys for an amateur shooter as opposed to another lossless format such as psd.

 

It makes a bit more sense to me for the pro. A pro may well want a highly marketable image 'frozen' in it's most marketable and highest quality state yet retain the capability to 'redevelop' the image in a totally different manner if market tastes change or a new market opens up that may require a different interpretation.

 

But a hack amateurlike myself, may well feel converting the raw file as needed is good enough. (and even might enjoy the serendipity of minor accidental differences)

TIFF and PSD files are uncompressed, have layers and can be 16 bits/channel. The principal alternative, JPEG, are compressed to varying degrees and limited to 8-bit and a single layer.

 

By using layers, particularly adjustment layers, all the masks and compatible adjustments can be saved and revisited upon reopening. Most images can be improved by post-processing, and TIFF files are the best way to save these changes - in other words, the master file.

 

TIFF files are usually converted to JPEG files for printing and distribution. JPEGs are much smaller, as a rule, and layered TIFF files are not compatible with as many programs as JPEGs.

 

With a D2x, RAW files are 19.6MB, TIFF files (single layer) are nearly 70MB and high quality JPEGs are 15MB or less. If you convert everything to TIFF, it takes 3 or 4 DVDs to back up a single, 4GB memory card. These days I'm inclined to do simple processing and save the results as JPEGs. I only do TIFFs for things that need the master's touch ;-) I save a few trees (or whatever) that way.

.

 

Opening a can of paint with a screwdriver is not in the screw driver's instruction manual ... but that's how we all open a can of paint, right?

 

Same here for RAW, TIF, PSD, BMP and JPG.

 

We're all sharing the definitions and design intentions of these files but NOT saying how we actually use them in our work flow and archive!

 

Earlier: "... It would seem that people (at least many people) who DO use Tiff's from Raw aren't converting to Tiff and deleting the raw. Instead, they are saving BOTH! This makes more sense to me than what I though was occurring ..."

 

See? Glen, we left some details out about our work flow and archive -- we NEVER delete anything, especially the original master RAW! So, to help explore our goals, let me ask you: how are you gonna reprint an image to look the same as the last print, at any time later in your life?

 

In the chemical darkroom days, people took notes on the back of a master print and saved it, notes saying the amount of exposure time and where to dodge and burn and so on. Photographers often made a "limited edition" print series because it was impossible to make an "UNlimited edition" series with the same appearance if we interrupted ourselves by a day, months or years, let alone hours. If we needed to replace a print or make an additional one, we had to do have exacting notes, and if too much time had passed, if we changed to fresh chemicals, if we changed the bulb in our enlarger, there was no way our subsequent print would match or original print.

 

Earlier: "... But a hack amateur like myself, may well feel converting the raw file as needed is good enough. (and even might enjoy the serendipity of minor accidental differences) ..."

 

Yep. Gotcha. But you'll notice on photo.net discussion forums, we're CONTROL freaks, and there is NO "serendipity of minor accidental differences" between prints in a series for us! So we HAVE to keep master files for each repurposing of the original RAW, and for me, that's often 2, 3 or more master TIFs or PSDs in addition to keeping (of course) the original RAW source / resource. I also keep copies of the distribution JPGs so I can resend them on demand. Anyone who makes photographic decisions based on the size of their hard drive is an economist, not a photographer!

 

And another thing, converting the RAW file is also a control freak area of concern since each conversion program has it's own flavor and varied controls. We have many threads on how each RAW converter behaves and how we can master their different features and benefits. I have no problem with the FREE RAW conversions from http://www.irfanview.com/ and http://picasa.google.com/ but I definitely also own my own MASTER set of professional RAW converters when I want complete repeatable control.

 

Click! Love and hugs, Peter Blaise peterblaise@yahoo.com Konica Minolta Sony Photographer http://www.peterblaisephotography.com/

While there are many adjustments that can be performed on a RAW file using ACR, many cannot. In particular, I use dodging and burning techniques (adjustment masks) in creating the master TIFF file, much as I have done in the darkroom. You can't do this in ACR, and shouldn't in a JPEG file (because 8 bits isn't enough, and compression artifacts accumulate). Another post-ACR technique is masking and compositing. For example, you mask open sky when sharpening, to prevent amplification of noise or grain.

 

This is simply the craft of photography, not some philosophical diversion.

While some operations can only be done on the original RAW files, many still can be done starting from 16-bit TIFFs. TIFF files can be read by almost any image editing program and don't require a special raw converter to be present. Yet they don't have compression artifacts unlike JPEGs.

 

The RAW file is really an anomaly caused by the Bayer pattern in use in most of today's digital sensors. It is not something you need for a Foveon-style sensor or film scan. TIFF is a standard image format, irrespective of source.

 

For my own purposes I convert the RAW file to a TIFF before opening it in PS and then I make a slight adjustment before printing. Processing RAW files is just slow to consider doing when just making more prints of the same image (IMO). I do appreciate the fact that the NEF format contains tags which describe the operations to be carried out to get to an RGB image. This makes doing all changes slow as all the operations are applied to the raw data simultaneously (or can be). However, the processing speed is IMO a major drawback of RAW. Yes, newer computers process them faster than older ones, but the camera sensors deliver progressively larger files as computers get faster, as well. So it will be slow for at least the next decade as far as I can see.

<ul><em>The RAW file is really an anomaly caused by the Bayer pattern in use in most of today's digital sensors. It is not something you need for a Foveon-style sensor or film scan. TIFF is a standard image format, irrespective of source.</em></ul>

 

<p>Well, RAW is the <em>only</em> format that existing Sigma Foveon sensor cameras save in. They can't save images as JPEG. Or TIFF. (The not-yet-released Sigma SD14 will be able to save JPEGs, though.)

 

<p>But I'm not sure what your point is. The significance to RAW isn't about Bayer vs. Faveon; it is that RAW is the data straight from the sensor, before any (well, almost any) processing has been done to it. Whereas, by the time an image becomes a TIFF, all kinds of processing transformations have been done to the image. If you want to do something to an image, you have your best chance if you start from the RAW file. Because from there on out, any processing destroys data.

 

<p>Data has been destroyed to make a TIFF file. This is the point of the RAW file--it is the only format that contains all the data that the sensor captured.

 

<p>As far as archiving goes, it is a good idea, IMO, to save both the RAW file and your working file(s). TIFF, JPEG, etc. As a practical matter, when you revisit an image, most of the time the TIFF (JPEG) that you made from the RAW file is what you want anyway. Unless you are revisiting it for the purpose of reworking it.

>But I'm not sure what your point is. The significance to RAW isn't >about Bayer vs. Faveon; it is that RAW is the data straight from >the sensor, before any (well, almost any) processing has been done >to it.

 

Yes it is precisely about Bayer vs. Foveon. A 10 MP Bayer sensor has 10 million pixels, 5 of which are green, 2.5 blue, and 2.5 red. This is not an RGB image by default because the different pixels are recorded at different spatial positions on the sensor. To generate a useable digital image, this needs to be interpolated to an RGB bitmap, which is what the TIFF contains. On the Foveon, AFAIK, they are recorded at the same position in the xy plane, at different depths. This means that no interpolation (=processing) is needed and the RAW file is in fact the RGB image which can be stored directly as a TIFF without any loss of information.

 

> Whereas, by the time an image becomes a TIFF, all kinds of >processing transformations have been done to the image. If you want >to do something to an image, you have your best chance if you start >from the RAW file. Because from there on out, any processing >destroys data.

 

TIFF images are a generic image format for RGB data. No processing is needed to store a Foveon image as a TIFF.

 

>Data has been destroyed to make a TIFF file. This is the point of >the RAW file--it is the only format that contains all the data that >the sensor captured.

 

On a Foveon sensor, if the camera saves the data as a 16-bit TIFF, no image data is lost. If the Sigmas don't do it that just means they probably can't get 16 good bits per color.

<ul><em>TIFF images are a generic image format for RGB data. No processing is needed to store a Foveon image as a TIFF.</em></ul>

 

<p>No. Substantial amounts of processing are needed to turn a Sigma/Foveon image into a TIFF (or JPEG). They don't use Bayer demosaicing, but the Foveon RAW files need a lot of processing to get the color correct.

 

<p>Don't you think if they could get TIFF quality right off the sensor, that the camera would do this already? And skip the RAW file? The reason why the Sigma SD9 and SD10 are RAW only is because of the complex amount of processing is needed to do this, and Sigma didn't have the resources to do this in hardware. (But they solved this for the upcoming SD14. It can create JPEGs in the camera.)

 

<p>Again, I don't know why you are trying to turn this into a Foveon vs. Bayer issue. The issue of the thread is about whether or not to archive RAW files. Not about the supposed advantages of Foveon sensor.