TIFF vs. RAW confusion

Getting to like my E-1 (bought used in January but actually new) more every day... including the colour rendering and lack of CA from my only lens, the kit 14-45. It is usually my walkabout camera of choice rather than a Nikon D200, F2, F3 or Leica M3.

Not having any software for the E-1 I’m doing all my image work in Photoshop CS (I have to import to Mac using Nikon’s Picture Project Transfer software which came with the D200... and always notice how the E-1’s Firewire transfer is soooo much faster than Nikon’s USB connection.)

However, I’m confused with my workflow because of previous professional (or current mal-) practices. In my photo magazine publishing and editing days everything I scanned was saved as a 275 or 300dpi CMYK tiff file for the off-set litho printer which used a 135dpi screen for my magazines. It was/is normal practice to save files at twice the final print resolution... I suppose many people know this but I’m a bit of a Luddite and have to remind myself of the basics quite often. (Sheesh... I remember when full page files used to take about half an hour to open on the fast Mac of the day... which probably put me off technology there and then!)

I’ve just noticed in the E-1 handbook that the TIFF “record mode” file size is 14.4mb whereas the RAW is 10.2mb from the 5mb sensor. I mainly shoot TIFFs because that’s the file designation I used during a decade of publishing... and open them in PS where the file size is around 57mbs... so I downsample the files using “bicubic smoother” or “bicubic sharpener” (can’t see any difference!) to 50mb as that is the size generally required by stock agencies. I only use the JPEG option for my web/blog pages.

Do other folk use the TIFF file option? And if for a stock library, then how are the results shaping up? Having always sharpened my CYMK files for off-set printing use I’m now finding it very difficult NOT to sharpen saved images for stock use because that’s what agencies requirements are.

TIA
--
Ed Buziak
http://blogs.salon.com/0004217/

TIFF is basically the uncompressed JPEG. It is created from the in-camera settings, like a JPEG file.

Raw is just that, it is the raw image data recorded by the sensor. It allows maximum flexibility in post processing. You can adjust exposure (within reason), amend white balance and set saturation and contrast. Raw files are not affected by in camera settings such as contrast, sharpness, saturation etc.

My own workflow with the E1 is to transfer the .ORF raw files to the computer, convert to adobe DNG format (it uses lossless compression, so instead of 10.2MB files you get 3.2MB files). The process the DNG raw files using Adobe Camera Raw in photoshop CS2.

After getting the Raw file right, I export to CS2 as 16bit sRGB files and adjust levels, curves etc, and any other techniques, such as blending two images to get a properly exposed foreground and sky. Finally I apply noise reduction and sharpening as needed, before saving the file as a 16bit TIFF.

I then convert to 8bit TIFF and apply my border action, then save the files as 8bit JPEGS for publishing to the web.

That probably doesn't fully answer your question, but I hope it helps a little bit at least.
--
Cheers, Neil
http://dodger.photoblog.com/
http://www.flickr.com/photos/dodger421/

very few people shoot TIFF. The only reason I have heard for shooting TIFF is if you want too publish the shot without the need for post processing; as some professionals do at events. TIFF takes longer to write; requires more memory and isn't as flexible when you are post processing.

Regards
Doug

--
See my best images @ http://dougjen.net/alamy
And many many more @ http://dougjen.net

The E-1 specs tell us the E-1 tiff images are 8bit/channel rgb files, while the raw images are 12 bit/pixel. ( see http://www.olympus-esystem.com/dea/products/e1/spec/index.html ) So in principle information is lost by the in-camera conversion to tiff.

There many different workflows out there, and what you describe for magazine work based on tiff certainly makes sense. Most here talk about using either jpeg or raw as their 'out-of-camera' preferred format, but then most here are not doing magazine work.

My own preference is to use Raw exclusively for out-of-camera. The tools for working with raw have evolved enormously, to the point where (IMO) raw workflow gives better results with less effort than jpeg. There are other well informed opinions on this with different conclusions.

Upgrading from CS to CS2 would give you access to the newer version of Adobe Bridge, their version of a browser and workflow tool. You might consider that upgrade (and the soon to come CS3 upgrade). Lightspeed is also just now released in version 1.0.

You could then edit in raw, then produce tiff (by batch scripts or by hand) for delivery to your magazine or agencies.

By the way, you might also take a look at the DAMBook, an remarkabel book describing 'Digital Asset Management'. It sure changed the way I manage my images.

Jeff

Ed Buziak said:

Getting to like my E-1 (bought used in January but actually new)
more every day... including the colour rendering and lack of CA
from my only lens, the kit 14-45. It is usually my walkabout camera
of choice rather than a Nikon D200, F2, F3 or Leica M3.

Not having any software for the E-1 I’m doing all my image work in
Photoshop CS (I have to import to Mac using Nikon’s Picture Project
Transfer software which came with the D200... and always notice how
the E-1’s Firewire transfer is soooo much faster than Nikon’s USB
connection.)

However, I’m confused with my workflow because of previous
professional (or current mal-) practices. In my photo magazine
publishing and editing days everything I scanned was saved as a 275
or 300dpi CMYK tiff file for the off-set litho printer which used a
135dpi screen for my magazines. It was/is normal practice to save
files at twice the final print resolution... I suppose many people
know this but I’m a bit of a Luddite and have to remind myself of
the basics quite often. (Sheesh... I remember when full page files
used to take about half an hour to open on the fast Mac of the
day... which probably put me off technology there and then!)

I’ve just noticed in the E-1 handbook that the TIFF “record mode”
file size is 14.4mb whereas the RAW is 10.2mb from the 5mb sensor.
I mainly shoot TIFFs because that’s the file designation I used
during a decade of publishing... and open them in PS where the file
size is around 57mbs... so I downsample the files using “bicubic
smoother” or “bicubic sharpener” (can’t see any difference!) to
50mb as that is the size generally required by stock agencies. I
only use the JPEG option for my web/blog pages.

Do other folk use the TIFF file option? And if for a stock library,
then how are the results shaping up? Having always sharpened my
CYMK files for off-set printing use I’m now finding it very
difficult NOT to sharpen saved images for stock use because that’s
what agencies requirements are.

TIA
--
Ed Buziak
http://blogs.salon.com/0004217/

Click to expand...

--
Jeff

Thanks folks for your clear answers...

And apologies on the confusing numbers and questions in my original post... I must have been very tired this morning and not thinking straight because I’ve rechecked my Olympus E-1 TIFF files size on opening and they are 14.1mg in 8-bit mode. The approx. 57mb file size I quoted previously - and was puzzled by - is in fact from my Nikon D200 NEF (Raw) files opening in 16-bit mode in PS/CS.
--
Ed Buziak
http://blogs.salon.com/0004217/

I'm a graphic designer, but would never shoot TIF format. The highest quality JPG settings are pretty much lossless in the real world, and save a lot of space. The only time I use TIF is for intermediate editing stages from RAW or JPG to avoid any cumulative losses, and then finally saving for print as a CMYK TIF at the required DPI (without resampling).

Workflows for print are changing though, some people maintaining an RGB workflow until the final ripping, and using high quality JPG rather than TIF.

If you are downsampling your TIFs, you will lose far more than saving as high quality JPG.

Bruce
--
http://www.flickr.com/photos/bruce-clarke/

you will need to use some interpolation. When processing for Alamy I double the size of my E-1 images with Photoshop's Bicubic smoother interpolation. This is slightly bigger than 50MB but using a whole number seems to give better results. I develop from RAW and save as TIFF for Alamy.

Regards
Doug

--
See my best images @ http://dougjen.net/alamy
And many many more @ http://dougjen.net

My process (not sure I can call it a workflow yet) is similar to Neil's process. Although, if I read it correctly, he transfers the .orf file to the computer and then converts to dng. I use a card reader and convert from the CF card to dng at that stage. In other words, the .orf file is never saved on the computer. I check to make sure the dng files transferred before clearing the .orf files from the CF card. Not sure if there are any disadvantages to that.
--
http://www.swimswithsharks.com

Ed Buziak said:

Getting to like my E-1 (bought used in January but actually new)
more every day... including the colour rendering and lack of CA
from my only lens, the kit 14-45. It is usually my walkabout camera
of choice rather than a Nikon D200, F2, F3 or Leica M3.

Click to expand...

Not to get off topic but why the E1 over the d200. I am seriously looking at picking up a d200. Would like some user feedback from someone who has used both.

Hehe, I do that too. I must have been tired and confused when I posted that first.

Should have read along the lines of:

Convert .ORF to .DNG on the card, then transfer to PC (or rather external HDD).

I think I needed that nap I just had (and the coffee I've got now!

I've never noticed a problem with the card after putting it back in the E1, but then I always make sure to format the card in camera straight after putting it back in...
--
Cheers, Neil
http://dodger.photoblog.com/
http://www.flickr.com/photos/dodger421/

I like to do a quick culling of shots before the dng converstion. So the steps are (1) "Import from Camera" with a script to rename, add basic metadata, put in a directory on the laptop. Then (2) rank & cull shots which deletes a 1/3 or so of the shots. Then (3) do a batch DNG conversion whenever its convenient to leave the computer to sit alone and crunch on the images.

dodger421 said:

Hehe, I do that too. I must have been tired and confused when I
posted that first.

Should have read along the lines of:

Convert .ORF to .DNG on the card, then transfer to PC (or rather
external HDD).

I think I needed that nap I just had (and the coffee I've got now!

I've never noticed a problem with the card after putting it back in
the E1, but then I always make sure to format the card in camera
straight after putting it back in...
--
Cheers, Neil
http://dodger.photoblog.com/
http://www.flickr.com/photos/dodger421/

Click to expand...

--
Jeff

Ed Buziak said:

I’ve just noticed in the E-1 handbook that the TIFF “record mode”
file size is 14.4mb whereas the RAW is 10.2mb from the 5mb sensor.
I mainly shoot TIFFs because that’s the file designation I used
during a decade of publishing... and open them in PS where the file
size is around 57mbs... so I downsample the files using “bicubic
smoother” or “bicubic sharpener” (can’t see any difference!) to
50mb as that is the size generally required by stock agencies. I
only use the JPEG option for my web/blog pages.

Ed Buziak

Click to expand...

Olympus RAW files carry more information than in-camera TIFF files, even though they take less space to store. The reason for this is, that each pixel in an ORF file is represented by a single 12-bit number, describing the luminosity at the photosite between 0 and 4095. Each photosite on your CCD sensor is covered by either a red, a green, or a blue filter (a Bayer filter). The RAW converter that you use (whether in-camera or on your Mac) knows which pixels represent which color and interpolates to a normal image. (Click here and the above will make more sense: http://en.wikipedia.org/wiki/Bayer_filter )

An 8-bit TIFF file, by contrast, stores each pixel as three 8-bit numbers, which represent the luminosity on each of the red, blue, and green channels for that pixel between 0 and 255. Thus, the TIFF file has significantly less dynamic range than the 12-bit ORF file.

It takes 12 bits (which rounds to 2 bytes) to store a pixel's information in an ORF file, but it takes 8*3=24 bits (which is 3 bytes) to store a pixel's information in an 8-bit TIFF file. That is why an uncompressed TIFF takes about 50% more disk space to store.

As to your downsampling, the number that you see in photoshop that tells you the file size is 57MB does not actually mean that the file size is 57MB! It just means that that particular file, when opened in PS, is using 57MB of your RAM. If you were to resave it as an uncompressed TIFF, you'd see the size go back down to 14ish MB.

For instance, I just opened a 600 kB jpeg (1916 x 1437) in my image editor (The GIMP). It showed a size at the bottom of 21MB. Just for fun, I duplicated the layer, and the number at the bottom then showed 42MB. If I resaved it as a JPEG at the same compression, it would still be a 600 kB file on my hard drive.

When a stock photo company says they want a 50 MB file, they mean they want one that takes up 50 MB of disk space, not 50 MB of RAM when they open it in photoshop. A 50MB tiff is roughly a 16 megapixel image (my E-500 takes 25MB TIFFs, so I just doubled the MP to arrive at that number). A 16 megapixel image in 4:3 format would have a resolution of 4616 by 3462—a resolution that lets them print an 11x14 at 300DPI. Thus, you can see that you should be UPsampling, not DOWNsampling.

As for my workflow, I shoot RAW (about 14 MB per image) and export as TIFFs out of Lightroom. I save them with LZW compression (which is totally lossless), so it cuts their file size to half of what it would be uncompressed (about 12 MB per image). If I want to put the image on the web, I use the GIMP to downsize it to 1600x1200 and save as a JPEG at 85% quality (about 200 kB per image). The total file sizes then add up to about what they'd be if I only shot TIFF and overwrote the original with my edits.

--
http://www.flickr.com/photos/hicsuget/

minor salazar said:

Ed Buziak said:

Getting to like my E-1 (bought used in January but actually new)
more every day... including the colour rendering and lack of CA
from my only lens, the kit 14-45. It is usually my walkabout camera
of choice rather than a Nikon D200, F2, F3 or Leica M3.

Click to expand...

Not to get off topic but why the E1 over the d200. I am seriously
looking at picking up a d200. Would like some user feedback from
someone who has used both.

Click to expand...

Hello Minor,

I've basically used Nikons (and several other marques at times) since the mid-60s and only picked up an E-1 this year after stumbling upon this forum (grrr!) However, the E-1 feels so nice in the hand, produces beautiful colours... AND... there's NO dust on the sensor, which, compared to my D200, could be the MAIN reason I change to Oly completely. If the news in the next couple of weeks is good concernng an interesting E-whatever with a professional specification, and the forecast 12-60 lens is available in April, and not April 2008... I will jump and pick up a 7-14 plus something longish.
--
Ed Buziak
http://blogs.salon.com/0004217/

Great plan. The 7-14 is reason alone to buy into the Oly system. Hopefully there will be even more reasons after PMA.

Ed Buziak said:

minor salazar said:

Ed Buziak said:

Getting to like my E-1 (bought used in January but actually new)
more every day... including the colour rendering and lack of CA
from my only lens, the kit 14-45. It is usually my walkabout camera
of choice rather than a Nikon D200, F2, F3 or Leica M3.

Click to expand...

Not to get off topic but why the E1 over the d200. I am seriously
looking at picking up a d200. Would like some user feedback from
someone who has used both.

Click to expand...

Hello Minor,

I've basically used Nikons (and several other marques at times)
since the mid-60s and only picked up an E-1 this year after
stumbling upon this forum (grrr!) However, the E-1 feels so nice in
the hand, produces beautiful colours... AND... there's NO dust on
the sensor, which, compared to my D200, could be the MAIN reason I
change to Oly completely. If the news in the next couple of weeks
is good concernng an interesting E-whatever with a professional
specification, and the forecast 12-60 lens is available in April,
and not April 2008... I will jump and pick up a 7-14 plus something
longish.
--
Ed Buziak
http://blogs.salon.com/0004217/

Click to expand...

--
Jeff