Olympus-OM
[Top] [All Lists]

Re: [OM] Interesting !

Subject: Re: [OM] Interesting !
From: Moose <olymoose@xxxxxxxxx>
Date: Wed, 25 Feb 2015 11:49:21 -0800
On 2/25/2015 6:40 AM, Chris Trask wrote:

...
      Nice, but with the 8MP E-500 I can record uncompressed TIF images of 
24.5M, which is as raw as you're going to get.   An interesting note here is 
that there is no TIF RAW format available in the E-510, but they do note that 
the ORF RAW format does NOT undergo any changes in white balance, sharpness, 
contrast, or colour.  Hmmmm...  The E-500 manual does not mention that.

I can't tell if you are being intentionally obtuse, or simply have no 
understanding of the technical issues.

A RAW file such as Oly's ORF format, consists of simply brightness values for each pixel. So if pixel x,y has a blue filter in front of it, the value in the file for that pixel records the brightness of the light that fell on it through a blue filter. Each pixel is thus a red, green or blue value. There are no red or green values for any blue pixel, and so on.

The RAW converter imputes for each pixel its missing color channel values. So for our blue pixel x,y, the red and green values are computed based on the adjacent and surrounding values from green and red filtered pixels. Differences in the algorithms that do this are the reason various RAW converters give different results.

A TIFF or JPEG file has all three color values for each pixel, created by in-camera RAW conversion. Above, you seem to be confusing size with nature and information content, "...24.5M, which is as raw as you're going to get." It is NOT a RAW file at all, merely larger than the potential RAW file you could have had because it has three times as many brightness values for each pixel. It appears to me that the only advantage of TIFFs over JPEGs for your cameras is that the JPEGs will have gone through a second layer of compression.

Worse, however, is the information lost in the conversion process. A 24 MB TIFF from an 8 MB sensor means that the data is 8 bit, 256 possible values for each color for each pixel. (3 colors x 1 byte x 8 MP = 24 MB*) (I don't remember if the three digit 4/3 E-thingies are 12 or 14 bit - doesn't much matter for this point.) The in-camera RAW conversion has already had to compress 4096 or 16384 steps of brightness data into 256. In the process, it may well make decisions about what to compromise that you would not.

Next, and I have not the patience, time or interest to check this out; for reasons I hope are clear here, above and below that you shouldn't be shooting to TIFF, it is possible/likely that the TIFFs include the various camera settings you mention above, "white balance, sharpness, contrast, or colour"

Finally (perhaps?), you are relying on a relatively ancient, low powered processor with old algorithms to process your images and lock in the results forever. If you read AG's recent posts, you will know that he finds that different RAW converters work better for different Oly cameras (and the semi Oly L1) AND that the latest version of Adobe ACR/LR give better results with his E-1 files than anything that came before.

I may have more to say about your take on filters and your cameras, if we ever get any clouds (let alone the rain we desperately need). I will say that your comments about using glass filters on digital are almost certainly wrong because you are misinformed:

"We proved that back in the B&W thread where the RAW photos came out exactly the same regardless of the Wratten filter that was used." From what you say above, you were not, in fact, using RAW files at all.

'It reduces or eliminates the opportunities for the camera to exercise its "opinions". If you find that your camera is "expressing opinions" in its output, it's because you have camera settings, esp. WB or scene modes, that are MEANT to affect the output, set wrong AND are using outputs that are affected by those settings. "... the ORF RAW format does NOT undergo any changes in white balance, sharpness, contrast, or colour. Hmmmm... The E-500 manual does not mention that. " Again, this is true (of the E-500, and of every 4/3 and µ4/3 camera ever made) - they do not express their opinions in their RAW files.

One further word. At least part of the reason you are having trouble with DR is that you are using old sensor systems with rather limited DR and high shadow noise, then letting the TIFF/JPEG conversions throw away much of that DR. This is FAR less of a problem with contemporary sensor systems, RAW files and correct technique. (Yes, with things like red/orange/yellow flowers, that means "underexposing" and pulling shadows and midtones up in RAW conversion and/or post.)

In a way, you would be better off with color neg. film. At least it has great highlight exposure latitude. By sticking with early, low end, Oly 4/3 camera bodies, you have selected a worst of both worlds position as to DR. Old tech, with film, or newer digital, are both better in that regard.

Tech Corner Moose

* TIFFs may also be compressed in the way JPEGs are, and may also be 16 bit, but I don't believe Oly did either with in-camera conversion.



Chris

When the going gets weird, the weird turn pro
      - Hunter S. Thompson


--
What if the Hokey Pokey *IS* what it's all about?
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/

<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz