Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Yes.So you are saying Sony cRAW only contains 8-bit instead of 12-bit
data found in RAW.
File sizes would indicate that that's how it is. It's also the reason for it.Are you certain, and where might I read an explanation of this?
In most pictures, 12-bit is more than adequate, and 8-bit is sufficient in the end result. By eliminating the "unnecessary" levels, a lot of data can be cut from the files.I had always assumed the cRAW was just a lossless compression of the RAW data. Would you help me out here.
I waded through the dyxum thread you suggested. Seems cRAW is technically lossy but, in effect, preserves virtually all the RAW information. Thanks for helping me learn the details. Since the difference is negligible, I can understand why Sony did not wish to reveal that it is lossyYes.So you are saying Sony cRAW only contains 8-bit instead of 12-bit
data found in RAW.
For further reference: http://www.dyxum.com/dforum/topic22802.html
File sizes would indicate that that's how it is. It's also the reasonAre you certain, and where might I read an explanation of this?
for it.
In most pictures, 12-bit is more than adequate, and 8-bit isI had always assumed the cRAW was just a lossless compression of the RAW data. Would you help me out here.
sufficient in the end result. By eliminating the "unnecessary"
levels, a lot of data can be cut from the files.
Chris:There are two different things going on here. cRAW is a fileimplies that cRAW is compressed RAW. I assume lossless and 12-bit as
in RAW file.
Bert
containing less data because it describes every pixel in less bit
depth. The top and bottom value for each pixel will be the same as
for 12 bit image files, but there will be less steps in between the
top and bottom value.
No. From http://www.sony.co.uk/hub/learnandenjoy/block/7/subblock/1 I copy-pasted this:That isn't always the case. By and large compression in RAW is
non-destructive, but that isn't always the case and you can't assume
that. CRAW from Sony IS destructive compression they make that very
clear.
Actually it would have to be a change in levels that is more than 8 bits.. the compression in CRAW from all I have read has a 12 bit sample.. than encodes 8 bit deltas from the start value.. much less likely to have to toss or compromise date.. not impossible.. but unlikely.No. From http://www.sony.co.uk/hub/learnandenjoy/block/7/subblock/1 IThat isn't always the case. By and large compression in RAW is
non-destructive, but that isn't always the case and you can't assume
that. CRAW from Sony IS destructive compression they make that very
clear.
copy-pasted this:
Lossless compression is available on some models for cRAW files: this
reduces the file size (by around 60-70%) without discarding any data,
but processing times may be increased.
Lossy compression discards unnecessary data to create JPEG files: the
amount of data discarded or lost depends on the quality setting. The
more data lost, the smaller the file but there is some loss in
quality visible to critical examination.
So Sony does not make it clear that compressed raw is lossy or
destructive. Quite the contrary.
But of course it is in a number of cases, namely where the scene
contains more levels than can be reproduced with 8 bits per channel.
That's why they use 12 bits per channel in the first place, some even
14.
--Also, the file size is not reduced by 60-70 %, but to those values
(from around 18 MB to around 12 MB).
I thought that by "they" in "CRAW from Sony IS destructive compression they make that very clear" you meant Sony.Actually, the A900 book that I read says that the Craw for the A900
is lossy, though one is unlikely to see it. Which in my experience
except with the smallest amount of compression one can't say about
jpeg.
The book in question is the one from Gary L. Friedman.
...GaryW wrote:
Ah. Well, that's not the case on my A100 -- it uses "fine" either way as the max quality -- so I guess I'm safe! ;-)I think Walt was referring to the jpeg is fine as opposed to
xfine when one shoots RAW+JPG
--Bert
Here's a quote from that other thread:Yes.So you are saying Sony cRAW only contains 8-bit instead of 12-bit
data found in RAW.
For further reference: http://www.dyxum.com/dforum/topic22802.html
File sizes would indicate that that's how it is. It's also the reasonAre you certain, and where might I read an explanation of this?
for it.
In most pictures, 12-bit is more than adequate, and 8-bit isI had always assumed the cRAW was just a lossless compression of the RAW data. Would you help me out here.
sufficient in the end result. By eliminating the "unnecessary"
levels, a lot of data can be cut from the files.
Well, the others have answered your Raw/Jpeg question, but the Kodak part of your question I find of interest.But Kodak, or any
other kiosk, won't take RAW files. So if I shoot RAW I would have to
convert them to JPEG anyway.
Depends on your JPG setting in the camera.Ken,
I'm trying to understand the final quality difference using an Alpha
700 in RAW then converted to a jpeg using Sony's IDC3 vs. Xfine jpeg.
I understand that for ultimate control and under challenging
lighting/exposure situations RAW wins without question. I also
realize that some RAW converters work better than others. Your
example in the dark dramatically proves those points. However, what
is the difference (for a detailed scenic) betwen RAW converted to a
jpeg using Sony's IDC3 vs. Xfine jpeg for an exposure that the jpeg
"nails" easily? In other words, the Xfine jpeg exposure is 'perfect",
the Xfine jpeg WB is "perfect" and so on. Does IDC3 produce a 12 bit
jpeg vs the in-camera 8 bit version? Will the raw converted file
contain more MB (detail) than the in-camera version?
Thank you,
f8f8f8
