A2 RAW to Standard jpg differences?

Receiving a 1gb card (Kingston Elite Pro & its very slow!) from Santa has prompted me to see how much difference there is between shots taken in RAW thru to a Standard jpeg. I have posted five comparison shots here http://www.grovestreet.com/jsp/picview.jsp?album=56220 make sure you click on each picture twice to see at 100% view.

This little test threw up the expected superior quality of RAW, however I have found it difficult to see much difference between the others. I could have been a bit more scientific and put the cam on a tripod and used manual focus, but I am reasonably satisfied that all are well focused and aimed at the centre of the bar code. In further shots, I seemed to be consistently finding the standard jpeg was giving as good a quality as the other jpeg/tif settings? This is pixel peeping I guess but I find it interesting as I can have 433 standard jpegs on my card versus 255 Fine, 128 Extra fine jpeg, 43 tiff, 85 RAW or 63 RAW + jpeg.

Hope you may find this of interest?

Stewart C

Stewart Cocker said:

Receiving a 1gb card (Kingston Elite Pro & its very slow!) from
Santa has prompted me to see how much difference there is between
shots taken in RAW thru to a Standard jpeg. I have posted five
comparison shots here
http://www.grovestreet.com/jsp/picview.jsp?album=56220 make sure
you click on each picture twice to see at 100% view.

This little test threw up the expected superior quality of RAW,
however I have found it difficult to see much difference between
the others. I could have been a bit more scientific and put the cam
on a tripod and used manual focus, but I am reasonably satisfied
that all are well focused and aimed at the centre of the bar code.
In further shots, I seemed to be consistently finding the standard
jpeg was giving as good a quality as the other jpeg/tif settings?
This is pixel peeping I guess but I find it interesting as I can
have 433 standard jpegs on my card versus 255 Fine, 128 Extra fine
jpeg, 43 tiff, 85 RAW or 63 RAW + jpeg.

Hope you may find this of interest?

Stewart C

Click to expand...

--
AussieToo

FWIW - on my screen (Apple Studio 17") the TIFF image is clearest, the Standard is definitely inferior but there's not much to choose between the rest. I judged by the text on the packaging, not the bar code. V interesting! And helpful! Sometimes I think we can get carried away with the advice to shoot everything in RAW. This may all be very well for those at the professional edge, but for others of us who don't want or need to PP everything in Photoshop, the high standard JPGs are fine. If the picture is well composed, the most that may be needed is a little touch of brightness or contrast available in a basic archiving program like Apple's iPhoto. I have had shots published from SuperFine images since all but the highest quality colour magazines still require only 75-100 dpi. I did shoot everything in one asignment in TIFF on my 7Hi, but TIFF images can't easily be read by many publishers, so they ended up converted into JPGs anyway!

If I may be pardoned for saying this - this is a classic 'discovery' by new users. It is true that for standard average shots that are well exposed & that do not require further processing that Jpegs compare very well to processed RAW shots. They are superior in one sense ,in that due to compression, they take up less disc space.

Often Jpegs look better because processing itself is an art that does require some practice. RAW does, however, ultimately give more latitude to the photographer. The ability to fine tune white balance being the most important.

I also find that the in-camera conversion does introduce changes into Jpegs that are quite difficult to modify. Greens ,for instance, are often too vivid although that can come down to a matter of individual taste.

Anyone who has followed the development of the Dimage series from the D7 thru to the A200 will have noticed that each successive generation has bumped up colour to give more impact which most people want and I've no doubt that the A200 meets more peoples expectations than the D7 did.

Anyone who searches for more quality & control will eventually end up with shooting most frames in RAW & its a lot easier to contemplate that nowadays due to the relative low cost of storage cards.

Keith-C

Stew,

thanks for doing the test, but on my monitor, the results are illogical -- I opened all of them in separate windows at 100% so that I could go back and forth and compare -- either the shots were mislabled, or there was camera shake because the Fine jpg is the worst of the bunch, the Standard jpg is better. I was using the small greek lettering under the flag to do the comparison. Listed below is what looks best to me with no. 1 being the best:
1) Extra Fine jpg
2) Standard jpg
3) Fine jpg
4) RAW
5) TIFF

Now, let me say, that the RAW has the best color saturation of all of the pics, but I was judging on "overall" value, and the jpg's are sharper images in that I can see more of the individual characters of the greek words. I still think that the Fine jpg being less clear than the Standard is a mistake or anomaly.

Cheers

Stewart Cocker said:

Receiving a 1gb card (Kingston Elite Pro & its very slow!) from
Santa has prompted me to see how much difference there is between
shots taken in RAW thru to a Standard jpeg. I have posted five
comparison shots here
http://www.grovestreet.com/jsp/picview.jsp?album=56220 make sure
you click on each picture twice to see at 100% view.

This little test threw up the expected superior quality of RAW,
however I have found it difficult to see much difference between
the others. I could have been a bit more scientific and put the cam
on a tripod and used manual focus, but I am reasonably satisfied
that all are well focused and aimed at the centre of the bar code.
In further shots, I seemed to be consistently finding the standard
jpeg was giving as good a quality as the other jpeg/tif settings?
This is pixel peeping I guess but I find it interesting as I can
have 433 standard jpegs on my card versus 255 Fine, 128 Extra fine
jpeg, 43 tiff, 85 RAW or 63 RAW + jpeg.

Hope you may find this of interest?

Stewart C

Click to expand...

--
DiMage A1 & Nikon 6006

You are seeing what I see and having used autofocus and no tripod it could well be that the fine shot is not representitive. However, what you have described was the key reason for making my post. Over a range of shots I can either see very little difference in all the jpeg settings or the Standard jpeg's were if anyhting slightly sharper with slightly less noise certainly in comparison with the fine setting. I guess the way to really test the issue is to print them. From what I have seen so far I reckon the difference could be marginal?

Re the RAW shot, I only added a very slight amount of contrast and I think it is generally the case that RAW needs post processing. The smoothness of the RAW pics is more akin to DSLR results. Which leads me to another thought, which is why can the seemingly significantly better picture quality obtained with RAW and the 8m sensor not be translated better when processing into jpegs. I am not really bothered by the noisy 8m sensor debate, but I would be interested to hear thoughts on this. Is it a software issue?

Stewart C

JimmyJet said:

thanks for doing the test, but on my monitor, the results are
illogical -- I opened all of them in separate windows at 100% so
that I could go back and forth and compare -- either the shots were
mislabled, or there was camera shake because the Fine jpg is the
worst of the bunch, the Standard jpg is better. I was using the
small greek lettering under the flag to do the comparison. Listed
below is what looks best to me with no. 1 being the best:
1) Extra Fine jpg
2) Standard jpg
3) Fine jpg
4) RAW
5) TIFF

Now, let me say, that the RAW has the best color saturation of all
of the pics, but I was judging on "overall" value, and the jpg's
are sharper images in that I can see more of the individual
characters of the greek words. I still think that the Fine jpg
being less clear than the Standard is a mistake or anomaly.

Cheers

Stewart Cocker said:

Receiving a 1gb card (Kingston Elite Pro & its very slow!) from
Santa has prompted me to see how much difference there is between
shots taken in RAW thru to a Standard jpeg. I have posted five
comparison shots here
http://www.grovestreet.com/jsp/picview.jsp?album=56220 make sure
you click on each picture twice to see at 100% view.

This little test threw up the expected superior quality of RAW,
however I have found it difficult to see much difference between
the others. I could have been a bit more scientific and put the cam
on a tripod and used manual focus, but I am reasonably satisfied
that all are well focused and aimed at the centre of the bar code.
In further shots, I seemed to be consistently finding the standard
jpeg was giving as good a quality as the other jpeg/tif settings?
This is pixel peeping I guess but I find it interesting as I can
have 433 standard jpegs on my card versus 255 Fine, 128 Extra fine
jpeg, 43 tiff, 85 RAW or 63 RAW + jpeg.

Hope you may find this of interest?

Stewart C

Click to expand...

--
DiMage A1 & Nikon 6006

Click to expand...

I think the messages in this thread suggest folks don't understand just how JPEG compresses files. JPEG does not reduce the number of pixels! What it does is tend to change a pixel's color by "leaking" from adjacent pixels. If adjacent pixels have the same color, information re the subject pixel needn't be included - it's constructed from the adjacent pixels. Thus reducing the number of bits to represent the image, or "compression".

At this site, I show how the same pixels of a random (lawn grass) scene are rendered in TIFF, SHQ JPEG and HQ (I think) JPEG.

http://www.pbase.com/russ/image/606198/medium

You can see no pixels are "lost", but with high compression, some pixel colors are "contaminate" by the colors of adjacent pixels. Think how this is going to affect details like fine print! It probably won't have much effect, hence users' conclusions that JPEG has no degradation!

Russell

i think you won't see the diff between fine and extra fine until you start doing solid colors that gradiate from one shade to the next...

for example, a wall of solid color that has bright light fading to dark or a sunrise/set

thats the only place i notice it anyway....

jamey

In regard to the Raw image...well, it isn't one. A Raw file is not strictly speaking an image at all, only a grey-scale mosaic of values for each pixel. Every digital picture starts life as a Raw file - some are converted to a picture format in the camera, some are converted afterward in a computer. Presumably the image which you have labelled "Raw" was converted to a jpg in order to produce a color image. This being the case, how can we compare it to other jpgs at varying levels of compression unless you tell us at what compression it was saved ? Moreover, it is important to indicate what Raw converter was used. Some don't sharpen at all , some sharpen according to what the camera setting was unless you override it, some offer sophisticated sharpening, Also, the choice of conversion algorithm can greatly influence sharpness, detail rendition and noise. And contrast also affects the perception of sharpness. Bottom line, the subject is somewhat complex.

By the way, that tif that you uploaded to your web host was converted by them to a jpg in order to be displayed, browsers don't sopport tif, and they for sure didn't tell you at what compression level.
Elie

Thanks for your reply, my post was more aimed at the seemingly very small difference in the quality between the various jpeg settings on the A2.

Re the Raw, I take your point but converted RAW files from the A2 seem to me, to have always given me better quality jpegs, especially with regard to noise. For this particular conversion I used Dimage viewer and added a touch of contrast in Graphic Converter before saving as a Jpeg at the highest quality setting. Taking this approach does not seem to me to affect quality significantly and certainly gives better results than jpegs straight from the camera.

Re the tiff you are spot on and I sheepishly had to convert it to a jpeg after being reminded that I could not upload it! However, it still exhibits the same characteristics of the original tif. Considering the enormous size of tif files on the A2 I find it interesting that there seems to be so little difference in quality?

Stewart C

EliezerDinur said:

In regard to the Raw image...well, it isn't one. A Raw file is not
strictly speaking an image at all, only a grey-scale mosaic of
values for each pixel. Every digital picture starts life as a Raw
file - some are converted to a picture format in the camera, some
are converted afterward in a computer. Presumably the image which
you have labelled "Raw" was converted to a jpg in order to produce
a color image. This being the case, how can we compare it to other
jpgs at varying levels of compression unless you tell us at what
compression it was saved ? Moreover, it is important to indicate
what Raw converter was used. Some don't sharpen at all , some
sharpen according to what the camera setting was unless you
override it, some offer sophisticated sharpening, Also, the choice
of conversion algorithm can greatly influence sharpness, detail
rendition and noise. And contrast also affects the perception of
sharpness. Bottom line, the subject is somewhat complex.
By the way, that tif that you uploaded to your web host was
converted by them to a jpg in order to be displayed, browsers don't
sopport tif, and they for sure didn't tell you at what compression
level.
Elie

Click to expand...

Stewart Cocker said:

Receiving a 1gb card (Kingston Elite Pro & its very slow!) from
Santa

Click to expand...

I had the same card, and it used 43 sec. to store an raw image.

I switched it for an Dane-elec Xs card, that card stores an raw image in 11 sec.

--
-
--

Yours faithfully

Henrik Høy

Here's the image set I referred to:

Russ Yost said:

I think the messages in this thread suggest folks don't understand
just how JPEG compresses files. JPEG does not reduce the number of
pixels! What it does is tend to change a pixel's color by
"leaking" from adjacent pixels. If adjacent pixels have the same
color, information re the subject pixel needn't be included - it's
constructed from the adjacent pixels. Thus reducing the number of
bits to represent the image, or "compression".

At this site, I show how the same pixels of a random (lawn grass)
scene are rendered in TIFF, SHQ JPEG and HQ (I think) JPEG.

http://www.pbase.com/russ/image/606198/medium

You can see no pixels are "lost", but with high compression, some
pixel colors are "contaminate" by the colors of adjacent pixels.
Think how this is going to affect details like fine print! It
probably won't have much effect, hence users' conclusions that JPEG
has no degradation!

Russell

Click to expand...

--
Russell

Amazing difference!! I will keep an eye out for that make of card for when I am allowed to consign the Kingston card to a back up role.

Thanks
Stewart C

Henrik Høy said:

Stewart Cocker said:

Receiving a 1gb card (Kingston Elite Pro & its very slow!) from
Santa

Click to expand...

I had the same card, and it used 43 sec. to store an raw image.

I switched it for an Dane-elec Xs card, that card stores an raw
image in 11 sec.

--
-
--

Yours faithfully

Henrik Høy

Click to expand...