PEF quality not the same as DNG... What the?? and jpeg :-(

davemdsn · 10-23-2007, 11:37 AM

There is a very scientific way to test whether there are any differences in PEF and DNG files of the same scene, as long as you don't budge the camera even the slightest amount. Open both files in Photoshop, stack one on top of the other in the same document and change the blend mode of the top image layer to "difference". If the image goes black they are identical, if they are different there will be a visible image - how visible and how much is visable depends on how different they are.

I used this test several years ago to make sure that the DNG converter didn't do any damage to the PEF files from my *istD. It did not. This test will likely show some difference when using two different image captures, no matter how careful you are, but will still work to see if those differences are image quality issues or just CCD movement between captures.

RAW image data is exactly that - RAW. No processing happens in camera, no matter what file format it is. The differences are in the post processing. I am very impressed with the sharpness and IQ of the K10D PEF files (which I convert to DNG on import to LR).

Last edited by davemdsn; 10-23-2007 at 11:38 AM. Reason: spelling

davemdsn · 10-23-2007, 11:39 AM

I guess I missed the other thread. Ooops.

m8o · 10-23-2007, 11:48 AM

yet please don't delete the previous post. It was MOST informative!

photog · 10-24-2007, 09:58 PM

Since acquiring my K-10D I have made just under 8000 exposures and I love the camera. At the outset I tested both PEF and DNG as I did not find the statement in the manual that both made the number of exposures to 4GB when viewing the files with any program on my computer showed the PEF to be about 35% smaller! I believe in personal testing.
I find the claim that there is ANY difference in quality quite unacceptable. My tests using a 4GB card and PsCs3 with Bridge and ACR 4.2 did show a markedly better image with the DNG images using the default settings! However, after processing both I found that there was NO difference.
The conclusion that I reached was that ACR 4.0 read more detail from the DNG metadata than from PEF. The increased capacity to store long shoots won me. I must say that I have not repeated the tests with the most recent ACR plugin, 4.2 which is much improved
Tests with the bundled software are not worth the trouble.
BTW, not all Aussie photographers believe everything they hear, especially when it does not fit the observed facts.

photog · 10-24-2007, 10:01 PM

Originally posted by Mandi

ChrisA, it's nothing to do with "Aussie solidarity" - in fact I didn't notice that the original poster was from Australia until you pointed it out. If that was the reason for MY posting, I wouldn't think it something to "admire," I would just think it stupid!!! Why agree with someone on a subject like this just because they happen to live in the same part of the world?? (As it happens, I wasn't even born here, and for all I know neither was 123K10D or SupremeMoFo - we're a nation of migrants! pentaxbling, I notice, is from Virginia and last time I looked that wasn't part of Australia. ;-) )

No, it has to do with the fact that someone has put some time and effort into doing a test that (until Carpents) none of the rest of us had done so - shouldn't we just be a teensy bit grateful for that instead of casting aspertions just because they've just joined the forum?? I personally don't know enough about either RAW format to be able to choose on any other criterion, so why not take his/her word for it and change, and see if I can also see some sort of difference and THEN decide for myself?? Yes, I agree that seeing the test images would be useful - and that's what I said, as did others. Perhaps Carpents could put his/her images up as well, and then we'd all have something to actually debate about. And when you've done your test, ditto...?

Sorry, this was originally posted in the wrong topic.
Since acquiring my K-10D I have made just under 8000 exposures and I love the camera. At the outset I tested both PEF and DNG as I did not find the statement in the manual that both made the number of exposures to 4GB when viewing the files with any program on my computer showed the PEF to be about 35% smaller! I believe in personal testing.
I find the claim that there is ANY difference in quality quite unacceptable. My tests using a 4GB card and PsCs3 with Bridge and ACR 4.2 did show a markedly better image with the DNG images using the default settings! However, after processing both I found that there was NO difference.
The conclusion that I reached was that ACR 4.0 read more detail from the DNG metadata than from PEF. The increased capacity to store long shoots won me. I must say that I have not repeated the tests with the most recent ACR plugin, 4.2 which is much improved
Tests with the bundled software are not worth the trouble.
BTW, not all Aussie photographers believe everything they hear, especially when it does not fit the observed facts.

ChrisA · 10-25-2007, 03:13 AM

Originally posted by photog

I must say that I have not repeated the tests with the most recent ACR plugin, 4.2 which is much improved

The test I did (results earlier this thread) was with ACR 4.2.

All default settings, no differences seen.

photog · 10-26-2007, 01:19 AM

Originally posted by ChrisA

The test I did (results earlier this thread) was with ACR 4.2.

All default settings, no differences seen.

I see from earlier post that you used the defaults in PsCs3 (ACR 4.2) I'm sure that you will have observed that Digital Photography reviews etc has posted photographic tests of the latest DSLR's in which he compared the K10D with the 200D and the EOS 40. (favourably re RAW images for a change). He found that although the default image had a poor dynamic range the adjusted image was in fact superior to the Nikon and the Canon.
I feel that it is safe to say that almost all users of ACR 4.2 use far more conservative settings than the default. The images I get are more like Capture One than the defaults of the JPGS.
However, we all have our own likes and dislikes and it would e a disaster if there only one raw converter on the market!
Ron

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
RAW + JPEG with JPEG on One Star quality	laissezfaire	Pentax DSLR Discussion	58	12-10-2008 02:42 PM
PEF or DNG?	beaumont	Digital Processing, Software, and Printing	14	09-15-2008 08:34 AM
K10D JPEG, PEF & DNG lens metadata is fully revealed ....	Richard Day	Pentax SLR Lens Discussion	1	08-07-2007 02:54 PM
PEF vs DNG	Fireball	Pentax DSLR Discussion	19	07-21-2007 08:01 PM
Pentax PEF vs DNG camera RAW Quality	benjikan	Pentax DSLR Discussion	3	03-22-2007 09:42 PM

10-23-2007, 11:37 AM	#46
davemdsn Inactive Account Join Date: Oct 2006 Location: Boise, Idaho Photos: Gallery \| Albums Posts: 1,413	There is a very scientific way to test whether there are any differences in PEF and DNG files of the same scene, as long as you don't budge the camera even the slightest amount. Open both files in Photoshop, stack one on top of the other in the same document and change the blend mode of the top image layer to "difference". If the image goes black they are identical, if they are different there will be a visible image - how visible and how much is visable depends on how different they are. I used this test several years ago to make sure that the DNG converter didn't do any damage to the PEF files from my istD. It did not. This test will likely show some difference when using two different image captures, no matter how careful you are, but will still work to see if those differences are image quality issues or just CCD movement between captures. RAW image data is exactly that - RAW. No processing happens in camera, no matter what file format it is. The differences are in the post processing. I am very impressed with the sharpness and IQ of the K10D PEF files (which I convert to DNG on import to LR). Last edited by davemdsn; 10-23-2007 at 11:38 AM. Reason: spelling*

10-23-2007, 11:39 AM	#47
davemdsn Inactive Account Join Date: Oct 2006 Location: Boise, Idaho Photos: Gallery \| Albums Posts: 1,413	I guess I missed the other thread. Ooops.

10-23-2007, 11:48 AM	#48
m8o Veteran Member Join Date: Jul 2007 Location: 40°-55'-44" N / 73°-24'-07" W [on LI] Photos: Gallery \| Albums Posts: 3,092	yet please don't delete the previous post. It was MOST informative!

10-24-2007, 09:58 PM	#49
photog Senior Member Join Date: Mar 2007 Location: Brisbane, Qld, Australia Posts: 165	Since acquiring my K-10D I have made just under 8000 exposures and I love the camera. At the outset I tested both PEF and DNG as I did not find the statement in the manual that both made the number of exposures to 4GB when viewing the files with any program on my computer showed the PEF to be about 35% smaller! I believe in personal testing. I find the claim that there is ANY difference in quality quite unacceptable. My tests using a 4GB card and PsCs3 with Bridge and ACR 4.2 did show a markedly better image with the DNG images using the default settings! However, after processing both I found that there was NO difference. The conclusion that I reached was that ACR 4.0 read more detail from the DNG metadata than from PEF. The increased capacity to store long shoots won me. I must say that I have not repeated the tests with the most recent ACR plugin, 4.2 which is much improved Tests with the bundled software are not worth the trouble. BTW, not all Aussie photographers believe everything they hear, especially when it does not fit the observed facts.

10-24-2007, 10:01 PM	#50
photog Senior Member Join Date: Mar 2007 Location: Brisbane, Qld, Australia Posts: 165	Originally posted by Mandi ChrisA, it's nothing to do with "Aussie solidarity" - in fact I didn't notice that the original poster was from Australia until you pointed it out. If that was the reason for MY posting, I wouldn't think it something to "admire," I would just think it stupid!!! Why agree with someone on a subject like this just because they happen to live in the same part of the world?? (As it happens, I wasn't even born here, and for all I know neither was 123K10D or SupremeMoFo - we're a nation of migrants! pentaxbling, I notice, is from Virginia and last time I looked that wasn't part of Australia. ;-) ) No, it has to do with the fact that someone has put some time and effort into doing a test that (until Carpents) none of the rest of us had done so - shouldn't we just be a teensy bit grateful for that instead of casting aspertions just because they've just joined the forum?? I personally don't know enough about either RAW format to be able to choose on any other criterion, so why not take his/her word for it and change, and see if I can also see some sort of difference and THEN decide for myself?? Yes, I agree that seeing the test images would be useful - and that's what I said, as did others. Perhaps Carpents could put his/her images up as well, and then we'd all have something to actually debate about. And when you've done your test, ditto...? Sorry, this was originally posted in the wrong topic. Since acquiring my K-10D I have made just under 8000 exposures and I love the camera. At the outset I tested both PEF and DNG as I did not find the statement in the manual that both made the number of exposures to 4GB when viewing the files with any program on my computer showed the PEF to be about 35% smaller! I believe in personal testing. I find the claim that there is ANY difference in quality quite unacceptable. My tests using a 4GB card and PsCs3 with Bridge and ACR 4.2 did show a markedly better image with the DNG images using the default settings! However, after processing both I found that there was NO difference. The conclusion that I reached was that ACR 4.0 read more detail from the DNG metadata than from PEF. The increased capacity to store long shoots won me. I must say that I have not repeated the tests with the most recent ACR plugin, 4.2 which is much improved Tests with the bundled software are not worth the trouble. BTW, not all Aussie photographers believe everything they hear, especially when it does not fit the observed facts.

10-25-2007, 03:13 AM	#51
ChrisA Veteran Member Join Date: Feb 2007 Location: London Posts: 393	Originally posted by photog I must say that I have not repeated the tests with the most recent ACR plugin, 4.2 which is much improved The test I did (results earlier this thread) was with ACR 4.2. All default settings, no differences seen.

10-26-2007, 01:19 AM	#52
photog Senior Member Join Date: Mar 2007 Location: Brisbane, Qld, Australia Posts: 165	Originally posted by ChrisA The test I did (results earlier this thread) was with ACR 4.2. All default settings, no differences seen. I see from earlier post that you used the defaults in PsCs3 (ACR 4.2) I'm sure that you will have observed that Digital Photography reviews etc has posted photographic tests of the latest DSLR's in which he compared the K10D with the 200D and the EOS 40. (favourably re RAW images for a change). He found that although the default image had a poor dynamic range the adjusted image was in fact superior to the Nikon and the Canon. I feel that it is safe to say that almost all users of ACR 4.2 use far more conservative settings than the default. The images I get are more like Capture One than the defaults of the JPGS. However, we all have our own likes and dislikes and it would e a disaster if there only one raw converter on the market! Ron