Secrets of K10D (Part 3 of 3): The IPU

trog100 · 02-02-2007, 10:03 AM

i dont have a K10 either jeffkrol.. i also have problems knowing what the correct amount of "sharpening" is..

life is made a little easier for me thow.. i own a k100 which together with the k110 seems to be the only pentax dslr which hasnt been knocked for its jpeg abilities..

i shoot jpegs mostly.. i tend to produce lots of pictures as opposed to the carefully produced single method.. the whole raw proccess as it stands at the moment is just too slow and cluncky for me..

having my camera produced jpegs come out as good as i think they can be everything else taken into considerarion is important to me..

soo my message to mr pentax is .. better (as soft or as sharp as i want them) camera produced jpegs please.. so so ones are not good enough.. he he

one day i might own a k10..

just my views thow..

trog

Last edited by trog100; 02-02-2007 at 10:08 AM.

Brew1brew · 02-02-2007, 10:13 AM

This has been discussed and tested on DPR. One of the members did a test and found that in normal mode the jpg's are soft, and bumping sharpness didn't help. But if you set it to bright mode then the jpg's are very sharp.

Chris Miller's test results can be seen here

SharpnessTest.jpg photo - Chris Miller photos at pbase.com

Les

robert · 02-02-2007, 10:31 AM

Your right about not making a difference unless you view at 100% but a lot of my photos need to be heavily cropped and it will show....Bob

trog100 · 02-02-2007, 10:39 AM

i have tried to find out what "bright mode" does exactly on my k100.. i think it just boosts the contrast.. saturation and sharpness compared to natural mode.. so it figures biright mode will produce seemingly sharper looking images if the k10 behaves in the same way..

i find that my k100 in bright mode with a minus 1 on the saturation is similar to natural with a plus 1 on the sharpness.. the k100 has a tendency to blow the reds.. the k10 dosnt do this as badly..

everything connected with picture taking seems to be a trade off.. he he

trog

ps.. u must be a prime man bob.. another trade off.. he he

Simon · 02-02-2007, 10:39 AM

Thanks Brew1brew. For those who, like RiceHigh expect their jpegs to come out of the camera fully formed, the answer is in "Bright", and it was there all along (see the section headed "Make a wise choice of finishing touch to create a desired image"). The advice found there for "Natural" applies more to me.

trog100 · 02-02-2007, 12:01 PM

come on simon "fully formed" isnt the right description.. i could equally say "not fuzzy blobs"..

i wrote more but for fear of sounding too much like R/H i deleted it.. he he

trog

ps.. i dont think the k10 produces "fuzzy blobs".. but from what i see some of the earlier pentax offerings did..

Simon · 02-02-2007, 12:24 PM

Originally posted by trog100

come on simon "fully formed" isnt the right description.. i could equally say "not fuzzy blobs"... i dont think the k10 produces "fuzzy blobs".. but from what i see some of the earlier pentax offerings did..

Sorry, but that was my attempt at pleasing everybody. Jpegs often need some tweaking to get the most out of them, though they aren't "half formed" in the way raw files are. I would rather apply sharpening after this tweaking (if at all). If you want to get what some (not necessarily me) might consider unaturally sharp results straight out of the camera, that's what "Bright" is for. Not everyone thinks the characteristically digital supersharp "look" is a good thing.

I tend not to sharpen many images, and not much on the others. Although I have previously only had a digicam I have always had it set to do as little sharpening as possible in camera - it works best for me that way

Simon

trog100 · 02-02-2007, 01:10 PM

just recently i have presented some of my images on a photo critique forum simon.. more to compare my own likes and dislikes with those of others.. i am more lost and confused now than when i started as to what exactly makes a good photograph..

i go for natural.. they say it needs more "pop".. they think i am a bit of a purist.. i think in pursuit of art they have gone too far.. he he

one think i like about my k100 is the lack of digital noise.. i can tweak my jpegs if i feel the need far more than i can tweak the images from my panasonic FZ20.. sharpness is never a problem with the FZ20 but the images it produces fall apart when any attempt to alter them afterwards is made..

the bottom line here is.. a decent artifact and noise free jpeg is worth its weight in gold.. to me at least it is.. it means i dont have to to put up with the extra work shooting raw brings..

a good noise and artifact free jpeg is "almost" as workable as raw data..

trog

Simon · 02-02-2007, 04:41 PM

Originally posted by trog100

the bottom line here is.. a decent artifact and noise free jpeg is worth its weight in gold.. to me at least it is.. it means i dont have to to put up with the extra work shooting raw brings..

a good noise and artifact free jpeg is "almost" as workable as raw data..

Oh dear, we seem to be agreeing again.

Simon

Xgear · 02-02-2007, 06:26 PM

Definition of JPEG format:
is Joint Photographic Experts Group file format. A compression algorithm for condensing the size of image files. JPEGs are helpful in allowing access to full screen image files on-line because they require less storage and are therefore quicker to download into a web page.

Definition of TIFF format:
is Tagged Image File Format, typically used for saving files to be used in print applications.

Definition RAW Image Format:
RAW image format allows the camera to store uncompressed or enhanced pictures.

Xgear · 02-03-2007, 06:38 AM

RAW is enhanced with extra data. DNG and PEF are RAW format what the diff you know wat i say.

RiceHigh · 02-03-2007, 08:53 AM

Originally posted by Xgear

RAW is enhanced with extra data. DNG and PEF are RAW format what the diff you know wat i say.

The K10D PEF RAW file format is compressed, but Pentax have never told us about that if the compression is lossy or lossless. Actually, Pentax have never mentioned about the PEF is compressed. In the user manual and for what the camera shows, it is the same file size as the DNG. But practically, the PEF is of smaller sizes (than the DNG).

So, everything remains a mystery. And no typical K10 users know about the actual image quality difference between the PEF and the DNG.

jeffkrol · 02-03-2007, 03:04 PM

Best analysis of the bright vs natural +2 sharpening comparison.
THANK YOU Bart..................
Bart's Blog

RiceHigh · 02-04-2007, 07:01 AM

:-))))))))

But some Pentaxians (or maybe mainly those fanboys) just can't accept or understand the contradictory thing that: A Pentax lover who bashes Pentax, or, A Pentax basher who loves Pentax! ;-D LOL!

Originally posted by trog100

i think R/H loves pentax.. he has as much "brand loyalty" as any fanboy.. but..

he hates the self imposed blindness which brand loyalty seems to bring about..

he thinks this is bad for pentax and is working against the odds to point out what he thinks could/should be improved which will ultimately be good for pentax and make his favoured brand a better brand..

i think his heart is in the right place.. he he

trog

ps.. but in the popularity stakes he aint gonna win any prizes..

RiceHigh · 02-04-2007, 07:25 AM

Originally posted by jeffkrol

Best analysis of the bright vs natural +2 sharpening comparison.
THANK YOU Bart..................
Bart's Blog

Very interesting experiment and analysis. Thanks for the info.

However, and maybe unfortunately, what is said by Bart in his conclusion is exactly what Phil Askey says in his review but *not* what Roland Mabo always says (bright mode will resolve the "edge sharpness" issue).

So, now everything are agreeing so far, this explains also why Imatest shows low LW/PH resolution figures with the K10D for the other tests I have seen.

Actually, to do Fourier transform to see the frequency response is interesting. But Imatest measurement on a set MTF (typically at 50%) is the most direct way to get and intrepret the results. That's what Photodo did carry out extensive tests in this similar way (despite they fixed the test chart spatial frequency first and then measure the MTF instead, and to plot the charts).

Doing the test in "physical" domain would be a more practical approach than in "frequency" domain, I think, just because there is actually not much processing in the frequency domain to be done, I bet (unlike audio signal processing, etc.)

Phil Askey do use eyes to judge where the resolution have arrived, but this will not be wrong neither, as everything is indeed so easy to be judged on the shot chart. And, he posts every resolution chart test shots for us to verify, too.

02-02-2007, 10:13 AM	#17
Brew1brew Senior Member Join Date: Sep 2006 Location: Plano Texas USA Posts: 121	JPG sharpness not an issue This has been discussed and tested on DPR. One of the members did a test and found that in normal mode the jpg's are soft, and bumping sharpness didn't help. But if you set it to bright mode then the jpg's are very sharp. Chris Miller's test results can be seen here SharpnessTest.jpg photo - Chris Miller photos at pbase.com Les

02-02-2007, 10:39 AM	#20
Simon Veteran Member Join Date: Jan 2007 Location: Bristol, England Photos: Gallery Posts: 314	Thanks Brew1brew. For those who, like RiceHigh expect their jpegs to come out of the camera fully formed, the answer is in "Bright", and it was there all along (see the section headed "Make a wise choice of finishing touch to create a desired image"). The advice found there for "Natural" applies more to me. Last edited by Simon; 02-02-2007 at 10:51 AM. Reason: correct typo

02-02-2007, 06:26 PM	#25
Xgear Guest	jpeg for snap shots Definition of JPEG format: is Joint Photographic Experts Group file format. A compression algorithm for condensing the size of image files. JPEGs are helpful in allowing access to full screen image files on-line because they require less storage and are therefore quicker to download into a web page. Definition of TIFF format: is Tagged Image File Format, typically used for saving files to be used in print applications. Definition RAW Image Format: RAW image format allows the camera to store uncompressed or enhanced pictures.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
For Sale - Sold: Pentax K10D for part (US/CAN)	dvdman45	Sold Items	2	08-17-2010 08:05 AM
6 Pro Photographers Share Their Most Guarded Digital Secrets	lukevega	Photographic Technique	33	11-10-2009 12:52 PM
You tube - Professional Photography Secrets	Peacekeeper	General Talk	2	08-31-2008 11:00 PM
Secrets of K10D (Part 2 of 3) - The Bridge: A to D Convertor	RiceHigh	Pentax DSLR Discussion	8	12-09-2006 07:14 AM
The Secrets of K10D - Part 1 of 3: The CCD imager	RiceHigh	Pentax DSLR Discussion	1	12-05-2006 04:35 PM

02-02-2007, 10:03 AM	#16
trog100 Veteran Member Join Date: Dec 2006 Location: UK Posts: 642	i dont have a K10 either jeffkrol.. i also have problems knowing what the correct amount of "sharpening" is.. life is made a little easier for me thow.. i own a k100 which together with the k110 seems to be the only pentax dslr which hasnt been knocked for its jpeg abilities.. i shoot jpegs mostly.. i tend to produce lots of pictures as opposed to the carefully produced single method.. the whole raw proccess as it stands at the moment is just too slow and cluncky for me.. having my camera produced jpegs come out as good as i think they can be everything else taken into considerarion is important to me.. soo my message to mr pentax is .. better (as soft or as sharp as i want them) camera produced jpegs please.. so so ones are not good enough.. he he one day i might own a k10.. just my views thow.. trog Last edited by trog100; 02-02-2007 at 10:08 AM.

02-02-2007, 10:31 AM	#18
robert Veteran Member Join Date: Jan 2007 Location: ohio Posts: 336	Your right about not making a difference unless you view at 100% but a lot of my photos need to be heavily cropped and it will show....Bob

02-02-2007, 10:39 AM	#19
trog100 Veteran Member Join Date: Dec 2006 Location: UK Posts: 642	i have tried to find out what "bright mode" does exactly on my k100.. i think it just boosts the contrast.. saturation and sharpness compared to natural mode.. so it figures biright mode will produce seemingly sharper looking images if the k10 behaves in the same way.. i find that my k100 in bright mode with a minus 1 on the saturation is similar to natural with a plus 1 on the sharpness.. the k100 has a tendency to blow the reds.. the k10 dosnt do this as badly.. everything connected with picture taking seems to be a trade off.. he he trog ps.. u must be a prime man bob.. another trade off.. he he

02-02-2007, 12:01 PM	#21
trog100 Veteran Member Join Date: Dec 2006 Location: UK Posts: 642	come on simon "fully formed" isnt the right description.. i could equally say "not fuzzy blobs".. i wrote more but for fear of sounding too much like R/H i deleted it.. he he trog ps.. i dont think the k10 produces "fuzzy blobs".. but from what i see some of the earlier pentax offerings did..

02-02-2007, 12:24 PM	#22
Simon Veteran Member Join Date: Jan 2007 Location: Bristol, England Photos: Gallery Posts: 314	Originally posted by trog100 come on simon "fully formed" isnt the right description.. i could equally say "not fuzzy blobs"... i dont think the k10 produces "fuzzy blobs".. but from what i see some of the earlier pentax offerings did.. Sorry, but that was my attempt at pleasing everybody. Jpegs often need some tweaking to get the most out of them, though they aren't "half formed" in the way raw files are. I would rather apply sharpening after this tweaking (if at all). If you want to get what some (not necessarily me) might consider unaturally sharp results straight out of the camera, that's what "Bright" is for. Not everyone thinks the characteristically digital supersharp "look" is a good thing. I tend not to sharpen many images, and not much on the others. Although I have previously only had a digicam I have always had it set to do as little sharpening as possible in camera - it works best for me that way Simon

02-02-2007, 01:10 PM	#23
trog100 Veteran Member Join Date: Dec 2006 Location: UK Posts: 642	just recently i have presented some of my images on a photo critique forum simon.. more to compare my own likes and dislikes with those of others.. i am more lost and confused now than when i started as to what exactly makes a good photograph.. i go for natural.. they say it needs more "pop".. they think i am a bit of a purist.. i think in pursuit of art they have gone too far.. he he one think i like about my k100 is the lack of digital noise.. i can tweak my jpegs if i feel the need far more than i can tweak the images from my panasonic FZ20.. sharpness is never a problem with the FZ20 but the images it produces fall apart when any attempt to alter them afterwards is made.. the bottom line here is.. a decent artifact and noise free jpeg is worth its weight in gold.. to me at least it is.. it means i dont have to to put up with the extra work shooting raw brings.. a good noise and artifact free jpeg is "almost" as workable as raw data.. trog

02-02-2007, 04:41 PM	#24
Simon Veteran Member Join Date: Jan 2007 Location: Bristol, England Photos: Gallery Posts: 314	Originally posted by trog100 the bottom line here is.. a decent artifact and noise free jpeg is worth its weight in gold.. to me at least it is.. it means i dont have to to put up with the extra work shooting raw brings.. a good noise and artifact free jpeg is "almost" as workable as raw data.. Oh dear, we seem to be agreeing again. Simon

02-03-2007, 06:38 AM	#26
Xgear Guest	RAW is enhanced with extra data. DNG and PEF are RAW format what the diff you know wat i say.

02-03-2007, 08:53 AM	#27
RiceHigh Veteran Member Join Date: Nov 2006 Location: Hong Kong Posts: 1,934 Original Poster	Originally posted by Xgear RAW is enhanced with extra data. DNG and PEF are RAW format what the diff you know wat i say. The K10D PEF RAW file format is compressed, but Pentax have never told us about that if the compression is lossy or lossless. Actually, Pentax have never mentioned about the PEF is compressed. In the user manual and for what the camera shows, it is the same file size as the DNG. But practically, the PEF is of smaller sizes (than the DNG). So, everything remains a mystery. And no typical K10 users know about the actual image quality difference between the PEF and the DNG.

02-03-2007, 03:04 PM	#28
jeffkrol Veteran Member Join Date: Sep 2006 Location: Wisconsin USA Photos: Gallery Posts: 8,434	Best analysis of the bright vs natural +2 sharpening comparison. THANK YOU Bart.................. Bart's Blog

02-04-2007, 07:01 AM	#29
RiceHigh Veteran Member Join Date: Nov 2006 Location: Hong Kong Posts: 1,934 Original Poster	:-)))))))) But some Pentaxians (or maybe mainly those fanboys) just can't accept or understand the contradictory thing that: A Pentax lover who bashes Pentax, or, A Pentax basher who loves Pentax! ;-D LOL! Originally posted by trog100 i think R/H loves pentax.. he has as much "brand loyalty" as any fanboy.. but.. he hates the self imposed blindness which brand loyalty seems to bring about.. he thinks this is bad for pentax and is working against the odds to point out what he thinks could/should be improved which will ultimately be good for pentax and make his favoured brand a better brand.. i think his heart is in the right place.. he he trog ps.. but in the popularity stakes he aint gonna win any prizes..

02-04-2007, 07:25 AM	#30
RiceHigh Veteran Member Join Date: Nov 2006 Location: Hong Kong Posts: 1,934 Original Poster	Originally posted by jeffkrol Best analysis of the bright vs natural +2 sharpening comparison. THANK YOU Bart.................. Bart's Blog Very interesting experiment and analysis. Thanks for the info. However, and maybe unfortunately, what is said by Bart in his conclusion is exactly what Phil Askey says in his review but not what Roland Mabo always says (bright mode will resolve the "edge sharpness" issue). So, now everything are agreeing so far, this explains also why Imatest shows low LW/PH resolution figures with the K10D for the other tests I have seen. Actually, to do Fourier transform to see the frequency response is interesting. But Imatest measurement on a set MTF (typically at 50%) is the most direct way to get and intrepret the results. That's what Photodo did carry out extensive tests in this similar way (despite they fixed the test chart spatial frequency first and then measure the MTF instead, and to plot the charts). Doing the test in "physical" domain would be a more practical approach than in "frequency" domain, I think, just because there is actually not much processing in the frequency domain to be done, I bet (unlike audio signal processing, etc.) Phil Askey do use eyes to judge where the resolution have arrived, but this will not be wrong neither, as everything is indeed so easy to be judged on the shot chart. And, he posts every resolution chart test shots for us to verify, too.