K7 compared to K20/K10 CA & PF

knumbnutz · 10-03-2009, 03:21 AM

Hi All,

Just interested who has either FE10-17, Da12-24 or DA*16-50 (or any other lens that PF badly) and noticed the purple fringing on the K10 or K20 but now find that their K7 has cured the PF.

Do you guys find it is cured/ half cured ?

Share your experiences please .

Cheers Neil

ytterbium · 10-03-2009, 04:52 AM

This is not an observation or hard fact. Just some thoughts.

Some time ago i remember reading that PF is digital only, sensor related problem, and CA is lens one, much less pronounced on film. Or the other way around. Actually i think its almost the same thing, just PF is effect produced as a result of CAs (like you could get green/red aberrations if the lens was constructed such way), since it is pink it's called PF.

Now, for the difference between cameras, sensors and film, and why theoretically it could be cured. If you know a little about lens corrections, you will also know that they do not bend all light equally, but are more or less optimized to visible spectrum. Simple illustration borrowed from wikipedia shows how well each type of improved lens correction helps focusing all colours at the same point.

Grap for APO chromatic lens:

Graph for Super APO chromatic lens:

You can clearly see that even with Super APO lens the graph rapidly deflects when leaving visible spectrum. If you have some older lens with the red infra red focus correction mark, you can imagine how much out of focus invisible portions of light spectrum can be. I think it's even worse with UV.

Since the film has different type of colour layers (Yellow/Magenta/Cyan) than digital cameras (Red, Green, Blue) you will have different spectral responses and final rendering of those colours (like you get bluish tint from IR component on hot objects glowing red). If you've took pictures of UV bulbs, you'd see a lot of additional blue/pink colour when shooting with most digital cameras, where film would have more correct representation of visible scene.

From this we can conclude that improving colour filters and better filtering invisible light components can have effect on reducing the registration of light that is not any more
corrected by lens, so reducing haze, CA's and improving image contrast.

Btw this is an interesting topic showing one of the areas where sensor design can give great benefit to picture quality other than noise and resolution improvements.

**Lowell Goudge** · 10-03-2009, 05:08 AM

i think you need to split this into 2 questions.

PF from overflow of the sensor is body/sensor specific. I have never really noticed this on any of my cameras, *istD, K10D or K7D.

CA, is a lens issue, and my SMC300F4 gives the same amount of CA in baacklit situations purple one side, green the other on out of focus subjects like twigs etc on all 3 bodies, as it should because it is a lens issue

ytterbium · 10-03-2009, 05:33 AM

By overflow, do you mean saturation, or spectral overflow (sensor registering wider spectrum than lens is corrected for).
Simple blown highlight type overflow would not have a colour cast.

Btw, has anyone experienced reduced PF while using additional UV/Warming/Skylight and IR filters?
But i guess it must be very steep, to have noticeable effect.

knumbnutz · 10-03-2009, 05:30 PM

Ok ..
I will say this .
I had a K10D and DA*16-50 and noticed a lot PF. Seemed like it was on everything.

I then bought a K20D and IMMEDIATELY noticed a difference with a lot less PF and spoke to a fellow with a K20D who said the same thing, an immediate noticable reduction in PF. ALL GOOD

BUT in the new spec for the K7 they mention that it corrects PF and distortion when you put on a lens which is recognised by the camera ie a pentax lens.

So I am seeking those that have noticed this effect and can say that it is good and works.
Cheers Neil

**jbinpg** · 10-03-2009, 06:26 PM

Same here. PF with strong backlight was a given on my K10D. Same lighting with my K20D and no PF. I understand that pixel overflow was staunched in the K20D by better gating. And yes, PF is sensor specific whereas chromatic abberations are lens-dependent.

Jack

kittykat46 · 10-03-2009, 07:58 PM

The K-7's CA correction is just JPEG post-capture processing, so its not magic. It does make a visible improvement in situations which generate a lot of CA.

And it only works with DA and DFA lenses, which have the necessary lens-characteristic electronic data.
The correction data does get embedded with the RAW file, and you can choose to use it or not in RAW development. The Pentax Digital Camera Utility can certainly apply the data, but I'm not sure if all the other common RAW developers can interprete the correction, though.

I found the penalty in shooting speed to much, so I usually turn Off the camera CA correction function.

brecklundin · 10-03-2009, 08:17 PM

Originally posted by kittykat46

The K-7's CA correction is just JPEG post-capture processing, so its not magic. It does make a visible improvement in situations which generate a lot of CA.

And it only works with DA and DFA lenses, which have the necessary lens-characteristic electronic data.
The correction data does get embedded with the RAW file, and you can choose to use it or not in RAW development. The Pentax Digital Camera Utility can certainly apply the data, but I'm not sure if all the other common RAW developers can interprete the correction, though.

I found the penalty in shooting speed to much, so I usually turn Off the camera CA correction function.

Thanks for that, I was just coming to ask if this was a JPG only feature...and it appears it is...

kenyee · 10-04-2009, 05:40 PM

Originally posted by knumbnutz

I then bought a K20D and IMMEDIATELY noticed a difference with a lot less PF and spoke to a fellow with a K20D who said the same thing, an immediate noticable reduction in PF. ALL GOOD

Ditto my transition from K10D to K20D.
The Tamron 70-300 which PF like crazy on the K10D was a lot better, but when it did PF on the K20D, it was a huge amount of PF...

flyer · 10-04-2009, 06:57 PM

Originally posted by kittykat46

The K-7's CA correction is just JPEG post-capture processing, so its not magic.

It works with RAW as long as you use Pentax Digital Utility 4. The information is imbedded with the pictures and PDU 4 can use it.

brecklundin · 10-04-2009, 07:21 PM

Originally posted by flyer

It works with RAW as long as you use Pentax Digital Utility 4. The information is imbedded with the pictures and PDU 4 can use it.

Does anyone know if there are any LR2.x or PSCS3 plug-in's that will access this data? I can usually manually address and PF in either program but it soure would be nice to have a way to deal with it automagically. Cuz, well, I am a lazy, lazy guy.

10-03-2009, 03:21 AM	#1
knumbnutz Veteran Member Join Date: Jul 2007 Location: Melbourne, Australia Photos: Gallery Posts: 394	K7 compared to K20/K10 CA & PF Hi All, Just interested who has either FE10-17, Da12-24 or DA*16-50 (or any other lens that PF badly) and noticed the purple fringing on the K10 or K20 but now find that their K7 has cured the PF. Do you guys find it is cured/ half cured ? Share your experiences please . Cheers Neil

10-03-2009, 04:52 AM	#2
ytterbium Veteran Member Join Date: Jan 2008 Posts: 1,076	This is not an observation or hard fact. Just some thoughts. Some time ago i remember reading that PF is digital only, sensor related problem, and CA is lens one, much less pronounced on film. Or the other way around. Actually i think its almost the same thing, just PF is effect produced as a result of CAs (like you could get green/red aberrations if the lens was constructed such way), since it is pink it's called PF. Now, for the difference between cameras, sensors and film, and why theoretically it could be cured. If you know a little about lens corrections, you will also know that they do not bend all light equally, but are more or less optimized to visible spectrum. Simple illustration borrowed from wikipedia shows how well each type of improved lens correction helps focusing all colours at the same point. Grap for APO chromatic lens: Graph for Super APO chromatic lens: You can clearly see that even with Super APO lens the graph rapidly deflects when leaving visible spectrum. If you have some older lens with the red infra red focus correction mark, you can imagine how much out of focus invisible portions of light spectrum can be. I think it's even worse with UV. Since the film has different type of colour layers (Yellow/Magenta/Cyan) than digital cameras (Red, Green, Blue) you will have different spectral responses and final rendering of those colours (like you get bluish tint from IR component on hot objects glowing red). If you've took pictures of UV bulbs, you'd see a lot of additional blue/pink colour when shooting with most digital cameras, where film would have more correct representation of visible scene. From this we can conclude that improving colour filters and better filtering invisible light components can have effect on reducing the registration of light that is not any more corrected by lens, so reducing haze, CA's and improving image contrast. Btw this is an interesting topic showing one of the areas where sensor design can give great benefit to picture quality other than noise and resolution improvements. Last edited by ytterbium; 10-03-2009 at 05:27 AM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
K10/K20 image compared to Canon 20D	SParker48	Pentax DSLR Discussion	7	09-16-2010 06:37 PM
k10 or k20	houstonmacgregor	Pentax DSLR Discussion	13	03-06-2010 10:37 AM
used K10 VS used K20	MacKay	Pentax DSLR Discussion	7	02-17-2010 11:23 PM
K10 & K20 over 10000 shutter releases	RussellW	Pentax DSLR Discussion	6	02-06-2010 07:04 AM
K10 vs K20 - go	thePiRaTE!!	Pentax DSLR Discussion	11	03-09-2008 03:28 PM

10-03-2009, 05:08 AM	#3
Lowell Goudge Site Supporter Join Date: Jan 2007 Location: Toronto Photos: Gallery \| Albums Posts: 17,887	i think you need to split this into 2 questions. PF from overflow of the sensor is body/sensor specific. I have never really noticed this on any of my cameras, *istD, K10D or K7D. CA, is a lens issue, and my SMC300F4 gives the same amount of CA in baacklit situations purple one side, green the other on out of focus subjects like twigs etc on all 3 bodies, as it should because it is a lens issue

10-03-2009, 05:33 AM	#4
ytterbium Veteran Member Join Date: Jan 2008 Posts: 1,076	By overflow, do you mean saturation, or spectral overflow (sensor registering wider spectrum than lens is corrected for). Simple blown highlight type overflow would not have a colour cast. Btw, has anyone experienced reduced PF while using additional UV/Warming/Skylight and IR filters? But i guess it must be very steep, to have noticeable effect.

10-03-2009, 05:30 PM	#5
knumbnutz Veteran Member Join Date: Jul 2007 Location: Melbourne, Australia Photos: Gallery Posts: 394 Original Poster	Ok .. I will say this . I had a K10D and DA*16-50 and noticed a lot PF. Seemed like it was on everything. I then bought a K20D and IMMEDIATELY noticed a difference with a lot less PF and spoke to a fellow with a K20D who said the same thing, an immediate noticable reduction in PF. ALL GOOD BUT in the new spec for the K7 they mention that it corrects PF and distortion when you put on a lens which is recognised by the camera ie a pentax lens. So I am seeking those that have noticed this effect and can say that it is good and works. Cheers Neil

10-03-2009, 06:26 PM	#6
jbinpg Site Supporter Join Date: Mar 2008 Location: Prince George, BC Photos: Gallery \| Albums Posts: 3,546	Same here. PF with strong backlight was a given on my K10D. Same lighting with my K20D and no PF. I understand that pixel overflow was staunched in the K20D by better gating. And yes, PF is sensor specific whereas chromatic abberations are lens-dependent. Jack

10-03-2009, 07:58 PM	#7
kittykat46 Veteran Member Join Date: Dec 2007 Posts: 923	The K-7's CA correction is just JPEG post-capture processing, so its not magic. It does make a visible improvement in situations which generate a lot of CA. And it only works with DA and DFA lenses, which have the necessary lens-characteristic electronic data. The correction data does get embedded with the RAW file, and you can choose to use it or not in RAW development. The Pentax Digital Camera Utility can certainly apply the data, but I'm not sure if all the other common RAW developers can interprete the correction, though. I found the penalty in shooting speed to much, so I usually turn Off the camera CA correction function.

10-03-2009, 08:17 PM	#8
brecklundin Veteran Member Join Date: Jun 2009 Photos: Albums Posts: 1,674	Originally posted by kittykat46 The K-7's CA correction is just JPEG post-capture processing, so its not magic. It does make a visible improvement in situations which generate a lot of CA. And it only works with DA and DFA lenses, which have the necessary lens-characteristic electronic data. The correction data does get embedded with the RAW file, and you can choose to use it or not in RAW development. The Pentax Digital Camera Utility can certainly apply the data, but I'm not sure if all the other common RAW developers can interprete the correction, though. I found the penalty in shooting speed to much, so I usually turn Off the camera CA correction function. Thanks for that, I was just coming to ask if this was a JPG only feature...and it appears it is...

10-04-2009, 05:40 PM	#9
kenyee Veteran Member Join Date: Mar 2007 Location: Boston, PRofMA Photos: Albums Posts: 3,026	Originally posted by knumbnutz I then bought a K20D and IMMEDIATELY noticed a difference with a lot less PF and spoke to a fellow with a K20D who said the same thing, an immediate noticable reduction in PF. ALL GOOD Ditto my transition from K10D to K20D. The Tamron 70-300 which PF like crazy on the K10D was a lot better, but when it did PF on the K20D, it was a huge amount of PF...

10-04-2009, 06:57 PM	#10
flyer Veteran Member Join Date: Jan 2008 Location: Ste-Anne des Plaines, Qc., Canada Posts: 2,013	Originally posted by kittykat46 The K-7's CA correction is just JPEG post-capture processing, so its not magic. It works with RAW as long as you use Pentax Digital Utility 4. The information is imbedded with the pictures and PDU 4 can use it.

10-04-2009, 07:21 PM	#11
brecklundin Veteran Member Join Date: Jun 2009 Photos: Albums Posts: 1,674	Originally posted by flyer It works with RAW as long as you use Pentax Digital Utility 4. The information is imbedded with the pictures and PDU 4 can use it. Does anyone know if there are any LR2.x or PSCS3 plug-in's that will access this data? I can usually manually address and PF in either program but it soure would be nice to have a way to deal with it automagically. Cuz, well, I am a lazy, lazy guy.