K-7 image quality concern

**AlexM** · 06-09-2009, 05:48 PM

Ya know, even with the firmware being at .35, an image that I saw from that Flickr set was VERY good, even at ISO 800....

Case in point:

This is not the full sized image, but when you view the full sized one, I could not see evidence of much noise even at ISO 800...

Full sized image: http://farm4.static.flickr.com/3371/3592113867_52273afe35_o.jpg

So if this is from firmware version 0.35, what must v1.0 be like?

Alex

claude21 · 06-10-2009, 07:42 AM

Originally posted by falconeye

Doing a head-to-head comparison of images from K-7 vs. K20D, same subject, same lens, same RAW conversion, is high up on my list. Probably to be done this week end.

Partly because of threads like this. I'll also run higher iso tests later this month when the updated firmware gets available.

From what I've seen so far, I think that banding is a non issue with K-7. I'll research this more as well, when doing my iso tests.

the K-7 seem's a very capable camera on paper, nice specs etc, still I found that low iso performance is not as good as K10D. But, it may be due to early hardware like Roland has said and I believe he's right. Until we have access to production units, I recognize it's too early to conclude.

creampuff · 06-10-2009, 08:38 AM

Originally posted by claude21

the K-7 seem's a very capable camera on paper, nice specs etc, still I found that low iso performance is not as good as K10D. But, it may be due to early hardware like Roland has said and I believe he's right. Until we have access to production units, I recognize it's too early to conclude.

Is it so difficult to accept the fact that the K-7's image quality is much improved over the K10D? Believe what you want to believe but having owned the K10D, K20D and used the K-7 briefly, there is no doubt in my mind the K-7's image quality is excellent and dare I say surpasses the K10D.

claude21 · 06-10-2009, 08:59 AM

Originally posted by creampuff

Is it so difficult to accept the fact that the K-7's image quality is much improved over the K10D? Believe what you want to believe but having owned the K10D, K20D and used the K-7 briefly, there is no doubt in my mind the K-7's image quality is excellent and dare I say surpasses the K10D.

I don't agree, at iso 100-200 k10d ccd sensor is better than (beta) K-7 at the same iso. That was discussed already for the K20D, samsung cmos sensor is noisier at low iso.

mrt10x · 06-10-2009, 02:46 PM

Originally posted by claude21

I don't agree, at iso 100-200 k10d ccd sensor is better than (beta) K-7 at the same iso. That was discussed already for the K20D, samsung cmos sensor is noisier at low iso.

This is why the k10d has a higher DXO mark than the K20d

wlachan · 06-10-2009, 02:54 PM

I am cool with manageable noise, but I hate BANDING!!!

raz-0 · 06-10-2009, 09:31 PM

I took the large version of the picture mentioned by the OP, and the big versions of the zoo pictures also referenced in the thread, and did horrible things to them in Photoshop.

The pic form this thread has more noise than the k10 pic if you push the dark areas, but the k10 pic also has better lighting and less problem areas.

After molesting the picture, I get no banding artifacts, even in the problem areas.

As for the zoo pictures, I can't find anything horribly wrong with them even when beating up on it in photoshop. At least nothing that is any more limiting that what you get out of a k10d.

The only thing I am finding is that there is more noise in browns in the pictures relative to the rest of the scene than with the k10d, and slightly more chroma noise overall. However, it is quite controllable noise with decent noise reduction software.

I'd be shocked if it didn't improve a good chunk before it hits the street. If I were going to be concerned about anything, it would be about the dynamic range of the pictures as the samples I am seeing that have nice inky black areas are clipping them in spots, which has potential to limit you in post production.

Class A · 06-25-2009, 09:16 PM

Originally posted by claude21

The Pentax image is official one found on pentax site. State of the art star lens show a pretty lack of sharpness and contrast on K-7, ...

I agree. I would not have put up this image to advertise the camera.

OK, it is a portrait and you are not supposed to see all the imperfections of a woman's skin but in order to impress people you want really sharp lashes for example.

I trust the K-7 will not be found lacking in sharpness when it hits the stores but I'm slightly puzzled why so many shots are soft and why so few people take issue with it.

Psynema · 06-25-2009, 09:22 PM

I'm not impressed with the 55mm headshot for the K7 - it looks good but prosumer color and a bit soft, but really , it's firmware .35 and there are ton's of pics of outstanding quality on past cameras...so bleh.

That headshot looks a bit like she's too close for min. focus by a tad...I may be overanalyzing it though, it is softish

benjikan · 06-26-2009, 03:11 AM

I would most certainly wait till the final production model hits the street before jumping to any conclusions.

Yonnie · 06-26-2009, 03:17 AM

Originally posted by benjikan

I would most certainly wait till the final production model hits the street before jumping to any conclusions.

Thank you! Common sense would suggest we all do this.

Maffer · 06-26-2009, 03:26 AM

Originally posted by Psynema

I'm not impressed with the 55mm headshot for the K7 - it looks good but prosumer color and a bit soft, but really , it's firmware .35 and there are ton's of pics of outstanding quality on past cameras...so bleh.

That headshot looks a bit like she's too close for min. focus by a tad...I may be overanalyzing it though, it is softish

The headshot with da55mm looks like misfocused. For example those eyelashes - no way they can be that soft when properly in focus unless some softener etc. was used in purpose.

falconeye · 06-26-2009, 03:41 AM

You have all seen my K-7 detailed resolution analysis, haven't you?

Falk Lumo: Comparative resolution study K-7 vs. K20D

Basically, it found that resolution is at least on par with K20D.

benjikan · 06-26-2009, 03:54 AM

I would most certainly wait till the final production model hits the street before jumping to any conclusions.

falconeye · 06-26-2009, 06:36 AM

Originally posted by benjikan

I would most certainly wait till the final production model hits the street before jumping to any conclusions.

How many more times are you going to repeat?

I have carefully examined with Pentax what can already be said and what would better await full production models. Consequently, there is some material which I did publish and there is some material (already prepared for my blog) which I didn't publish.

BTW, since when do you await final production models hitting the street before saying anything about forthcoming models?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Are you satisfied with the K-x image quality?	rjm	Pentax DSLR Discussion	37	01-21-2010 06:27 AM
How can I get good image quality from a K-x??	Manfred	Troubleshooting and Beginner Help	38	12-20-2009 08:18 PM
K or M Series - Which has the best image quality	8540tomg	Pentax SLR Lens Discussion	20	10-05-2009 07:53 AM
DA 16-45mm image quality	sveinmb	Pentax SLR Lens Discussion	25	09-18-2008 03:58 PM
Some concern about image quality.	Bart	Pentax SLR Lens Discussion	17	07-23-2007 05:54 PM

06-09-2009, 05:48 PM	#16
AlexM Site Supporter Join Date: Mar 2008 Location: Prescott Valley, AZ Photos: Gallery Posts: 171	Ya know, even with the firmware being at .35, an image that I saw from that Flickr set was VERY good, even at ISO 800.... Case in point: This is not the full sized image, but when you view the full sized one, I could not see evidence of much noise even at ISO 800... Full sized image: http://farm4.static.flickr.com/3371/3592113867_52273afe35_o.jpg So if this is from firmware version 0.35, what must v1.0 be like? Alex

06-10-2009, 07:42 AM	#17
claude21 Forum Member Join Date: Jan 2008 Posts: 55 Original Poster	Originally posted by falconeye Doing a head-to-head comparison of images from K-7 vs. K20D, same subject, same lens, same RAW conversion, is high up on my list. Probably to be done this week end. Partly because of threads like this. I'll also run higher iso tests later this month when the updated firmware gets available. From what I've seen so far, I think that banding is a non issue with K-7. I'll research this more as well, when doing my iso tests. the K-7 seem's a very capable camera on paper, nice specs etc, still I found that low iso performance is not as good as K10D. But, it may be due to early hardware like Roland has said and I believe he's right. Until we have access to production units, I recognize it's too early to conclude.

06-10-2009, 08:38 AM	#18
creampuff Veteran Member Join Date: May 2007 Location: Singapore Photos: Gallery Posts: 3,953	Originally posted by claude21 the K-7 seem's a very capable camera on paper, nice specs etc, still I found that low iso performance is not as good as K10D. But, it may be due to early hardware like Roland has said and I believe he's right. Until we have access to production units, I recognize it's too early to conclude. Is it so difficult to accept the fact that the K-7's image quality is much improved over the K10D? Believe what you want to believe but having owned the K10D, K20D and used the K-7 briefly, there is no doubt in my mind the K-7's image quality is excellent and dare I say surpasses the K10D.

06-10-2009, 08:59 AM	#19
claude21 Forum Member Join Date: Jan 2008 Posts: 55 Original Poster	Originally posted by creampuff Is it so difficult to accept the fact that the K-7's image quality is much improved over the K10D? Believe what you want to believe but having owned the K10D, K20D and used the K-7 briefly, there is no doubt in my mind the K-7's image quality is excellent and dare I say surpasses the K10D. I don't agree, at iso 100-200 k10d ccd sensor is better than (beta) K-7 at the same iso. That was discussed already for the K20D, samsung cmos sensor is noisier at low iso.

06-10-2009, 02:46 PM	#20
mrt10x Senior Member Join Date: Mar 2009 Location: Woodland Park CO Posts: 102	Originally posted by claude21 I don't agree, at iso 100-200 k10d ccd sensor is better than (beta) K-7 at the same iso. That was discussed already for the K20D, samsung cmos sensor is noisier at low iso. This is why the k10d has a higher DXO mark than the K20d

06-10-2009, 02:54 PM	#21
wlachan Veteran Member Join Date: Sep 2006 Location: Canada Photos: Albums Posts: 2,625	I am cool with manageable noise, but I hate BANDING!!!

06-10-2009, 09:31 PM	#22
raz-0 Veteran Member Join Date: May 2007 Location: Central NJ Posts: 470	I took the large version of the picture mentioned by the OP, and the big versions of the zoo pictures also referenced in the thread, and did horrible things to them in Photoshop. The pic form this thread has more noise than the k10 pic if you push the dark areas, but the k10 pic also has better lighting and less problem areas. After molesting the picture, I get no banding artifacts, even in the problem areas. As for the zoo pictures, I can't find anything horribly wrong with them even when beating up on it in photoshop. At least nothing that is any more limiting that what you get out of a k10d. The only thing I am finding is that there is more noise in browns in the pictures relative to the rest of the scene than with the k10d, and slightly more chroma noise overall. However, it is quite controllable noise with decent noise reduction software. I'd be shocked if it didn't improve a good chunk before it hits the street. If I were going to be concerned about anything, it would be about the dynamic range of the pictures as the samples I am seeing that have nice inky black areas are clipping them in spots, which has potential to limit you in post production.

06-25-2009, 09:16 PM	#23
Class A Pentaxian Join Date: Aug 2008 Location: Wellington, New Zealand Posts: 11,251	Originally posted by claude21 The Pentax image is official one found on pentax site. State of the art star lens show a pretty lack of sharpness and contrast on K-7, ... I agree. I would not have put up this image to advertise the camera. OK, it is a portrait and you are not supposed to see all the imperfections of a woman's skin but in order to impress people you want really sharp lashes for example. I trust the K-7 will not be found lacking in sharpness when it hits the stores but I'm slightly puzzled why so many shots are soft and why so few people take issue with it.

06-25-2009, 09:22 PM	#24
Psynema Senior Member Join Date: Jun 2009 Posts: 136	I'm not impressed with the 55mm headshot for the K7 - it looks good but prosumer color and a bit soft, but really , it's firmware .35 and there are ton's of pics of outstanding quality on past cameras...so bleh. That headshot looks a bit like she's too close for min. focus by a tad...I may be overanalyzing it though, it is softish

06-26-2009, 03:11 AM	#25
benjikan Veteran Member Join Date: Jan 2007 Location: Paris, France Photos: Gallery Posts: 4,308	I would most certainly wait till the final production model hits the street before jumping to any conclusions.

Thread Tools	Search this Thread
	Search this Thread: Advanced Search

06-26-2009, 03:17 AM	#26
Yonnie Senior Member Join Date: Nov 2007 Location: Tennessee, USA Posts: 110	Originally posted by benjikan I would most certainly wait till the final production model hits the street before jumping to any conclusions. Thank you! Common sense would suggest we all do this.

06-26-2009, 03:26 AM	#27
Maffer Inactive Account Join Date: Sep 2008 Location: Finland Photos: Gallery Posts: 308	Originally posted by Psynema I'm not impressed with the 55mm headshot for the K7 - it looks good but prosumer color and a bit soft, but really , it's firmware .35 and there are ton's of pics of outstanding quality on past cameras...so bleh. That headshot looks a bit like she's too close for min. focus by a tad...I may be overanalyzing it though, it is softish The headshot with da55mm looks like misfocused. For example those eyelashes - no way they can be that soft when properly in focus unless some softener etc. was used in purpose.

06-26-2009, 03:41 AM	#28
falconeye Veteran Member Join Date: Jan 2008 Location: Munich, Alps, Germany Photos: Gallery Posts: 6,871	You have all seen my K-7 detailed resolution analysis, haven't you? Falk Lumo: Comparative resolution study K-7 vs. K20D Basically, it found that resolution is at least on par with K20D.

06-26-2009, 03:54 AM	#29
benjikan Veteran Member Join Date: Jan 2007 Location: Paris, France Photos: Gallery Posts: 4,308	I would most certainly wait till the final production model hits the street before jumping to any conclusions.

06-26-2009, 06:36 AM	#30
falconeye Veteran Member Join Date: Jan 2008 Location: Munich, Alps, Germany Photos: Gallery Posts: 6,871	Originally posted by benjikan I would most certainly wait till the final production model hits the street before jumping to any conclusions. How many more times are you going to repeat? I have carefully examined with Pentax what can already be said and what would better await full production models. Consequently, there is some material which I did publish and there is some material (already prepared for my blog) which I didn't publish. BTW, since when do you await final production models hitting the street before saying anything about forthcoming models?