K5 ISO worth the upgrade form K7? Compare and decide.


Smeggypants

Link Posted 11/11/2010 - 16:43
Good stuff - cheers for posting.

On the low light test the K7 6400 looks as noisy as my K20D is at 6400,and is about the same as the 51200 on the K-5. That's a 3 stop improvement. Excellent news.
[i]Bodies: 1x K-5IIs, 2x K-5, Sony TX-5, Nokia 808
Lenses: Pentax DA 10-17mm ED(IF) Fish Eye, Pentax DA 14mm f/2.8, Sigma 17-70mm f/2.8, Pentax-A 28mm f/2.8, Sigma 30mm F1.4 EX DC, Pentax-A 50mm f/1.2, Pentax-A 50mm f/1.4, Pentax-FA 50mm f/1.4, Pentax-A 50mm f/1.7, Pentax DA* 50-135mm f/2.8, Sigma 135-400mm APO DG, and more ..
Flash: AF-540FGZ, Vivitar 283
Last Edited by Smeggypants on 11/11/2010 - 16:44

amoringello

Link Posted 11/11/2010 - 21:51
OK, I reposted with a couple of "REDO"s for ISO 100 and ISO 6400.

http://moringello.com/public/K5_vs_K7_ISO.html

Tripod, WB the same, double-checked contrast, sharpness, etc... (although should not affect RAW),
made sure SR was off. Used LiveView with ContrastDetection for focusing.
I believe everything was as close to constant and consistent.

The focusing sharpness is now very close for both K5 and K7.
Since the tripod mounts were slightly different position on each camera, the last set might have focused slightly different -- need to build a custom rig or use laser guide to line up better, who knows.
Anyway I think I got them more in-line this time as well.
But WB is a little bit off, but much closer.

I also posted the ISO 6400 images with some processing.
K5 used about 25 for luminance and color noise reduction. The K7 required 100% but still looks pretty rough.

FYI, I noticed I said if you click on the initial images that you'll see 100% crops. That wasn't quite true.
I missed something and you have to additionally click on the [Download] link at the upper-right of the larger image. This will display the full sized 100% crop image.

My first time using iWeb, sorry about that. Nice program for quick mockups, but overly simplistic without a lot of control.

Anvh

Link Posted 11/11/2010 - 22:01
Thank you Amoringello , it looks very good.
I've an little request, if you don't mind.
Could you compare shots with shutter speeds around 1/125?
The K7 had quite some shutter blur there so hopefully they have put a new shutter in it so it doesn't do that anymore.
Stefan


K10D, K5
DA* 16-50, DA* 50-135, D-FA 100 Macro, DA 40 Ltd, DA 18-55
AF-540FGZ

Pentaxophile

Link Posted 11/11/2010 - 22:19
Anvh wrote:
Thank you Amoringello , it looks very good.
I've an little request, if you don't mind.
Could you compare shots with shutter speeds around 1/125?
The K7 had quite some shutter blur there so hopefully they have put a new shutter in it so it doesn't do that anymore.

He will have to take it off tripod for that, & use a wide angle lens.

But you've already gone above and beyond, thank you amoringello!
[link=https://500px.com/will_brealey/[/link]
Last Edited by Pentaxophile on 11/11/2010 - 22:24

Anvh

Link Posted 11/11/2010 - 22:45
Depends on how sturdy the tripod is and I believe you will actually see it better with long lenses because any small movement by the camera is more visible then right?
If you want a real test you can also do a macro shot
Stefan


K10D, K5
DA* 16-50, DA* 50-135, D-FA 100 Macro, DA 40 Ltd, DA 18-55
AF-540FGZ

Pentaxophile

Link Posted 12/11/2010 - 00:39
Don't think so Stefan. As I understand it, that would just pick up normal camera shake. The issue is the sensor moving relative to the camera body and lens, rather than the whole camera moving. Go back to Falk Lumo's 'white paper' and check what he says. He has provisionally confirmed that the K-5 doesn't seem to be noticeably affected, based on some of his friend's hands-on testing, btw.
[link=https://500px.com/will_brealey/[/link]
Last Edited by Pentaxophile on 12/11/2010 - 00:44

amoringello

Link Posted 12/11/2010 - 02:38
I've read on tripod also shows the issue, unless the tripod is sable as the Rock of Gibraltar.
I'm not certain I have seen consistent behavior with the K7 so I may not be able to prove or disprove the existence with the K5.

I am hoping to try to find out, since most of my studio work is at or around 1/125 and the K7 is a bit soft far too often -- but not consistently. But I don't use a tripod, so I cannot say for certain if it is me or the camera.

amoringello

Link Posted 12/11/2010 - 21:08
OK...I tried to see if the shutter bounce was an issue but I guess I'm not really sure how to determine if it was a success or not.

BUT... five or six times AF was absolutely soft on the K5 and ten out of ten times on the K7.
BUT, BUT... ten out of ten times with Live View and contrast AF, focus was spot on for both the K5 and K7!

So, does this sound like an issue with my technique or an issue with standard AF?
Possibly an issue with the mirror bounce since LiveView keeps the mirror locked up???

Any opinions on if these finding mean anything? I'm honestly not sure what to look for. (I'm trying to find Falk's info)
I'm certainly not going to make any claims on findings from less than a dozen shots with each camera.

Sadly, it is getting too dark today to get images at 1/125s shutter speed any more, so I had to cut things short. I'll have to get more valid tests during mid-daylight over the weekend.
Last Edited by amoringello on 12/11/2010 - 21:20

Anvh

Link Posted 12/11/2010 - 21:22
Will if you use a 2 second delay it's only the shutter that cause the camera movement.
It's god to hear the problem "shifted" (most likely) with the K5 1/125 is precisely in the zone many shoot I believe.

amoringello wrote:
So, does this sound like an issue with my technique or an issue with standard AF?
Possibly an issue with the mirror bounce since LiveView keeps the mirror locked up???

Sounds like you need to calibrate the phase AF in the camera, that's the standard AF.
It's not uncommon it isn't spot on and the K5 has the option to calibrate that for each lens.
If you notice all lenses are out of focus you can do a global adjustment with the MODSET hack so you can get in the debug menu... the only problem is that it isn't know how to do that just yet with the k5.

Contrast AF doesn't look at the lens AF or anything fancy, it just try to create as large of a contrast possible between the pixels.
Phase AF is a bit more complicated and it can even see if it needs to focus forward or backwards and by how much, contrast AF can't do that.
Stefan


K10D, K5
DA* 16-50, DA* 50-135, D-FA 100 Macro, DA 40 Ltd, DA 18-55
AF-540FGZ
Add a Comment
You must be registered or logged-in to comment.