Pages: 1 [2] 3 4   Go Down

Author Topic: Quick Dynamic range comparison - Leaf AFi-ii 12, CFii-39MS, Phase P20, Canon 5D2  (Read 18927 times)

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA

By the way the Leaf crops were presented to be about the same portion of the frame - here's a 100% crop with no noise reduction.   I should also mention, I was doing two tests at once here because I wanted to test for purple fringing and for infinity focus.  The lens was only stopped down to f/5.6 and focused at full infinity stop so this region is not as sharp as it would be were I to have either stopped down or focused closer in from infinity.  I only mention this as you will be looking at a 100% crop and don't want people to believe the back can't get detail - that it can do and in spades. 
« Last Edit: May 19, 2011, 04:40:15 pm by EricWHiss »
Logged
Rolleiflex USA

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA

Conclusions:  
After manipulating the files in various programs and experimenting to see which methods and programs (ie. shadows, fill light, exposure adjusts, curves, NR, etc),  I've found where I personally would consider the maximum usable DR for each camera/back.  I simply tried to find the best shadow lift combination where the noise reduction could still control the noise level enough to make a satisfactory print. I looked at the wedge and recorded the darkest discernible level and converted to stops.  Probably Imatest would do much much better at quantifying these data, and even then I suspect that many of you out there are working with different definitions of DR, so for the sake of argument (or rather to avoid arguing) I'll call these numbers Eric's usable DR figures or EDR.  I find looking at what will be usable in print a much more useful number than a manufacturers spec or a number from DXO but YMMV.    Note that this does not really take in account for the differences in pixel count.  Also worth noting the first aptus 12 I had for testing seemed to perform a bit better.

Canon 26 steps = 8.7 stops EDR
Hasselblad CFii-39 in single shot   33 steps = 11 stops EDR
Hasselblad CFii-39 in multi-shot    35 steps = 11.7 stops EDR
Leaf Aptus 12  32 steps = 10.7 stops EDR
Phase p20 (non plus) 29 steps = 9.7 stops EDR



« Last Edit: May 20, 2011, 02:33:41 pm by EricWHiss »
Logged
Rolleiflex USA

nazdravanul

  • Full Member
  • ***
  • Offline Offline
  • Posts: 132
    • Stefan Iacob - visual artist

Thank you for your time, effort and articulate presentation. Seems like quite useful stuff. I would have liked to see a newer generation Hasselblad sensor (one of the Kodak Truesense something I believe it's called, from the h4d40, h4d50, and/or h4d50 MS )  thrown in the mix, but hey  ... you can't have everything in life :) . Also a P65+ , IQ160, H4d60 thrown in the mix (the other Dalsa sensor before the 80) would have been interesting, and hell why not an iq180 at iso 32 :))) ... I know, I'm being shameless here ... :) 
Logged

henrikfoto

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 899

Hi Eric!

Seems like the Leaf 12 is coming out poorly in your test.

What do you think about this back asides from the dr?

Henrik
Logged

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA

Hi Eric!

Seems like the Leaf 12 is coming out poorly in your test.

What do you think about this back asides from the dr?

Henrik
Hi Henrik,
Well I would not say it was doing poorly exactly ESP. considering it has twice the pixels. What did you think from the sample images?

I must say that this is my third aptus 12 back. The first one was a loaner for testing generously provided by Paul Slotboom of Optechs Digital and it was nothing short of brilliant. That one pulled a 34 step and would have been the best overall single shot plus it had none of the purple fringes that I'm seeing on occasion with the other two. Based on the loaner's performance, I purchased one which had unluckily had some problems and needed to be sent back. The replacement does not appear to be as good as the loaner. The promise is there though. I like a lot about it and would be 100% satisfied with an equal to the loaner as I like the features of the back overall. 
« Last Edit: May 20, 2011, 05:54:21 pm by EricWHiss »
Logged
Rolleiflex USA

henrikfoto

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 899

That is strange. Do you know which one is the newer?

I have heard that there have been two different productions of the sensor because
Leaf ran dry. Could there be a quality-differense between the two batches?

Henrik
Logged

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA

Henrik,
Thats interesting - I wonder if there is a way to tell from the file EXIF data?   I can see the loaner had V 5.2 firmware while the others have V 5.1 but Yair didn't think that would make a difference.  I can see a difference in the black file when really pushed - the sensor is made from 4 smaller rectangles and did note a difference - one chip the the diagonally opposed squares were matched and on the other the halves were matched (like the CFii-39MS)  but didn't think this was important.
Eric
Logged
Rolleiflex USA

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA

Thank you for your time, effort and articulate presentation. Seems like quite useful stuff. I would have liked to see a newer generation Hasselblad sensor (one of the Kodak Truesense something I believe it's called, from the h4d40, h4d50, and/or h4d50 MS )  thrown in the mix, but hey  ... you can't have everything in life :) . Also a P65+ , IQ160, H4d60 thrown in the mix (the other Dalsa sensor before the 80) would have been interesting, and hell why not an iq180 at iso 32 :))) ... I know, I'm being shameless here ... :) 

Yes, I'd really like to have seen the IQ 180 and the newer hasselblad backs but neither are made to fit on my cameras. My understanding from Doug Peterson is that the IQ 180 mostly will differ in long exposure times and sensor+ but we'll probably see once users have had them for a while.
Logged
Rolleiflex USA

gazwas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 539

I'm not trying to make excuses for Phase/Leaf but the 80MP chip is still very new and I imagine the firmware in the back is still not as optimised as the other much more established Hasselblad backs used. The P65+ got quite a few FW upgrades after launch to improve IQ so I suspect the IQ180 will get the same.

Excellent test and brilliant nerdy reading!  ;)
Logged
trying to think of something meaningful........ Err?

marcmccalmont

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1780

Eric
If you can get your hands on a pentax k5 it would be interesting to see the results
I'm continually blown away with the DR of this little camera. It's going to be interesting which has wider DR the k5 or the IQ180?
Marc
Logged
Marc McCalmont

chiek

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 284
    • chiek imaging

Hi Eric.
Excellent test. Thanks for sharing.
How about image quality feeling? No mechanical, only feeling...
As you know, I'm using cf-39ms but I like leaf back tonality.
I'm very interested in new aptus12.
Cheers.
Logged
chiek imaging, in Seoul, SOUTH-KOREA.
Sinar P2, Hasselblad CFv-50c medium format and a7R systems
major job is products shot, especially for electronic products.
but interested in Landscapes and Portraits, Still-life.
my hobby is Designing camera…
www.chiek.co.kr

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA

Hi Eric.
Excellent test. Thanks for sharing.
How about image quality feeling? No mechanical, only feeling...
As you know, I'm using cf-39ms but I like leaf back tonality.
I'm very interested in new aptus12.
Cheers.

Thanks Guys,

Chiek, good point to bring up -  the tonality or feel is very important aspect, and I did do some testing vs the other cameras.  I'll be happy to post that stuff, but probably best to start another thread which I'll do in a couple days.    I also looked at how each camera could render detail through the aperture range and really quite a lot of other stuff.   

Mark - yes would love to get my hands on the K5 as I've heard good things

Gawas - probably you are right and I hope so too!  I got sample files from Leaf that had higher rev versions of firmware, and the loaner back also had a higher rev so we at least know they are working on it.



Logged
Rolleiflex USA

marcmccalmont

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1780

Here is a K5 example, I've never been able to point a camera straight up and capture both highlight and shadow detail until now!
Marc
Logged
Marc McCalmont

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA

Marc,
Looks impressive, thanks for sharing.  I checked DP Review who also shoot a stouffer wedge and they are reporting 29 stops which would put it between Canon 5D and the digital backs but their testing method may differ from mine.  http://www.dpreview.com/reviews/pentaxk5/page14.asp

Eric
Logged
Rolleiflex USA

marcmccalmont

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1780

Unfortunately they test an in camera jpeg not a RAW. The K5 trumps my P45+ with regards to DR but I don't have them side by side for a good RAW comparison
I'm hoping the IQ180 is at least close (within a 1/2 stop) of the K5 I think DxO mark tests are valid so time will tell
Marc
Logged
Marc McCalmont

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA

Marc,
Sounds really impressive.  Would it be possible for you to post a side by side of the K5 vs. P45+? 

Thanks,
Eric
Logged
Rolleiflex USA

marcmccalmont

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1780

Marc,
Sounds really impressive.  Would it be possible for you to post a side by side of the K5 vs. P45+?  

Thanks,
Eric


sent the P45+ in for trade on a IQ180 so I no longer have it, I have my fingers crossed that the IQ180 will come close to the K5 with respect to DR
Marc
« Last Edit: May 22, 2011, 04:44:41 am by marcmccalmont »
Logged
Marc McCalmont

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/

sent the P45+ in for trade on a IQ180 so I no longer have it, I have my fingers crossed that the IQ180 will come close to the K5 with respect to DR
Marc

How come you are allowed to say about the K5 what I have been saying about the D3x for 2.5 years, and nobody calls you a DSLR troll?  ;D

Cheers,
Bernard

John R Smith

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1357
  • Still crazy, after all these years

How come you are allowed to say about the K5 what I have been saying about the D3x for 2.5 years, and nobody calls you a DSLR troll?  ;D

Because no MF shooter feels threatened by a Pentax K5. Whereas they do feel threatened by a top-end Nikon.

John
Logged
Hasselblad 500 C/M, SWC and CFV-39 DB
an

bcooter

  • Guest

Because no MF shooter feels threatened by a Pentax K5. Whereas they do feel threatened by a top-end Nikon.

John


I don't understand why anyone would feel threatened unless they have a financial arrangement with a specific brand, but what anyone uses really is up to them.

I know 4 very good photographers that went from their backs to the d3x and never blinked and I thought about it since I have a full case of Nikon glass.  The only thing that stopped me was I like the 1ds3 skin tones and with the Nikon I was told you can't see the lcd when tethering.

But cameras, it's all up to the person.  I'd rather use my p21+ than my p30+ because it's so much faster and responsive.  My best friend just returned from shooting the stars at Cannes and he only uses a p21+ on a V system and he is one of the few photographers I know that for commerce, with his contract,  every print goes to 60 inches minimum.

It's all about what you are comfortable with.

IMO

BC
Logged
Pages: 1 [2] 3 4   Go Up