Pages: 1 ... 3 4 [5] 6 7 8   Go Down

Author Topic: Mini Medium Format Shootout  (Read 55281 times)

Rob C

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 24074
Re: Mini Medium Format Shootout
« Reply #80 on: November 19, 2010, 03:45:02 pm »

It also prompts the question "Are we assessing the camera, the end product or the shortcomings of the people behind the cameras?"


Ah, Keith, if we exclude the last in the group, we may as well all go back to watching the tv. It's always the human element that throws the googly!

And how to assess the first two without the last?

As I have concluded in the past, nobody else can really do the testing for us. Unfortunately! Or maybe not; depends on what you want tested.

Rob C

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Mini Medium Format Shootout
« Reply #81 on: November 19, 2010, 07:30:41 pm »

Hi,

Which of course gets us into proper sharpening. Sharpening affects noise, doesn't it? I guess sharpening methods differ a lot and FocusMagic may be one of the better ones in this regard. Would sharpening affect SNR and DR either in the physical or metaphysical (perceived) sense?

Best regards
Erik

Yes, that's with the TS-E 90 mm @ f/7.1 . I used Capture One as Raw converter (extracted the most detail from the 1Ds3 Raws) without sharpening, and I used FocusMagic for the capture sharpening.

Cheers,
Bart
Logged
Erik Kaffehr
 

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8913
Re: Mini Medium Format Shootout
« Reply #82 on: November 19, 2010, 08:06:33 pm »

Would sharpening affect SNR and DR either in the physical or metaphysical (perceived) sense?

Hi Erik,

Depends on the sharpening method, and the noise spectrum. Random (white) noise has a power spectrum that goes from low spatial frequencies to high spatial frequencies. Deconvolution sharpening is usually restricted to a limited range, most likely the highest spatial frequencies, so the noise there will be boosted with the signal while leaving the rest alone. However, there are methods that specifically seek to boost the signal more than the noise, thus increasing the S/N ratio at the targeted spatial frequencies. The adaptive version of the Richardson Lucy algorithm is an example of such an algorithm, and FocusMagic also does a good job.

This manifests itself as a boost of the MTF curve especially at the highest spatial frequencies, near the Nyquist frequency, while lower spatial frequencies benefit much less. So with a proper algorithm, the benefits are greater than the drawbacks, and it results in visibly restored resolution.

Cheers,
Bart
« Last Edit: November 20, 2010, 05:46:59 am by BartvanderWolf »
Logged
== If you do what you did, you'll get what you got. ==

Rob C

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 24074
Re: Mini Medium Format Shootout
« Reply #83 on: November 20, 2010, 06:43:24 am »

At last we have an example that does justice to the capture device.

Bart, thank you.




And more!

Would have also been interesting to know which optic was used, not that I'm about to swap over to C from N, or anything much else, to be honest about it; but interesting knowledge to have, regardless.

I did note the bricks.

;-)

Rob C

Rob C

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 24074
Re: Mini Medium Format Shootout
« Reply #84 on: November 20, 2010, 07:14:40 am »

Rob, see above.


As my wife used to say when I didn't know where I thought she'd hidden my things: Rob, have you tried looking under your nose?

Thanks!

Rob C

P.S. If I thought that the Nikkor 24mm tiltshifter could be as good, I'd be sorely tempted. But then, I always am. Tempted, I mean.
« Last Edit: November 20, 2010, 07:17:29 am by Rob C »
Logged

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8913
Re: Mini Medium Format Shootout
« Reply #85 on: November 20, 2010, 08:23:49 am »

At last we have an example that does justice to the capture device.

Bart, thank you.

You're welcome. But it was not my intention to hyjack the tread, just to show that there are no bad cameras when the proper technique is used, which obviously wasn't the case with the example in the original test. Lenses can make a lot of difference, as does proper Raw conversion and capture sharpening (especially with an AA-filter fitted setup).

Cheers,
Bart
Logged
== If you do what you did, you'll get what you got. ==

Joseph Yeung

  • Newbie
  • *
  • Offline Offline
  • Posts: 21
Re: Mini Medium Format Shootout
« Reply #86 on: November 20, 2010, 11:23:51 am »

And with that, let's bring the discussion back on topic--not that the important people feel like participating, it seems.

Since no clarification from the writers is forthcoming, I'll assume the worst for now and suppose they shrunk the images because no more detail could be seen at higher magnification.  ::)

Since the centre and corner crops of the Canon were taken from roughly the same location as the MF backs, I'll assume that FOV of the images were the same, so 100% crops from two cameras with the same MP count would have the same magnification.

So without further ado...



Comparing the P40+ crop and the 1Ds3 crop, supposing the 1Ds3 crop was 100%, then if the P40+ crop were 100%, the P40+ would be a 13.3MP camera.  Ya think? ;)



If the magnifications were the same, the P40+ would be a 21MP camera just like the Canon...



And finally, this is the size one would have expected the P40+ 100% crop to be if it's a 40MP camera and the FOV are the same.

And until I see a clarification from the people that matter, I'll suppose that's how the P40+ looks like at 100%. :D

(edit: P40+ crops resized using Photoshop standard bicubic (neither smoother nor sharper), for anyone who cares.  The resizes were 125.6% to get up to 21MP size and 173.37% to get up to 40MP size.  13.3*(125.6%^2)=21MP, 13.3*(173.37%^2)=40MP)
« Last Edit: November 20, 2010, 11:50:56 am by Joseph Yeung »
Logged

dmerger

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 680
Re: Mini Medium Format Shootout
« Reply #87 on: November 20, 2010, 12:21:38 pm »

Joseph, as you amply demonstrated, it is difficult to compare photos on screen from cameras with different megapixels, especially with regards to resolution.   Perhaps the authors of the article will explain their procedure for the crops included in the article.  Without an explanation, it's probable that some people will draw misleading conclusions when comparing the crops.  I suspect that this crop difficulty has much more influence on the perceived short comings of the Canon sample than does the choice of lens.

I also noticed that many of the photos have an embedded color profile for what appears to be Mark's monitor rather than sRGB, aRGB or ProPhoto.  Using a monitor profile seems like an unusual choice.  Perhaps the authors of the article could explain why they chose to use the monitor profile, or perhaps it was inadvertent. 
Logged
Dean Erger

darr

  • Full Member
  • ***
  • Offline Offline
  • Posts: 145
Re: Mini Medium Format Shootout
« Reply #88 on: November 20, 2010, 01:12:26 pm »

Hmmm...

I do not know why all this cropping with a Canon compared to a P40+ has any relevance to the Pentax 645D.
Can we stay on topic?  Maybe a new thread addressing that isolated issue would produced many more comments and examples for the Canon users to compare.  Just a thought.
Logged

mgrayson

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 67
Re: Mini Medium Format Shootout
« Reply #89 on: November 20, 2010, 02:22:40 pm »

It WAS in another thread, but was completely ignored there. :-\

It is on-topic because a possibly valid proposition, e.g., the 645D is a great camera, is weakened by false supporting arguments. Now the whole article, which was almost certainly done in good faith, comes across as an exercise in self-justification.

Best,

Matt
Logged

tsjanik

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 720
Re: Mini Medium Format Shootout
« Reply #90 on: November 20, 2010, 05:35:05 pm »

Matt, I don’t understand how the presence or absence of any report of the Canon affects the validity of the 645D/Phase comparison.  If Nick and Mark were to edit out all reference to the Canon, would the comparison of the MFD cameras then be valid?  Although you can suggest that Nick is motivated by self-justification that certainly wouldn’t be the case for Mark, who recently parted with considerable cash for the Phase. Oh, I forgot, it has already been pointed out that he unconsciously caused the Canon to perform poorly to reduce any cogitative dissonance over his Phase purchase.

Really this has gotten quite silly.  Two guys test their cameras as well as they can, take the trouble to write a report for our free enjoyment and they get skewered.
Logged

mgrayson

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 67
Re: Mini Medium Format Shootout
« Reply #91 on: November 20, 2010, 06:22:03 pm »

It doesn't affect the 645D/Phase comparison. It does affect the credibility of the authors. That isn't silly or irrelevant. If an article displays either poor technique or outright bias in one area, doesn't that alter your opinion of the rest? It's sad because the rest of the comparison is so well done. By leaving out the Canon bashing they would have done this site a great service by enhancing its credibility.

And I did not suggest that the authors were motivated by self-justification. Please read what I wrote.
Logged

DaveCurtis

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 508
    • http://www.magiclight.co.nz
Re: Mini Medium Format Shootout
« Reply #92 on: November 20, 2010, 07:17:58 pm »

I can see why Michael gave up on the pixel-peeping reviews.  :)
Logged

Joseph Yeung

  • Newbie
  • *
  • Offline Offline
  • Posts: 21
Re: Mini Medium Format Shootout
« Reply #93 on: November 20, 2010, 08:54:10 pm »

Matt, I don’t understand how the presence or absence of any report of the Canon affects the validity of the 645D/Phase comparison.  If Nick and Mark were to edit out all reference to the Canon, would the comparison of the MFD cameras then be valid?  Although you can suggest that Nick is motivated by self-justification that certainly wouldn’t be the case for Mark, who recently parted with considerable cash for the Phase. Oh, I forgot, it has already been pointed out that he unconsciously caused the Canon to perform poorly to reduce any cogitative dissonance over his Phase purchase.

Really this has gotten quite silly.  Two guys test their cameras as well as they can, take the trouble to write a report for our free enjoyment and they get skewered.


The point is not that they've stacked the deck against Canon.  The point is that the 645D and P40+ 100% crops in the articles are not 100% crops.  (More like 57%) They are useless for judging pixel quality.  If you take a 57% crop and take that as the pixel quality at 100%, you're massively overestimating the pixel quality of both the 645D or the P40+.  Forget the Canon.  For all we know the 645D could have been massively superior to the P40+, or the other way round, we just can't see it.  Of course, you could just take Nick and Mark's words for everything, in which case I don't know why they bothered to post any pictures at all.



"seen at 100% screen magnification..." the title bars in my Photoshop window are about 300% the size of this "100% crop" in each direction...

OK, this will be my last post in this thread until I hear back from Nick and Mark.
« Last Edit: November 20, 2010, 09:16:24 pm by Joseph Yeung »
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
Re: Mini Medium Format Shootout
« Reply #94 on: November 22, 2010, 03:32:28 pm »

I'm wondering if there was any capture sharpening done in this example. The 24-105 is not that soft. I downloaded the center example and imported into Lightroom and found that with a bit of capture sharpening the example was as crisp as the Leica (almost).

Either there was no capture sharpening at all or only the default sharpening which is very soft was applied. In the Michael Reichmann and Jeff Schewe Lightroom 3 tutorial there is an excellent video on capture sharpening. For anybody who have seen this, it is very clear that using either no or just the default capture sharpening will be missing a lot of details. I suspect that the 1Ds mkIII example was not properly done wrt. the capture sharpening.

For your information, all images were capture sharpened using Photokit Sharpener hi-res digital, leaving the contours and the pass through layers all at their default opacities.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
Re: Mini Medium Format Shootout
« Reply #95 on: November 22, 2010, 03:40:51 pm »

The lack of a blur filter improves apparent sharpness but can result in alaising. Why does Nikon go to the expense of using a blur filter (as does Canon and almost all other dSRL makers)? I understand that a blur filter would be prohibitively expensive for a MFDB, so it is really not an option there.

Regards,

Bill

The Leica M9 isn't MF, but it also has no AA filter. I think this is a matter of design philosophy (sizing-up the relative value of the trade-offs) and marketing. There are hardware and software approaches to this problem - when it occurs.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

Dick Roadnight

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1730
Re: Mini Medium Format Shootout
« Reply #96 on: November 22, 2010, 03:49:49 pm »

For your information, all images were capture sharpened using Photokit Sharpener hi-res digital, leaving the contours and the pass through layers all at their default opacities.
One of the main differences between the MFs is the software they come with... If you were doing a comparative test between 400hp super-cars, would you insist on testing them all with mini tyres?
Logged
Hasselblad H4, Sinar P3 monorail view camera, Schneider Apo-digitar lenses

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
Re: Mini Medium Format Shootout
« Reply #97 on: November 22, 2010, 03:52:07 pm »

Not so, just waiting for a response from the author(s), they are the only ones who know what really happened. The crops were undoubtedly screen captures, thus 100% crops, but it seems like the images shown in the essay were resampled further. That would make them less useful as comparison material, to put it mildly.

Yes, screen captures of the image magnified "a la Photoshop" to 100%. Because what emerges from this process is a JPEG which is then up-loaded to the internet, for sure there is much compression. Nonetheless, the impression you get from looking at the images on this web forum is reasonably similar in a comparative sense to the impression you get seeing the actual images on display at 100%. Not perfectly so, but decent enough to confirm what we told you in the text.

As for the 1DsMk3, i use it extensively with the Canon 24~105 L lens and I consider this lens to be FAR ABOVE average, as have other reviewers and users. We exercised our judgment about whether to include the FF stuff in an essentially MF review, and we decided on balance of the pros and cons to do it - some will agree that it has merit, while others will disagree - so be it. We have already been around this issue here, and in the article, so I really have nothing more to add on that matter.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
Re: Mini Medium Format Shootout
« Reply #98 on: November 22, 2010, 04:09:54 pm »

Hey! Joseph. Don't spoil the fun  ;D . These guys are still exited by possession of their new, expensive cameras.

Don't expect them to get too analytical and serious. Once you've splashed out a lot of dough on a really expensive system, there's a strange reluctance to be completely objective during comparisons with a much less expensive system.

But my expectations of you were better than this. You are supposed to be a seasoned professional photographer who can see objectively without personalizing issues using phoney psyc-101 non-insights. Obviously, despite the content and context of this article you remain completely clueless about the mindsets brought to bear on its research and preparation. Comments of this ilk don't belong on this website because they contribute NOTHING to learning and understanding, and this site and forum are meant to be about that.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
Re: Mini Medium Format Shootout
« Reply #99 on: November 22, 2010, 04:23:45 pm »

One of the main differences between the MFs is the software they come with... If you were doing a comparative test between 400hp super-cars, would you insist on testing them all with mini tyres?

In Photokit Sharpener "hi-res" is one category for all images above a certain resolution level as explained in the documentation. There is no received theory or practice about the merits or techniques of adjusting opacities for different sensor formats. The pixel pitch of a P40+ back is pretty much the same as the pixel pitch of a Pentax 645D sensor is pretty much the same as the pixel pitch for the 1Ds3, Leica's being larger but needing less sharpening out of the box than any of the others.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."
Pages: 1 ... 3 4 [5] 6 7 8   Go Up