Pages: 1 ... 3 4 [5] 6 7 ... 13   Go Down

Author Topic: 645D vs D3x  (Read 148082 times)

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA
Re: 645D vs D3x
« Reply #80 on: February 14, 2011, 12:52:18 am »

Erik,
Yes of course I did use the live view for focusing the 5D2 in the comparison I did.  I shot both cameras with the Leica 100 apo macro lens.  Frankly I was surprised to see how good the DMR camera was. I don't have the RAWs to share anymore but I did post crops to one of the forums - its been a while though.... maybe Fred Miranda?  Since the lens was the same, I could only conclude it had to do with sensor or processing.  I used C1 Pro for both images.   I think the kodak chip for the DMR has microlenses but you'd have to double check that.   My feeling still is that it was the AA filter that made the difference and I am talking about very fine detail. 

In any case this is a long ways from the OT  - sorry for the side track.   I do conclude that the 645D was better than the D3x but not sure that difference has so much to do with AA filters or lack thereof.
Logged
Rolleiflex USA

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: 645D vs D3x
« Reply #81 on: February 14, 2011, 12:55:42 am »

I do conclude that the 645D was better than the D3x but not sure that difference has so much to do with AA filters or lack thereof.

At F11 the difference between an AA filter and lack of, is not significant.
Logged

Dennis Carbo

  • Full Member
  • ***
  • Offline Offline
  • Posts: 163
Re: 645D vs D3x
« Reply #82 on: February 14, 2011, 09:18:02 am »

Ray ,

I looked at your samples - nice job processing. I see very little of the "mushiness" i usually see from a dslr, files are actually very, very close. I am not shooting with a D3X either, so perhaps some of my perception of "Mushiness" is due to resolution ?(16mp dslr vs 22mp mfdb). So ...I have come to this conclusion....I may be biased ! ;D   I am comparing  shots done with a Sinar 54M and Schnieder Super Angulon 40/3.5  to a D7000 and NIKON 12-24. It probably comes down to the glass more than anything in my case - The Schnieder lens is in amazing, the nikon is fair at best IMO. I have rented a D3X in the past but probably didnt spend as much time as I should have with it.  I do think the 645D files were better than the D3X but it is closer than I would have thought.

Regards,

Dennis
Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: 645D vs D3x
« Reply #83 on: February 14, 2011, 09:46:49 am »

Ray ,

I looked at your samples - nice job processing. I see very little of the "mushiness" i usually see from a dslr, files are actually very, very close.

There really is nothing mushy about the d3x when correct sharpening is applied.

The full size sample below is shot handheld with a 300 f2.8 VR... the helicopter is moving.

http://www.flickr.com/photos/bernardlanguillier/3900647680/sizes/o/in/set-72157622461555672/

Cheers,
Bernard

Peter Devos

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 518
Re: 645D vs D3x
« Reply #84 on: February 14, 2011, 10:34:32 am »

Hi Bernard, the helicopter image to me is the exact reason why to use MF digital back, no mather if its a 16, a 22 or a 50Mp back. I am sure you will disagree with me but then i would symply ask you to shoot these things with a digital back and with a Conon/Nikon. Oh yes, i do not mean you have to use such a long telephoto lens as this oes not exist on MF but just the rendering of different tones and shades is so uggly, 35mm filmlike. With all respect, its a great shot and i really like your photography but i think there are to many things said over here abouth MF that simply aren't true or at least biased towards Nikon/Canon. I shoot with Nikon/Canon and sony 850 and none of them has the smoothness of any of my MF gear. Just my 2cents.
Logged

telyt

  • Guest
Re: 645D vs D3x
« Reply #85 on: February 14, 2011, 10:59:03 am »

... the helicopter image to me is the exact reason why to use MF digital back, no mather if its a 16, a 22 or a 50Mp back. I am sure you will disagree with me but then i would symply ask you to shoot these things with a digital back and with a Conon/Nikon. Oh yes, i do not mean you have to use such a long telephoto lens as this oes not exist on MF but just the rendering of different tones and shades is so uggly, 35mm filmlike.

I agree, the tonal rendition in this photo is quite poor compared with what I'd expect from MF digital.
Logged

Dennis Carbo

  • Full Member
  • ***
  • Offline Offline
  • Posts: 163
Re: 645D vs D3x
« Reply #86 on: February 14, 2011, 11:29:08 am »

Hi Bernard,

The mushiness I was referring to is what I usually see on branches/foliage and fine detail areas like grasses. it is similar to what u see if you get heavy handed with noise reduction.  You dont see much it until you compare it to a MFDB file - as I said though I may be biased as my MF rig is of far better quality glass wise than my Nikon rig

I do quite a few Aerials as well and find the D7000 fast and easy to handle with a 80-200 / 2.8 it does a great job, and is better suited for rough use inside a C-172 than my MF rig !

Nice copter shot

D

Logged

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA
Re: 645D vs D3x
« Reply #87 on: February 14, 2011, 11:31:14 am »

Yes, agree with the above posters.  MF has a smoother, deeper, richer look not just better detail.  I didn't think so myself until I made the switch. I find a lot of the discussion in these threads interesting but as Peter Devos suggests, there is no substitute for just working with the different platforms and seeing for yourself.  
Logged
Rolleiflex USA

Peter Devos

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 518
Re: 645D vs D3x
« Reply #88 on: February 14, 2011, 12:26:24 pm »

I still have a leaf Cantare XY camera back and i dare any D3x/5dMk2/1ds3/A900/A850 to shoot a setup against the Cantare( or Imacon 3020/Phase lightphase). They are only 6Mp, but man... what a great 6 Mp. When not using both systems it is futile to express an opinion abouth MF. These people only misinform possible users because of their " extremely limited internet exerience ( mostly from analising someone elses shot). Just get a MF set for a weekend and really try and use it as it is suppose d to... then you will see why so much pro's are using MF.
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: 645D vs D3x
« Reply #89 on: February 14, 2011, 01:19:42 pm »

Hi Eric,

What I have seen from Miles pictures the Pentax has a definitive advantage in absolute resolution. I made prints from 1/3 crops corresponding to 53x83 cm and the sharpness advantage of the Pentax is quite obvious when pixel peeping, but seeing the prints at say 80 cm the difference is not obvious to me, with my eyesight. I tried to match colors and tonality.

The colors are quite different but I´d suggest that may be a white balance issue. I rebalanced on a hopefully white door on a distant building and the color were much closer. The difference is still a bit to much for my taste.

Looking at shadow detail I'd say the Pentax image has a tiny advantage.

The Pentax having better resolution comes as no surprise as it has a bigger sensor with more pixels. This is also obvious from testing I have done using Imatest, see enclosed figure. The Imatest calculation also shows that the Pentax 645D transmits an unhealthy amount of contrast at Nyquist. Another image from the same test also shows a healthy amount of Moiré.

Regarding your finding on the DMR vs. the Canon 5DII it really seems that you have eliminated the obvious problem areas. What is pretty certain that the lens you have would produce a significant amount of false detail on a sensor with 6.8 micron pitch (like the DMR) unless stopped down to at least f/8 (Airy ring diameter is around 8 microns at f/8). So I have no good explanation. Thanks for sharing your experience! That's not sidetracking, BTW.

There was a posting recently by Marc McCalmont comparing his P45+ to his K5. The K5 seems to have better shadow detail than the P45+. The impression I have is that MFDBs are said to have magic properties. Both the Pentax tests and the Marc's image contradict this.
http://www.luminous-landscape.com/forum/index.php?topic=50895.0

There are several posters on the forum who have both MFDBs and DSLRs and don't really see the magic, like Jeff Schewe, Marc and Eronald. They acknowledge the quality but not the magic.

Best regards
Erik





Erik,
Yes of course I did use the live view for focusing the 5D2 in the comparison I did.  I shot both cameras with the Leica 100 apo macro lens.  Frankly I was surprised to see how good the DMR camera was. I don't have the RAWs to share anymore but I did post crops to one of the forums - its been a while though.... maybe Fred Miranda?  Since the lens was the same, I could only conclude it had to do with sensor or processing.  I used C1 Pro for both images.   I think the kodak chip for the DMR has microlenses but you'd have to double check that.   My feeling still is that it was the AA filter that made the difference and I am talking about very fine detail.  

In any case this is a long ways from the OT  - sorry for the side track.   I do conclude that the 645D was better than the D3x but not sure that difference has so much to do with AA filters or lack thereof.
« Last Edit: February 14, 2011, 02:04:09 pm by ErikKaffehr »
Logged
Erik Kaffehr
 

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA
Re: 645D vs D3x
« Reply #90 on: February 14, 2011, 02:09:59 pm »


There was a posting recently by Marc McCalmont comparing his P45+ to his K5. The K5 seems to have better shadow detail than the P45+. The impression I have is that MFDBs are said to have magic properties. Both the Pentax tests and the Marc's image contradict this. There are several posters on the forum who have both MFDBs and DSLRs and don't really see the magic, like Jeff Schewe, Marc and Eronald. They acknowledge the quality but not the magic.

Best regards
Erik



Erik,
Well what can I say except have you shot any of your own work on a MF camera with a digital back?   If you have to rely on what others post and think, you can't be sure yourself.

Look,  every piece of gear will have its strengths and it might well be that a person could find a test shot scenario where one camera will come out on top. For example if I wanted to show that then canon 5D2 was better than a MF camera, I'd simple take a shot at ISO 3200 or one requiring auto focus.   In those limited aspects one could say the DSLR was better.   And in the p45+ v K5 image - sure the shadows could be lifted, but that's an exercise only for peepers.  Which image looked better as posted by the OP? I'm sure that in most other examples and criteria the p45 would win easily. The point of the OP was to show the strength of the K5 and he did so. What would it look like if he wanted to show the strength of the p45?

Eric
Logged
Rolleiflex USA

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: 645D vs D3x
« Reply #91 on: February 14, 2011, 03:33:37 pm »

Hi,

I would like to test an MFDB, but there are a couple of issues.

1) Cost, renting a top MFDB for a couple of days like buying a Sony Alpha 850

2) Making good test shots is in no way easy, so even if I rented an MFDB a couple of days I would not know if I utilized it optimally

Also, I cannot see what is wrong with using other peoples tests. Diglloyd essentially has testing as his business, it is very probable he makes better test images than I would. Miles Hecker's images are made with considerable care. Why would I think that I would devise better tests?

Regarding Marc's image, it's clear that he makes a statement. But it is routinely said that DR on MFDBs is several stops wider than on DSLRs, Marc's image clearly contradicts this, at least for shadow detail. I don't have any issue with MFDBs, I sort of considered buying both the Mamya ZD and the Pentax 645D and I never considered buying a Nikon D3X.

What I'm doubtful about are statements that don't match physics, like MFDBs having better color, better tonality, significantly more DR.

I have seen samples from Phase One clearly indicating better DR than Canon 1DsIII and Miles's sample here really indicate quite different color between P645D and D3X. Nevertheless, I don't believe in magic. If there is a difference there must be some logical explanation to it.

Best regards
Erik

Erik,
Well what can I say except have you shot any of your own work on a MF camera with a digital back?   If you have to rely on what others post and think, you can't be sure yourself.

Look,  every piece of gear will have its strengths and it might well be that a person could find a test shot scenario where one camera will come out on top. For example if I wanted to show that then canon 5D2 was better than a MF camera, I'd simple take a shot at ISO 3200 or one requiring auto focus.   In those limited aspects one could say the DSLR was better.   And in the p45+ v K5 image - sure the shadows could be lifted, but that's an exercise only for peepers.  Which image looked better as posted by the OP? I'm sure that in most other examples and criteria the p45 would win easily. The point of the OP was to show the strength of the K5 and he did so. What would it look like if he wanted to show the strength of the p45?

Eric

Logged
Erik Kaffehr
 

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA
Re: 645D vs D3x
« Reply #92 on: February 14, 2011, 04:45:46 pm »

Erik,
If you aren't going to ever use a MF camera why do you waste so much time on the forums about it?   I really mean it that you won't know what some of us are talking about until you use one yourself.  
Eric
« Last Edit: February 14, 2011, 06:26:19 pm by EricWHiss »
Logged
Rolleiflex USA

telyt

  • Guest
Re: 645D vs D3x
« Reply #93 on: February 14, 2011, 06:15:16 pm »


What I'm doubtful about are statements that don't match physics, like MFDBs having better color, better tonality, significantly more DR.

I have seen samples from Phase One clearly indicating better DR than Canon 1DsIII and Miles's sample here really indicate quite different color between P645D and D3X. Nevertheless, I don't believe in magic. If there is a difference there must be some logical explanation to it.

I don't believe in magic either.  If the logic available to us doesn't explain observed differences then perhaps it's best to re-examine the logic.  In my day job I'm creating mathematical models of physical systems.  Where differences are observed between empirical data and mathematical models and in the empirical tests the variables have been adequately controlled, the differences are most likely caused by a weakness in the mathematical model, simplifying assumptions or inadequate understanding of the physics being the most likely culprits.  A logical explanation will always exist, but we might not have enough knowledge or have included enough parameters in our equations to explain it.
Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: 645D vs D3x
« Reply #94 on: February 14, 2011, 06:38:11 pm »

No worries, this image was for sure tuned for detail, tones are OK in print.

Cheers,
Bernard

spqr

  • Newbie
  • *
  • Offline Offline
  • Posts: 5
Re: 645D vs D3x
« Reply #95 on: February 14, 2011, 10:42:56 pm »

Erik,
Well what can I say except have you shot any of your own work on a MF camera with a digital back?   If you have to rely on what others post and think, you can't be sure yourself... [snip]  And in the p45+ v K5 image - sure the shadows could be lifted, but that's an exercise only for peepers.  Which image looked better as posted by the OP? I'm sure that in most other examples and criteria the p45 would win easily.

Ah now, the same advice might be taken in regards to the K-5. We do know that the K-5 dynamic range at low ISO exceeds all current MF and FF cameras, though that is unlikely to remain true for very long. So, in terms of shadow detail, I would expect the K-5 to fair very well in comparison. Mind you, as you note, this isn't the whole picture. It is important, though, to note that the technology around digital photography is not standing still and is moving far faster in the 35mm format than it is in the medium format market, that's simple economics, just as P&S is moving faster than the 35mm market. I think a lot of old opinions on this topic are either going to change or simply be shown to be incorrect in the face of stubborn reality. So, at the current rate, the 35mm form factor will surpass MF in IQ, the MF sensor technology is simply not keeping up, so the question is when.
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: 645D vs D3x
« Reply #96 on: February 15, 2011, 12:13:20 am »

Hi

I agree. Except that there is the definite possibility of the observation being wrong. Do you have any good explanation that we don't see fake detail in images with very high resolving lenses, large pixels and no AA-filtering? I have a few theories, please add yours:

1) Fake detail can be mistaken for real one

2) Stopping down reduces aliasing. Diglloyd indicated that stopping down to f/11 eliminate Moiré on Leica S2 images. The Airy circle diameter will be around 11 microns (depending on wavelength of light) at f/11 so it pretty much corresponds to twice the pixel pitch (6 microns on the S2).

3) Less than perfect focus would also reduce fake detail

My guess is that the first point is most important. We see fake detail and perceive it as real detail.

How detrimental is AA-filtering? That is another question, very well possible that the disadvantages outweigh the benefits. I see smaller pixels as the solution. That also unfortunately decreases DR both on the pixel level and when averaged over area. Binning will not reduce aliasing either, if done before demosaic. (As far as I understand)

A third question is how much of the edge contrast lost to AA-filtering can be regained with sharpening. Excessive sharpening will induce fake detail and enhance noise.

Best regards
Erik


I don't believe in magic either.  If the logic available to us doesn't explain observed differences then perhaps it's best to re-examine the logic.  In my day job I'm creating mathematical models of physical systems.  Where differences are observed between empirical data and mathematical models and in the empirical tests the variables have been adequately controlled, the differences are most likely caused by a weakness in the mathematical model, simplifying assumptions or inadequate understanding of the physics being the most likely culprits.  A logical explanation will always exist, but we might not have enough knowledge or have included enough parameters in our equations to explain it.
« Last Edit: February 15, 2011, 12:49:04 am by ErikKaffehr »
Logged
Erik Kaffehr
 

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: 645D vs D3x
« Reply #97 on: February 15, 2011, 12:42:05 am »

1) Fake detail can be mistaken for real one

Yes, I believe that this is very true.

Cheers,
Bernard

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: 645D vs D3x
« Reply #98 on: February 15, 2011, 01:01:19 am »

Hi,

This is a very good example of fake detail. It's from the Sigma DP2 having a Foevon sensor with relatively large pixels (7.8 micron pitch)

Note that no lines are resolved at the 18 mark but the lines reemerge below. They look very real but they are absolutely fake.

I have not seen it on DPReviews test of the Leica M8, but the image there was JPEG so it was preprocessed in camera.

Best regards
Erik



Yes, I believe that this is very true.

Cheers,
Bernard

Logged
Erik Kaffehr
 

kers

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4388
    • Pieter Kers
Re: 645D vs D3x
« Reply #99 on: February 15, 2011, 08:24:50 am »

Hi Bernard, the helicopter image to me is the exact reason why to use MF digital back, no mather if its a 16, a 22 or a 50Mp back. I am sure you will disagree with me but then i would symply ask you to shoot these things with a digital back and with a Conon/Nikon. Oh yes, i do not mean you have to use such a long telephoto lens as this oes not exist on MF but just the rendering of different tones and shades is so uggly, 35mm filmlike. With all respect, its a great shot and i really like your photography but i think there are to many things said over here abouth MF that simply aren't true or at least biased towards Nikon/Canon. I shoot with Nikon/Canon and sony 850 and none of them has the smoothness of any of my MF gear. Just my 2cents.

I disagree about 35mm not being able to be smooth.
I think it has a lot to do with the sensor but also with lenses and digital treatment.
Most 35mm images we usually see have other goals than smoothness - especially when used for press purposes.
Smoothness is clearly not the goal in the helicopter image.
It is only since recent that 35mm are being able to be 14 bit and that 24MP is possible. We will see that it will be able to be 16 bit and 40MP in the near future.
Here an image made with my 35mm Nikon d3x were smoothness is vital and achieved...
« Last Edit: February 15, 2011, 04:56:02 pm by kers »
Logged
Pieter Kers
www.beeld.nu/la
Pages: 1 ... 3 4 [5] 6 7 ... 13   Go Up