Pages: 1 ... 4 5 [6] 7 8 ... 13   Go Down

Author Topic: 645D vs D3x  (Read 148091 times)

telyt

  • Guest
Re: 645D vs D3x
« Reply #100 on: February 15, 2011, 09:00:10 am »

... It is only since recent that 35mm are being able to be 14 bit ...

My 5-year-old Leica DMR is 16-bit and the tonal rendition is gorgeous.  I have no idea why CaNikon et al have been dragging their feet.
Logged

eronald

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6642
    • My gallery on Instagram
Re: 645D vs D3x
« Reply #101 on: February 15, 2011, 03:09:20 pm »

You need to set the Nikon to 14 bit before comparing it to something else.

However, the Kodak chips have more orthogonal color filters, while the others go for more ISO. This translates to faster dSLRs with worse color discrimination.

Edmund
Logged
If you appreciate my blog posts help me by following on https://instagram.com/edmundronald

telyt

  • Guest
Re: 645D vs D3x
« Reply #102 on: February 15, 2011, 04:11:40 pm »

... the Kodak chips have more orthogonal color filters, while the others go for more ISO. This translates to faster dSLRs with worse color discrimination....

This is something I've suspected.  Do you have links or other references that would explain this further?
Logged

ejmartin

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 575
Re: 645D vs D3x
« Reply #103 on: February 15, 2011, 04:44:09 pm »

My 5-year-old Leica DMR is 16-bit and the tonal rendition is gorgeous.  I have no idea why CaNikon et al have been dragging their feet.

Because 16-bit recording of a <12 bit dynamic range is rather pointless.  Nikon has finally gotten their pixel DR above 12 bits; Canon not yet (the data indicates a photosite DR of around 14 bits, but they aren't able to get it cleanly off the sensor).
Logged
emil

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: 645D vs D3x
« Reply #104 on: February 15, 2011, 06:28:04 pm »

Because 16-bit recording of a <12 bit dynamic range is rather pointless.  Nikon has finally gotten their pixel DR above 12 bits; Canon not yet (the data indicates a photosite DR of around 14 bits, but they aren't able to get it cleanly off the sensor).

Those 16 bits specs are close to false advertising.

They remind me of the DPI claims of Epson scanners. :)

Cheers,
Bernard

telyt

  • Guest
Re: 645D vs D3x
« Reply #105 on: February 15, 2011, 07:03:20 pm »

Those 16 bits specs are close to false advertising.

Is this opinion based on a mathematical model or on empirical observation?
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: 645D vs D3x
« Reply #106 on: February 15, 2011, 11:38:59 pm »

Physics and mathematics...

Try to put lenscap on and make an exposure at medium time, so you get no noise. Increase exposure in raw developing like four stops if you see any noise you don't have true sixteen bits. Reason: 16 bits is a ratio of 1:64000. Sensor probably holds around 60000 electrons, by and large corresponding to photons. So with true 16 bits there would be no noise. In practice there is something usually called readout noise, and that would typically be around 15 electrons or so. It takes 4 bits to count 15 electrons. So the range for meaningful signal would be around 12 bits.

Another simple test. Underexpose four steps, adjust exposure in raw development to compensate, the image would still be free of readout noise. And yes, the camera would excel at high ISOs.

I have no DMR so I cannot make these simple tests. Kodak has published readout noise for many of their sensors, so it may be possible to find out the exact data. DR (what the bits are needed for) is defined by Full Well Capacity and readout noise. DR is log2(FWC/readout noise). It happens also be the number of useful bits in the processing pipeline.

Best regards
Erik

Is this opinion based on a mathematical model or on empirical observation?
« Last Edit: February 15, 2011, 11:47:59 pm by ErikKaffehr »
Logged
Erik Kaffehr
 

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: 645D vs D3x
« Reply #107 on: February 15, 2011, 11:43:08 pm »

Hi,

Or my Dimage Scan Multi Pro having 16 bits corresponding to a density range of 4.8, in reality perhaps 3.2. With a DR of 4.8 scanning Velvia would be peace of cake, but it was not. Velvia has a density range just below 4.

Best regards
Erik

Those 16 bits specs are close to false advertising.

They remind me of the DPI claims of Epson scanners. :)

Cheers,
Bernard

Logged
Erik Kaffehr
 

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: 645D vs D3x
« Reply #108 on: February 16, 2011, 12:08:03 am »

Hi,

Something similar is discussed in this article:

http://www.dxomark.com/index.php/en/Our-publications/DxOMark-Insights/Canon-500D-T1i-vs.-Nikon-D5000/Color-blindness-sensor-quality

Which also gives detailed spectral data for the two sensors.

DxO Mark also has "Sensitivity metamerism index", It would have been 100 if sensor spectral sensivities matched human eye. So higher figure is better. This figure is based on the standard Xrite Color Checker patches. It may not be a perfect standard but developed by professionals in the color reproduction business.

It is:

75 for the Pentax 645D
79 for the Nikon D3X
80 for the Canon EOS 5DII
87 for the Sony Alpha 900
76 for the Leica M9
76 for the Phase One P65

The DMR is not listed, sorry.

The sensivity curves for the eye have much overlap for red and green (see below):

http://upload.wikimedia.org/wikipedia/commons/thumb/2/2f/Cone-fundamentals-with-srgb-spectrum.png/287px-Cone-fundamentals-with-srgb-spectrum.png

So sensors need some overlap on green and red, too.

Pleasant color is not the same as true color, would true color exist. So it is quite possible that one set of colors would be preferred for subjective photography, while another set of colors would be preferred for reproduction. In film days I used Velvia for landscapes, but it was really awful for portraits. The sensor just delivers three channel monochrome data, the color interpretation is done in raw development.

Comparing Miles's image i color balanced each for same subject area (White door on distant building) and got a much better color match. The Nikon image still has a lot more yellow in the greens. I'd consider it possible that it may depend on more overlap between red and green sensors, and also that it may be more truthful representation of the scene than the image from the Pentax.

Nikon shows color that is not very present in the Pentax, is Pentax missing colors or does Nikon exaggerate, I don't know. This illustrates a part of the problem. It is almost impossible to put the scene in a spectrometer and actually measure!

I also converted the images with the well regarded Iridient Raw Developer and the results were quite similar to ACR.

Best regards
Erik

This is something I've suspected.  Do you have links or other references that would explain this further?
« Last Edit: February 16, 2011, 12:58:11 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 #109 on: February 16, 2011, 05:17:34 am »

Is this opinion based on a mathematical model or on empirical observation?

It is based on the result of actual raw file analysis performed by some researchers in imaging active on this very forum.

There appears to be no usable information in the last 2 bits.

Cheers,
Bernard

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA
Re: 645D vs D3x
« Reply #110 on: February 16, 2011, 10:33:24 am »

Is this opinion based on a mathematical model or on empirical observation?

I measured the DR of the DMR and the 5D with Imatest years back now and while the 5D had more DR under the technical definition of ISO something close to 12 stops and the DMR had only 11.5 stops, once the threshold is changed to something more useful to photographers (Imatest draws several curves) the 5D's DR had dropped to 8 stops or under while the DMR didn't drop much at all - in summary handily out performing the 5D in real world testing not pie in the sky arm chair mathematical talk.    This is the really important thing to keep in mind.  A lot of discussion on this forum still quote DXO figures rather than actually measuring things and deal with idealized case rather than real world performance.   It's just too bad that those folks so interested in this stuff can't get their hands on the different cameras and make real life tests because then maybe they'd see what the people who use the cameras daily are talking about.
« Last Edit: February 16, 2011, 10:38:10 am by EricWHiss »
Logged
Rolleiflex USA

Dennis Carbo

  • Full Member
  • ***
  • Offline Offline
  • Posts: 163
Re: 645D vs D3x
« Reply #111 on: February 16, 2011, 10:53:41 am »

"A lot of discussion on this forum still quote DXO figures rather than actually measuring things and deal with idealized case rather than real world performance.   It's just too bad that those folks so interested in this stuff can't get their hands on the different cameras and make real life tests because then maybe they'd see what the people who use the cameras daily are talking about."



+ 1
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: 645D vs D3x
« Reply #112 on: February 16, 2011, 11:37:58 am »

On one hand, the claim that 16 bits has nor real advantage over 14 bits is not only supported by the experiments that Bernard quotes, but by the experimental data reported by the sensor makers themselves: both Kodak and Dalsa consistently report S:N ratios (engineering DR) values less than 8000:1, or about what 13 bits can encode, and both report dark noise levels of around 10 electrons or more, so that in 16 bit output, where there are about enough levels to count each electron, the last three bits or so are counting dark noise electrons.

On the other hand, what EricWHiss says is plausible:
"I measured the DR of the DMR and the 5D with Imatest years back now and while the 5D had more DR under the technical definition of ISO something close to 12 stops and the DMR had only 11.5 stops, once the threshold is changed to something more useful to photographers (Imatest draws several curves) the 5D's DR had dropped to 8 stops or under while the DMR didn't drop much at all..."

DXO has a measure of something like "stops of tonal range" which is significantly lower than its reported DR, and seems to be of more practical relevance.

Raising the minimum "useful" S:N level used to judge useful DR makes the dark noise floor less relevant at full exposure (low ISO speed) so that larger well capacity and thus better ratio of signal to shot noise can shift the results in favor of sensor with larger well capacity. As I have argued before, lowering dark noise below about 5 electrons will improve some engineering specs (and is vital for astronomy and useful for surveillance cameras and cameras with tiny photosites) but it contributes little or nothing to aspects of observed image quality relevant to high end MF or 35mm DSLR usage: controlling shot noise by detecting enough photons becomes the main way forward.
Logged

Gandalf

  • Full Member
  • ***
  • Offline Offline
  • Posts: 112
Re: 645D vs D3x
« Reply #113 on: February 16, 2011, 04:04:10 pm »

My 5-year-old Leica DMR is 16-bit and the tonal rendition is gorgeous.  I have no idea why CaNikon et al have been dragging their feet.

Doug, only people who have actually owned the DMR know what it is capable of. For everyone else, it is like trying to describe a rainbow to a colorblind dog.

Back to the original subject, after downloading and working with the raw files I have to say I am both surprised how at how good the D3x file is, and at the same time how much better the Pentax file it.
Logged

Dennis Carbo

  • Full Member
  • ***
  • Offline Offline
  • Posts: 163
Re: 645D vs D3x
« Reply #114 on: February 16, 2011, 04:43:48 pm »

It seems many of the forum discussions about 12 bit vs. 14 bit vs. 16 bit sensors yell and scream about DR.  ......Correct me if I am wrong but a 16 bit sensor records a greater range of colors / tones right ?  Seems like a newbie question but I am starting to doubt what I thought to be the case !  I always thought the smoother tones of a MFDB were largely due to them being 16 bit......is that correct or not ?

Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: 645D vs D3x
« Reply #115 on: February 16, 2011, 04:50:25 pm »

Hi,

My experience is that DR is very seldom a major factor in the kind of photography I normally do, landscape mostly. Would I need extreme DR the resort would be HDR photography. I did experiment with HDR but I could more often than not achieve similar results from correct ETR exposures. I do sometime church interiors and the experience still applies.

I essentially agree with Bill on well capacity and photo shot noise being the most important factor.

On the other hand I have seen a lot of real images demonstrating superior DR in the Pentax K5. Guiilermo Luik has developed some quite advanced tools for extending dynamic range and he really demonstrated how well the K5 handles deep underexposures.

Marc McCalmont demonstrated the K5 having excellent shadow detail compared with his P45+. I´d say that is real evidence, in the P45+ vs. K5 case raw images I evaluated myself. That's a bit hard to ignore. In theory shot noise is more important than readout noise but the K5 images by Marc McCalmont and Guillermo Gluik really indicate that the advantages are for real.

I have quite a lot of respect for Imatest, which I'm using myself, but so do I have for DxO data. The measurements are probably correct, but the interpretation may be misleading. I had some discussion with Mark Dubovoy on the issue and I got the impression that lenses and MTF play a significant role in perceived dynamic range. A god lens can simply produce more contrast on small detail than a bad lens. Similarly, larger pixels will have higher contrast because MTF is higher for lower frequencies. But this has nothing to do with CCD vs. CMOS or bit depth.

But, I see a real problem. People like Miles, Lloyd Chambers, Mark Dubovoy, Marc Mccalmont and Guillermo Luik present a lot of evidence in form or real images, some even in "raw". Some of the folks who are most knowledgeable like Emil Martinez don't use MFDB but humble DSLRs.

It is very hard to find any good raw images usable for evaluation on MFDBs, but pretty much on DSLRs.

I'd also add that it is not particularly easy to make good tests. To many variables around. I don't see anything wrong with evaluating test images made by others, especially not with images made by people knowing what they do. Of course the images must be raw. I have done such evaluations:

On Pentax 645D vs. Pentax D3X: http://echophoto.dnsalias.net/ekr/index.php/photoarticles/51-a-closer-look-at-pentax-645d-image-quality
On Leica S2 vs. Nikon D3X: http://echophoto.dnsalias.net/ekr/index.php/photoarticles/38-observations-on-leica-s2-raw-images
On Phase One P65+ vs. Canon 1dSIII http://echophoto.dnsalias.net/ekr/index.php/photoarticles/41-phase-one-images-for-download

I'm not a Nikon fanboy, but it seems that Nikon D3X used to be the champ in AA-filtered CMOS technology. So for me the issue was more about the effect of AA-filtering and how much can be gained by moving to larger formats.

My finding was that:

- Regarding sharpness the bigger format wins.
- Regarding noise and DR Phase One P65+ is much better than Canon 1DsIII. Pentax 645D is about on par with Nikon D3X and Nikon D3X is better than Leica S2.

This observations are based on the images I had access to. The observations are pretty much consistent with DxO data. As a side note the Canon 1DsIII has issues with noise in shadow detail at low ISO, it has definitively been passed by D3X and Pentax K5. That said Canon keeps noise down at high ISO due to preamp implementation (probable explanation).

It would be very nice if MFDB and DMR owners would post comparison raw images from their CCD-based cameras and modern top of the line CMOS based DSLRs like Nikon D3X and Pentax K5. The only comparisons I have seen not that is the original posting here by Miles Hecker and the P45+ vs. K5 comparison posted by Marc McCalmont. Please feel free to contribute images!

Best regards
Erik




On one hand, the claim that 16 bits has nor real advantage over 14 bits is not only supported by the experiments that Bernard quotes, but by the experimental data reported by the sensor makers themselves: both Kodak and Dalsa consistently report S:N ratios (engineering DR) values less than 8000:1, or about what 13 bits can encode, and both report dark noise levels of around 10 electrons or more, so that in 16 bit output, where there are about enough levels to count each electron, the last three bits or so are counting dark noise electrons.

On the other hand, what EricWHiss says is plausible:
"I measured the DR of the DMR and the 5D with Imatest years back now and while the 5D had more DR under the technical definition of ISO something close to 12 stops and the DMR had only 11.5 stops, once the threshold is changed to something more useful to photographers (Imatest draws several curves) the 5D's DR had dropped to 8 stops or under while the DMR didn't drop much at all..."

DXO has a measure of something like "stops of tonal range" which is significantly lower than its reported DR, and seems to be of more practical relevance.

Raising the minimum "useful" S:N level used to judge useful DR makes the dark noise floor less relevant at full exposure (low ISO speed) so that larger well capacity and thus better ratio of signal to shot noise can shift the results in favor of sensor with larger well capacity. As I have argued before, lowering dark noise below about 5 electrons will improve some engineering specs (and is vital for astronomy and useful for surveillance cameras and cameras with tiny photosites) but it contributes little or nothing to aspects of observed image quality relevant to high end MF or 35mm DSLR usage: controlling shot noise by detecting enough photons becomes the main way forward.
« Last Edit: February 16, 2011, 04:57:06 pm by ErikKaffehr »
Logged
Erik Kaffehr
 

LKaven

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1060
Re: 645D vs D3x
« Reply #116 on: February 16, 2011, 05:36:33 pm »

It seems many of the forum discussions about 12 bit vs. 14 bit vs. 16 bit sensors yell and scream about DR.  ......Correct me if I am wrong but a 16 bit sensor records a greater range of colors / tones right ?  Seems like a newbie question but I am starting to doubt what I thought to be the case !  I always thought the smoother tones of a MFDB were largely due to them being 16 bit......is that correct or not ?
There is no camera made today capable of capturing more than 14 "good" bits of signal, and the 14th bit is usually no more than a dither bit.  The D3x and the Phase P65+ are about on a par with 13.7 stops of dynamic range, both a good bit better than the 5DII.  Cameras introduced this year from Nikon and/or Canon may push a little higher. 

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: 645D vs D3x
« Reply #117 on: February 16, 2011, 05:53:59 pm »

Hi,

+1

The advantage of MFDBs may not be 16 bits but larger sensor area thus collecting more photons.

These two posting may indicate how good todays DSLRs are:

http://www.luminous-landscape.com/forum/index.php?topic=50895.msg420348#msg420348
http://www.luminous-landscape.com/forum/index.php?topic=50895.msg421886#msg421886

Same images but different aspects. The K5 is an APS-C camera from Pentax and the P45+ e pretty up to date Phase One with CCD sensor.

And yes, I cannot explain why the K5 sensor is so good as it seems but have seen a lot of convincing evidence in it's favor.

Best regards
Erik


There is no camera made today capable of capturing more than 14 "good" bits of signal, and the 14th bit is usually no more than a dither bit.  The D3x and the Phase P65+ are about on a par with 13.7 stops of dynamic range, both a good bit better than the 5DII.  Cameras introduced this year from Nikon and/or Canon may push a little higher. 
Logged
Erik Kaffehr
 

Dennis Carbo

  • Full Member
  • ***
  • Offline Offline
  • Posts: 163
Re: 645D vs D3x
« Reply #118 on: February 16, 2011, 07:07:12 pm »

Ok...so just to clarify....a 16 Bit capture from a MFDB has the same amount of color information as a 14 bit capture from a DSLR ? 
Logged

ejmartin

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 575
Re: 645D vs D3x
« Reply #119 on: February 16, 2011, 07:24:38 pm »

One of the things that DxO does not measure that can visibly affect usable DR and image quality is pattern noise; it is the main reason why low ISO shadows on a D3x are dramatically better than a 1Ds3 which is dramatically better than a 5D2.  Even though the 5D2 and 1Ds3 measure for all practical purposes to have the same DR on DxO.
Logged
emil
Pages: 1 ... 4 5 [6] 7 8 ... 13   Go Up