645D vs 1DSmk3 vs D3X

[BernardLanguillier]

Maybe I am misreading you, but isn't the obvious reason that there are more pixels in same section the Pentax image because....there are more pixels in the Pentax sensor? 

The point is not to compare pixel-to-pixel, but rather similarly frame subject with similar perspective lens.

There should be approximately 27% more pixel in the Pentax image than in the D3x (approximately SQRT(40/24.5) - ignoring the different aspect ratio) for a similar framing of the image.

If the gap is larger, then this is not a fair comparison and gives only limited indication of the actual advantage of the 645D in terms of pixel quality and less information than it should in terms of actual image quality.

Cheers,
Bernard

[EricWHiss]

There should be approximately 27% more pixel in the Pentax image than in the D3x (approximately SQRT(40/24.5) - ignoring the different aspect ratio) for a similar framing of the image.

If the gap is larger, then this is not a fair comparison and gives only limited indication of the actual advantage of the 645D in terms of pixel quality and less information than it should in terms of actual image quality.

Cheers,
Bernard

Bernard,
The logic of going pixel to pixel is like wondering if the dollar bill in any joe's wallet is the same as the one in Bill Gates' wallet, completely missing the other differences.  I maybe wrong but I feel most people are looking at this comparison to see which camera in a price range can deliver the best files. The prices of the D3x, 1Ds3, and the 645D are close enough to make a comparison regardless of # of pixels each has.  And who said it had to be fair anyhow? 
Eric

[bjanes]

Hi All,

See http://www.imaging-resource.com/PRODS/645D/645DA.HTM

The 645D betters the 35mm champs.   
Look, download and see for yourself.

Enjoy,
Miles

Welcome to the discussion, Miles. For others, I would like to point that mheckler is listed as a newbie on this forum, but he is a well known and respected photographer. While reading the Road to Taos article Miles recently wrote for LuLa, I visited his web site and also noted that he has written an interesting article comparing dSLRs to MFDB--at least on paper. He mentioned the Pentax 645 but hasn't actually handled one.

[ErikKaffehr]

Bernard,

I downloaded some appropriate test images from Imaging Resource and measured MTF50 with Imatest. I got the following results:

Pentax 645D: 4605 LW/PH
Nikon D3X: 3057 LW/PH

So the Pentax gives something like 50% better resolution. Some of the improvement is probably because of better lens, no OLP filter and so on. I made a comparison of the bottle images scaled to 50x70 cm and 70x100 cm. On 50x70 cm file very little difference was seen but the 70x100 file was clearly better on the Pentax.

The way I do these comparisons I upscale the image so I can make a 50x70 (or 70x100) print at 200 pixels/inch. I choose 200 line/inch because that is what my pro lab defaults to at these sizes. I'm not really sure about pixels/inch on the 70x100cm, could be 360 pixels/inch although it seems they are 200 pixels/inch, too.

I don't know if Imaging review had the best lens on the Nikon D3X. The EXIF says 70/2.8. I'd presume 24-70/2.8 at 70 mm.

Best regards
Erik

Best regards
Erik

There should be approximately 27% more pixel in the Pentax image than in the D3x (approximately SQRT(40/24.5) - ignoring the different aspect ratio) for a similar framing of the image.

If the gap is larger, then this is not a fair comparison and gives only limited indication of the actual advantage of the 645D in terms of pixel quality and less information than it should in terms of actual image quality.

Cheers,
Bernard

[BernardLanguillier]

Bernard,
The logic of going pixel to pixel is like wondering if the dollar bill in any joe's wallet is the same as the one in Bill Gates' wallet, completely missing the other differences.  I maybe wrong but I feel most people are looking at this comparison to see which camera in a price range can deliver the best files. The prices of the D3x, 1Ds3, and the 645D are close enough to make a comparison regardless of # of pixels each has.  And who said it had to be fair anyhow? 

Eric,

All I am saying is that if you want to get useful data from a comparison of images online, they have to have been captured in comparable conditions, among which the framing. That's what I mean by fair. This is nothing but basic scientific experiment theory. You need to take that into account if the goal is to gather useful comparative information.

I expect the 645D to be clearly superior to the D3x when things are done perfectly. As mentioned before though, I believe that it will be much harder to get things perfect with the 645D because of the lack of live view.

So the real useful comparison would be to take 10 images with both cameras in real life situation and compare the actual image quality including that of images that are not optimally focused.

Cheers,
Bernard

[ErikKaffehr]

Ah,

Miles is WyoFoto! Lucky guy to have a lot of Pentax lenses!

Best regards
Erik

Welcome to the discussion, Miles. For others, I would like to point that mheckler is listed as a newbie on this forum, but he is a well known and respected photographer. While reading the Road to Taos article Miles recently wrote for LuLa, I visited his web site and also noted that he has written an interesting article comparing dSLRs to MFDB--at least on paper. He mentioned the Pentax 645 but hasn't actually handled one.

[Bart_van_der_Wolf]

Bernard,

I downloaded some appropriate test images from Imaging Resource and measured MTF50 with Imatest. I got the following results:

Pentax 645D: 4605 LW/PH
Nikon D3X: 3057 LW/PH

So the Pentax gives something like 50% better resolution. Some of the improvement is probably because of better lens, no OLP filter and so on.

Hi Erik,

That's totally cosistent with the expectations, no surprise there. The 645D has almost 35% more vertical pixels, so an increase in Line Widths per Picture Height of 35% would be expected, but the lack of an AA-filter boosts the MTF response even further. The reponse at and beyond Nyquist shows its potential for aliasing.

I made a comparison of the bottle images scaled to 50x70 cm and 70x100 cm. On 50x70 cm file very little difference was seen but the 70x100 file was clearly better on the Pentax.

The way I do these comparisons I upscale the image so I can make a 50x70 (or 70x100) print at 200 pixels/inch. I choose 200 line/inch because that is what my pro lab defaults to at these sizes. I'm not really sure about pixels/inch on the 70x100cm, could be 360 pixels/inch although it seems they are 200 pixels/inch, too.

Yes, that's a sensible test from a usability standpoint. The outcome is not surprising because, as I've said from the beginning, more pixels pay off when enlarging.

I don't know if Imaging review had the best lens on the Nikon D3X. The EXIF says 70/2.8. I'd presume 24-70/2.8 at 70 mm.

There are other factors to consider as well, depending on one's subject choice. To reach a certain DOF on the 645D one may need to stop down further, and the effects of diffraction will become visible at the pixel level with apertures narrower than f/6.3, and a larger sensor array requires more exposure, so also shutter speed might become relatively lower unless one can add extra light. Live View and/or tethering is often a critical need, because nothing kills MTF as fast as defocus. So handling (e.g. think about portrait orientation) and placements of controls becomes important as well.

So the demonstration of potential (I'd rather call it that than a comparison or test, due to the different platforms) is nice, but it doesn't tell the whole story ...

Cheers,
Bart

[ziocan]

I think some of the 35mm die hard have difficulty to digest the evidence and are finding any kind of excuse to deny the fact that once the focusing is there: more is more.

[ErikKaffehr]

Hi,

The LW/PH figure I measured was a little bit on the low side for a truly excellent lens, correctly focused, on the D3X. Your other observations are absolutely valid.

I sort of presume that the camera is sitting on top a decent tripod and MLU is used if we talk resolution.

Best regards
Erik

Hi Erik,

There are other factors to consider as well, depending on one's subject choice. To reach a certain DOF on the 645D one may need to stop down further, and the effects of diffraction will become visible at the pixel level with apertures narrower than f/6.3, and a larger sensor array requires more exposure, so also shutter speed might become relatively lower unless one can add extra light. Live View and/or tethering is often a critical need, because nothing kills MTF as fast as defocus. So handling (e.g. think about portrait orientation) and placements of controls becomes important as well.

So the demonstration of potential (I'd rather call it that than a comparison or test, due to the different platforms) is nice, but it doesn't tell the whole story ...

Cheers,
Bart

[BernardLanguillier]

I think some of the 35mm die hard have difficulty to digest the evidence and are finding any kind of excuse to deny the fact that once the focusing is there: more is more.

Why would anyone in his/her right mind resist the obvious fact that a 40MP 645D without AA filter has the potential to deliver significantly more resolution than a high end DSLR when focused properly?

Beyond that, why would anyone want to be a 35mm die hard??? I think most photographers look at their actual needs and budget and take a rational decision on what tool meets their minimal criteria for performance while offering a good price/performance ratio. It will be 35mm for some, MF for others.

There are objective information sources around to take such rational decisions.

Cheers,
Bernard

[ErikKaffehr]

Hi Bernard,

It seems to me that the Pentax 645D delivers. What it delivers is MF performance at nearly D3X price. That of course rises the question of lenses. Those lenses Nick tested seem to be pretty good.

Another question is how a future Nikon (D4X?) will perform with Zeiss lenses and other high end primes? We see that Canon, Nikon and Sony introduce APS-C cameras with 16-18 MP sensors and that sensor pitch would be around 40 MP on full frame. So I guess that kind of resolution is on the horizon. My guess is that present day zooms may lack resolution across the field to match a 4.5 micron pitch sensor, but high quality primes can perhaps utilize a sensor fully. Anyway smaller pitches mean "weaker" OLP filter.

Another question is the market place. How many actors can we have in the MF marketplace? The Pentax 645D is in a sense very similarly positioned to the Leica S2.
Lloyd Chambers will soon do some testing on the Pentax 645D and he has done a lot of shooting with the Leica S2. No doubt, some will buy the Leica for the magic "red dot", but I got the impression from that it has some real advantages.

On the other hand I got the impression from Lloyd's and your writing that at least some Zeiss lenses really deliver, once correctly focused.

Anyway, it seems that future is bright for photographers. Doesn't always feel that way in the Swedish winter.

Best regards
Erik

Why would anyone in his/her right mind resist the obvious fact that a 40MP 645D without AA filter has the potential to deliver significantly more resolution than a high end DSLR when focused properly?

Beyond that, why would anyone want to be a 35mm die hard??? I think most photographers look at their actual needs and budget and take a rational decision on what tool meets their minimal criteria for performance while offering a good price/performance ratio. It will be 35mm for some, MF for others.

There are objective information sources around to take such rational decisions.

Cheers,
Bernard

[BernardLanguillier]

Hi Bernard,

It seems to me that the Pentax 645D delivers. What it delivers is MF performance at nearly D3X price.

Yes, it does. Leaving economics, existing investments and personal preferences aside, it will be a no brainer for many of the landscape photographers confident in their ability to focus optimally a camera through its viewfinder, which should be the case for all existing MF owners (if not mine).

This remarkable achievement reflects the amount of money invested in the project by Pentax and a lot of very hard work, this itself resulting from a smart business model where volume funds R&D costs across product lines.

Besides the lack of live view, another problem with the Pentax system today is the lack of T/S lenses though. I have already mentioned this to them (along with the live view issue) and they are aware of these 2 limitations, acknowledge them, and are trying to find solutions.

I agree with your point that the 1ds4/A1000/D4x are not very far away. I will personally probably wait until the 2nd generation of the 645D, see if it does have live view and T/S lenses. If Pentax can pull that, chances are that I will sell most of my Nikon gear. If they don't, I might or might not buy a D4x that I would probably keep for 10 years or so. The D3x is already so damned close to perfection that even that will be a hard call.

Cheers,
Bernard

[bjanes]

Bernard,

I downloaded some appropriate test images from Imaging Resource and measured MTF50 with Imatest. I got the following results:

Pentax 645D: 4605 LW/PH
Nikon D3X: 3057 LW/PH

Eric,

Interesting results. I don't know which images you downloaded, but for some testing on own, I downloaded D3XhRES6048F for the D3x and 645DhRES7264F.JPG for the 645. My Imatest results are shown below. The 645 resolution varies in the horizontal and vertical axis, so I chose the vertical for testing.

The problem is that the 645 image does not appear to be sharpened and the D3X image appears oversharpened to my eye, even though Imatest reports slight undersharpening. My results for MTF50 are shown:

D3x: 2750 lw/ph unsharpened, and 2818 lw/ph with Imatest standard sharpening.
645: 3021 lw/ph unsharpened, and 3773 lw/ph with Imatest standard sharpening.

Sharpening is said to be less critical with cameras lacking a low pass filter, but the 645 does appear to need sharpening. The D3x has a slightly lower MTF at Nyquist, consistent with the low pass filter, but there appear to be a lot of artifacts beyond Nyquist, which I would consider undesirable.

I think that the characteristics of the two sensors are so different that it would be difficult to equalize the variables with Imatest, where oversharpening can lead to MTF50 well beyond Nyquist consistent with artifacts. Ideally, one would use deconvolution sharpening for the D3x, but I don't see a raw file. Looking at the rendered images, the 645 appears smoother, but the magnification is greater as Bart pointed out.

Regards,

Bill

[ErikKaffehr]

Bill,

Thanks for comments.

The results you got were quite different from mine. I was working with raw images and converted with Lightroom with standard sharpening.

The images I used were:

D3XhMULTI0100 (100 ISO)
645FhRES7264F  (200 ISO)

My figures were for standardized sharpening. I'm going to recheck my results.

Best regards
Erik

Eric,

Interesting results. I don't know which images you downloaded, but for some testing on own, I downloaded D3XhRES6048F for the D3x and 645DhRES7264F.JPG for the 645. My Imatest results are shown below. The 645 resolution varies in the horizontal and vertical axis, so I chose the vertical for testing.

The problem is that the 645 image does not appear to be sharpened and the D3X image appears oversharpened to my eye, even though Imatest reports slight undersharpening. My results for MTF50 are shown:

D3x: 2750 lw/ph unsharpened, and 2818 lw/ph with Imatest standard sharpening.
645: 3021 lw/ph unsharpened, and 3773 lw/ph with Imatest standard sharpening.

Sharpening is said to be less critical with cameras lacking a low pass filter, but the 645 does appear to need sharpening. The D3x has a slightly lower MTF at Nyquist, consistent with the low pass filter, but there appear to be a lot of artifacts beyond Nyquist, which I would consider undesirable.

I think that the characteristics of the two sensors are so different that it would be difficult to equalize the variables with Imatest, where oversharpening can lead to MTF50 well beyond Nyquist consistent with artifacts. Ideally, one would use deconvolution sharpening for the D3x, but I don't see a raw file. Looking at the rendered images, the 645 appears smoother, but the magnification is greater as Bart pointed out.

Regards,

Bill

[eronald]

I agree with your point that the 1ds4/A1000/D4x are not very far away. I will personally probably wait until the 2nd generation of the 645D, see if it does have live view and T/S lenses. If Pentax can pull that, chances are that I will sell most of my Nikon gear. If they don't, I might or might not buy a D4x that I would probably keep for 10 years or so. The D3x is already so damned close to perfection that even that will be a hard call.

Cheers,
Bernard

I pulled out the original 1Ds recently, I think that is 8 years old, and apart from the fact that it is VERY heavy, at normal ISO values and enlargement it will give any other camera a run for its money. I have the impression that much of the increase in ISO and noise since then is "fake", with ISO set conservatively to underexpose on the 1Ds and avoid burnout. Out of camera image processing has advanced considerably since then, so higher ISO and highlight recovery should be possible with the same hardware.

Note that I had Canon swap the sensor on my 1Ds, and I was doing fashion shows at 1250 ISO with it at the time ...

I did a bunch of portraits with the 1Ds and the old 135/2 Canon lens a couple of months ago, and can assure you it works. I would have been smarter not to be taken in by marketing and simply kept using the thing. Good big pixels trump bad small pixels every time.

Edmund

[KevinA]

So disregarding the mathematics and the formula, print by print does the Pentax better the 35mm DSLR's?

Kevin.

[ErikKaffehr]

Hi,

From the samples I would say yes, if printed larger than 50x70 cm and using perfect technique.

Moirés may be an issue, if technique is perfect. Stopping down to or past f/11 would probably much reduce moiré and aliasing, thanks diffraction. This observation is much based on Lloyd Chambers field test of the Leica S2, which had lot's of moiré that disappeared at f/11.

Cannot say more without getting the camera and shooting with it!

Best regards
Erik

So disregarding the mathematics and the formula, print by print does the Pentax better the 35mm DSLR's?

Kevin.

[ziocan]

I pulled out the original 1Ds recently, I think that is 8 years old, and apart from the fact that it is VERY heavy, at normal ISO values and enlargement it will give any other camera a run for its money.

Edmund, honestly the only cameras that the old 1ds can "get" a run for its money today, are: 12mp, 500 bucks APSc consumer DSLR.

Of course the pictures it can take are as good today are they were more than good enough back then.
IMO the 1ds was the first digital camera worth buying for replacing film on my work flow.

[ziocan]

So disregarding the mathematics and the formula, print by print does the Pentax better the 35mm DSLR's?

Kevin.

You betcha!

Once I made a comparative test between a Sony a900 with zeiss 135mm (as good as any D3x with equivalent lens) and a Phase P30 with 150mm mamiya.
processed  and sharpened to taste and printed in A3.
Showed the prints to 2 photographers that shoot only dslr, 2 art directors who normally works with high end photographers and an very fine architect.
Asked which one they preferred and one of the two 2 photographers could not tell the difference, the other 4 persons picked the Phase one print.
It was not a night and day difference, but it was there and could be seen.

From the files I have seen, IMO the Pentax is tangibly better than the phase P30 and will print better than any current DSLR.

[eronald]

Edmund, honestly the only cameras that the old 1ds can "get" a run for its money today, are: 12mp, 500 bucks APSc consumer DSLR.

Of course the pictures it can take are as good today are they were more than good enough back then.
IMO the 1ds was the first digital camera worth buying for replacing film on my work flow.

I disagree. The 1Ds focuses faster an better than my 5DII, and the images have more crispness.
In fact I just managed to locate my equally old 85/1.2, and will be taking this dumb-bell out for a walk today.
I won't say the 1Ds is a better camera than the Nikon, It feels a bit like digging out a roll of a different film.

Edmund