Pages: [1] 2   Go Down

Author Topic: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples  (Read 15035 times)

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto

Hi,

I have a short writeup on a comparison of the Pentax 645D with the Nikon D3X. The comparison is based on prints up to A0 size.

http://echophoto.dnsalias.net/ekr/index.php/photoarticles/51-a-closer-look-at-pentax-645d-image-quality

The article is still in progress and may be updated in the next couple of days.

The essence is: Pentax 645D delivers...

Best regards
Erik
Logged
Erik Kaffehr
 

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #1 on: December 18, 2010, 10:00:30 am »

Really not sure what I am seeing here in terms of comparsion...

Of course looking at an image on the monitor zoomed at 100% doesn't tell how the print of this particular image will look like (grain, sharpness, contrast etc. etc.).
But still two images zoomed at 100% tell something about the difference of the two respective images in terms of comparision.
Or maybe zoom them both at 50%... or if you use OpenGL zoom them both to print size ... either way.

But when you are comparing actual prints you also compare the method of uprezzing, the sharpening technique and you name it.
So the entire printing workflow...
How will the prints look like without resampling, i.e. at native pixel size? For instance if you print the 645D image at 360ppi and the D3X at 300ppi (to achieve a comparable print size).

The Pentax simply resloves more details and the actual image is larger.
You can see that at any (useful) zoom magnification on the monitor.
Therefore it comes to now surprise that the difference is also apparent in a print.

Your 645D processing obviously doesn't match the D3x image by the way. It's darker, less contrasty and the white balance is warmer.
All three aspects are an advantage for the D3x print here... as the 645D looks a bit dull in comparision.

Bottom line... larger prints benefit from captures with higher resolution.
But I was already aware of that before...
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #2 on: December 18, 2010, 10:05:46 am »

Hi,

Just in case you missed it, all images are scans from actual prints of A2 and A0 size. So the full pipeline is tested. Sorry for the color mismatch,I tried to adjust as well as I could, although I didn't spend that much effort.

I have also evaluated the prints visually but prints are hard to share over the net. This is the best approach I came up with,

Best regards
Erik
Logged
Erik Kaffehr
 

pcunite

  • Full Member
  • ***
  • Offline Offline
  • Posts: 205
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #3 on: December 18, 2010, 10:24:51 am »

I really think that 160mp'ish is going to be the sweet spot. Remember that 40mp is only a doubling of 20 and you must double to get appreciable gains. Of course I could be wrong if the future holds wall length paper thin LCD mounts in every home. :)

Personally I think that if a person is not happy with a 20mp system, they should move to at least 80mp. In the computer processing world, if a task takes 5 minutes on a 2Ghz system, it will take 2.5 minutes on a 4Ghz system. I would not want to spend that kind of money for such a small gain. Better to buy an 8Ghz system and get something that would change your life, you could feel every $ at work.

Just expressing a point of view...
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #4 on: December 18, 2010, 10:38:26 am »

Hi,

I don't know. Lot of people invest in 60 MP backs right now. Would the Pentax go to 160 MP it would have a pixel pitch of 3 microns. This may be good for aliasing but may be less than optimal for DR. I guess that 20 MP is good enough for many uses.

Also, the 40 MP's now are probably not easy to utilize. Dead on focus, no vibration and don't stop down beyond f/11. With 160 MP that would be f/5.6!

Perhaps a good reason not to go much beyond 5 microns. P65+ has a larger sensor so sensor pitch is still around 6 microns although it has around 60 MP.

Best regards
Erik

I really think that 160mp'ish is going to be the sweet spot. Remember that 40mp is only a doubling of 20 and you must double to get appreciable gains. Of course I could be wrong if the future holds wall length paper thin LCD mounts in every home. :)

Personally I think that if a person is not happy with a 20mp system, they should move to at least 80mp. In the computer processing world, if a task takes 5 minutes on a 2Ghz system, it will take 2.5 minutes on a 4Ghz system. I would not want to spend that kind of money for such a small gain. Better to buy an 8Ghz system and get something that would change your life, you could feel every $ at work.

Just expressing a point of view...
Logged
Erik Kaffehr
 

LKaven

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1060
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #5 on: December 18, 2010, 02:46:49 pm »

Erik, could you share what are now your preferred deconvolution sharpening methods for the D3x?  As I recall, you put quite a bit of work into finding a near-optimal setting for reversing the effects of the OLP filter.  I'm hoping I don't need CS5 to do it, or anything too expensive.

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #6 on: December 18, 2010, 03:12:21 pm »

Just in case you missed it, all images are scans from actual prints of A2 and A0 size. So the full pipeline is tested.
no, I did not miss it. Actually I was referring exactly to this particular kind of comparision. It's somehow useless as it involves a lot of image transformations (uprezzing, sharpening etc.). The printing worklfow for the D3x should be different to the workflow for the 645D... simply because the starting position is different.
For instance... I guess the D3x print would look somewhat better with less uprezzing but priniting at lower ppi. But that's just a guess as I don't know the actual resultion you applied to the prints.
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #7 on: December 18, 2010, 03:42:13 pm »

Hi,

480 PPI I think, but I'm not really sure. The idea was not to deliver the absolute truth but to give an idea to potential buyers of the P645D of the quality difference between a top end DSLR and an entry point MFDB. The comparison also discusses sharpness and none of the other purported advantages of medium format digital. Regarding resolution it seems to me that he Pentax 645D delivers.

Best regards
Erik


no, I did not miss it. Actually I was referring exactly to this particular kind of comparision. It's somehow useless as it involves a lot of image transformations (uprezzing, sharpening etc.). The printing worklfow for the D3x should be different to the workflow for the 645D... simply because the starting position is different.
For instance... I guess the D3x print would look somewhat better with less uprezzing but priniting at lower ppi. But that's just a guess as I don't know the actual resultion you applied to the prints.
Logged
Erik Kaffehr
 

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #8 on: December 18, 2010, 03:57:52 pm »

Hi,

I have used Lightroom 3.3. Develop has a slider called "detail" in the "detail" dialogue. This does unsharp masking at the left position and shifts to deconvolution at the rightmost position. Deconvolution is a way to restore detail. Normally deconvolution requires that the Point Spread Function, which is seldom the case. Lightroom estimates PSF using a model of a defocused image assuming a lens with a circular aperture, at least according to Eric Chan (Madmanchan) who is often posting on these forums.

Economical alternatives to CS5/Lightroom may be Iridient Raw Developer. Focus Magick or Topaz Labs In Focus (or their older product "Detail"). The settings I used in Lightroom are shown below.

By the way Bart van Der Wolf and others are much better references on the issue than myself!

Best regards
Erik




Erik, could you share what are now your preferred deconvolution sharpening methods for the D3x?  As I recall, you put quite a bit of work into finding a near-optimal setting for reversing the effects of the OLP filter.  I'm hoping I don't need CS5 to do it, or anything too expensive.
« Last Edit: December 18, 2010, 04:00:36 pm by ErikKaffehr »
Logged
Erik Kaffehr
 

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #9 on: December 18, 2010, 03:59:30 pm »

480 PPI I think
why that high?

Quote
The idea was not to deliver the absolute truth but to give an idea to potential buyers of the P645D of the quality difference between a top end DSLR and an entry point MFDB.
well, I think the RAW files make that pretty clear.
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #10 on: December 18, 2010, 04:08:28 pm »

That's what I normally use.

I seldom print larger than A2, when I do I'll send the image to a pro lab printing on Durst Lambda. For large sizes they recommend 200 PPI, but at that density there are visible staircase artifacts.

Also, I don't think that excess resolution matters as I actually show scanned prints. I don't think any of the prints would be better using a lower resolution. Regarding CPU-time I simply don't care about wasted CPU cycles!

Best regards
ERik

why that high?
well, I think the RAW files make that pretty clear.
Logged
Erik Kaffehr
 

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #11 on: December 18, 2010, 04:27:42 pm »


Honestly, I don't get it…

Din A2 = 59,4 x 42 cm

printed @ 300ppi you need 7016x4961 pixels
printed @ 360ppi you need 8419x5953 pixles
printed @ 480ppi you need 11225x7937 pixles

So the higher the print resolution the more you have to uprez the actual pixels.
A print at 300ppi (RGB printer etc.) or 360ppi (Epson) is basically super sharp and detailed to our eyes. 240ppi on an Epson is still very good (mostly).
So you beat the files back and forth just to gain a resolution that nobody could tell from 360ppi (300ppi respectively).
Something simply doesn't add up…
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #12 on: December 18, 2010, 04:31:44 pm »

I dont argue...

That said there is some rationale behind using a high resolution...

- Jeff Schewe has indicated that it matters sometime, not often but sometimes
- I would scan the images and do some pixel peeping
- The printer driver will resize to native dimensions anyway. My guess is Lightroom may be better at resizing.
- I wanted to avoid a situation where Pentax image would be downsized and Nikon image upsized.


Best regards
Erik

Honestly, I don't get it…

Din A2 = 59,4 x 42 cm

printed @ 300ppi you need 7016x4961 pixels
printed @ 360ppi you need 8419x5953 pixles
printed @ 480ppi you need 11225x7937 pixles

So the higher the print resolution the more you have to uprez the actual pixels.
A print at 300ppi (RGB printer etc.) or 360ppi (Epson) is basically super sharp and detailed to our eyes. 240ppi on an Epson is still very good (mostly).
So you beat the files back and forth just to gain a resolution that nobody could tell from 360ppi (300ppi respectively).
Something simply doesn't add up…

« Last Edit: December 18, 2010, 05:02:01 pm by ErikKaffehr »
Logged
Erik Kaffehr
 

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #13 on: December 18, 2010, 05:23:19 pm »

Jeff Schewe has indicated that it matters sometime, not often but sometimes
480ppi vs. 300ppi (360ppi) at the price of even more uprezzing???
I am curious to read about that...

Quote
The printer driver will resize to native dimensions anyway.
don't get the point I think...
if you prepare an image at a defined size and defined printer resolution... why should the printer resize anything??

Quote
I wanted to avoid a situation where Pentax image would be downsized and Nikon image upsized
see my suggstion in post #2... just print the D3x file without uprezzing at 300ppi and the 645D file without downrezzing at 360ppi.

Logged

madmanchan

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2115
    • Web
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #14 on: December 18, 2010, 06:13:31 pm »

Most printer drivers do not process input images at arbitrary resolutions (e.g., pixels/inch). There are usually at most a handful of resolutions at which the driver can process the image. If your provided image does not match one of those accepted resolutions, the image will be resampled to meet one of the accepted resolutions.
Logged
Eric Chan

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8914
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #15 on: December 18, 2010, 07:26:57 pm »

Honestly, I don't get it…

Din A2 = 59,4 x 42 cm

printed @ 300ppi you need 7016x4961 pixels
printed @ 360ppi you need 8419x5953 pixles
printed @ 480ppi you need 11225x7937 pixles

So the higher the print resolution the more you have to uprez the actual pixels.
A print at 300ppi (RGB printer etc.) or 360ppi (Epson) is basically super sharp and detailed to our eyes. 240ppi on an Epson is still very good (mostly).
So you beat the files back and forth just to gain a resolution that nobody could tell from 360ppi (300ppi respectively).
Something simply doesn't add up…

Hi Tho_mas,

While there will not be more image pixels carrying detail information, there will be more pixels to perform (output) sharpening on. This potentially boosts the possibility to render sharper detail and boost the MTF for the finest spatial frequencies that the image did record.

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

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #16 on: December 19, 2010, 06:34:36 am »

While there will not be more image pixels carrying detail information, there will be more pixels to perform (output) sharpening on. This potentially boosts the possibility to render sharper detail and boost the MTF for the finest spatial frequencies that the image did record.
I'd like to see this in real prints...
While this might be true for certain images (probably only for certain details of certain images actually) I suspect it very much depends on the amount of uprezzing. In every image there is a certain limit (threshold) of uprezzing. If you uprez too high you are just introducing artefacts such like doubling of edges, a kind of ghosting in straight vertical or horizontal lines etc. (at least as long as we talk about bicubic smoother). You certainly won't enhance those pseudo details...?
Within reasonable values I think a higher enlargement is leveled out by higher printer resolution. But if the enlarged image is actually destroyed by too heavy uprezzing I seriously doubt that the higher printer resolution will level out the artefacts (i.e. "new" details that were actually not captured).
I do prefer a higher amount of uprezzing and printing at 300ppi over less uprezzing and printing at 200ppi on a Chromira RGB printer. However printing on an Epson I would prefer 240ppi over 360ppi with some critical images. Though mostly I print at 360ppi. But I can't imagine that I would gain anything printing at 480ppi... at least not when heavy uprezzing is required to meet a printer resolution that high.

Logged

stever

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1250
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #17 on: December 19, 2010, 11:18:26 am »

for any non-subscribers, Lloyd Chambers is not testing the 645 (diglloyd.com) and finding his copy of the 55mm lens very disappointing - even compared to a 5DII and 24-70!
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #18 on: December 19, 2010, 01:49:56 pm »

Hi,

Yes, I have seen it. A note on the issue has been added to my article.

Best regards
Erik

for any non-subscribers, Lloyd Chambers is not testing the 645 (diglloyd.com) and finding his copy of the 55mm lens very disappointing - even compared to a 5DII and 24-70!
Logged
Erik Kaffehr
 

darr

  • Full Member
  • ***
  • Offline Offline
  • Posts: 145
Re: Pentax 645D compared to Nikon 3DX based on Imaging Resource's samples
« Reply #19 on: December 19, 2010, 01:54:01 pm »

for any non-subscribers, Lloyd Chambers is not testing the 645 (diglloyd.com) and finding his copy of the 55mm lens very disappointing - even compared to a 5DII and 24-70!

I think you mean "now testing"

http://diglloyd.com/dap/index.html#Pentax645D
http://diglloyd.com/diglloyd/index.html
Logged
Pages: [1] 2   Go Up