Pages: [1] 2 3   Go Down

Author Topic: Sensor to Sensor  (Read 15214 times)

CBarrett

  • Guest
Sensor to Sensor
« on: November 17, 2009, 09:13:59 am »

I did this little test for two reasons:

  • I can
  • I'm a geek

Shooting on the Arca with the Rodenstock HR 70mm, P65+ vs D3 body.  The P65+ is smoother tonally with greater range (as I would expect) and feels sharper.  Everything shot to and processed through C1 Pro 5 with base settings.  This is purely a sensor test and for most, the issue of LF Glass vs T/S lenses needs to play into the equation as well.  The results are the same as always... if you can justify the extra $35k and this is the way you want to work, cool.  If not, you'll do pretty good with the alternative.

 P65+  Full Frame
 D3 Full Frame
 P65+  100% Crop
 D3  100% Crop

/word.
« Last Edit: November 17, 2009, 09:18:11 am by CBarrett »
Logged

haefnerphoto

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 679
    • http://www.jameshaefner.com
Sensor to Sensor
« Reply #1 on: November 17, 2009, 09:28:15 am »

Nice test Chris!  I can't see 35,000.00 worth of improved quality in the P65 file, can you?  The Nikon, to me, looks a little crisper actually.  Jim
Logged

gwhitf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 855
Sensor to Sensor
« Reply #2 on: November 17, 2009, 09:29:27 am »

Quote from: CBarrett
If not, you'll do pretty good with the alternative.

Christopher:

I did a similar test, comparing my P30+ to my 1Ds3 and came to the exact conclusion: a tiny bit of difference, but nothing that would ever show in CMYK, and nothing that couldn't be corrected for, in a tiny contrast adjustment and USM in DPP. I purposely don't want to do this test with my P45+ and my 5D2, because I want to keep my Hasselblad. I just don't want to see the true reality.

I will say this: these cameras do look more similar at f8 or f13, where you're carrying focus deeper into the frame; but the differences are greater when shooting wide open. Not much difference, but a bit more difference.

And not to mention that, with 35, you have excellent T/S lenses to throw focus even moreso. And the ability to shoot in Available Darkness, and without a tripod.

Surprises me that more and more Dealers don't do these side by side tests, either between MF brands, or between MF vs 35, or maybe, it's better not to show the truth?

I'd love to see this side-by-side test, done with a 5D2 and an S2, where the Cost Factor is, literally, 10-to-1.

Again, I feel it's worth noting: when comparing MF to 35, the Resolution Game is over, unless you're splitting hairs, and if you're splitting hairs, you've already bought a P65+ and you're not interested in testing anything. The BIG DEAL between the two formats is how the lenses render focus and render "space". You've got to own it and test it to know what I mean. I know everyone rolls their eyes at that "3D Effect", but it truly does show up, especially shooting wide open. The rendering thing, is the reason you'd keep the MF, not resolution. I simply find that, with 35, the overall feeling is more squashed, or more flat, even with the 85 1.2, wide open. I'm not an Optician or a Scientist, but I know it when I see it.

I also think, finally, that Rob Haggart touches on something that's here in this equation: Perception. Read this article and tell me that the words "Leica" or "P65+" or "PODAS" or "Red Dot" don't flash to the forefront of your mind. The brain is a powerful thing: If you shoot a sunset with a $35k digital back, you just think, on some level, it's just GOT to be better than the same sunset, shot with a 5D2.

http://www.aphotoeditor.com/2009/11/17/per...-is-everything/
« Last Edit: November 17, 2009, 11:28:04 am by gwhitf »
Logged

asf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 510
    • http://www.adamfriedberg.com
Sensor to Sensor
« Reply #3 on: November 17, 2009, 09:45:21 am »

Thanks for posting.

D3 surprisingly good.

Says a lot about that sensor and also the excellent lens.
Logged

Snook

  • Guest
Sensor to Sensor
« Reply #4 on: November 17, 2009, 10:04:17 am »

A lot more dynamic range as expected with the P65.

I agree that I do not think the money difference is worth it... If you want MF you can pick up some great deals on used backs right now.
I shoot MF becasue I like the RZ P30 combo. plus the files hold up to a lot more abuse if needed..
Nice test, but it was expected. I think with the P65 you getting more room to crop etc.. but not "better" quality than a lot of other older backs.
Plus the D3 would blow away any back in the high ISO department...
Snook
Logged

Christopher

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1499
    • http://www.hauser-photoart.com
Sensor to Sensor
« Reply #5 on: November 17, 2009, 10:13:02 am »

Quote from: asf
Thanks for posting.

D3 surprisingly good.

Says a lot about that sensor and also the excellent lens.

Well I am not surprised at all. The main difference or benefit you get, is more resolution, which is only important for larger prints. Perhaps some smaller DR and color advantages, but these are certainly not worth the price.

However there are still more than enough people out there, who will tell you forever that their H50, P45 or 65, Sinar, leaf, DMR, m9 or whatever is SOOOOOOOOOOOOOOOOOOOO much superior to what a Canon or Nikon can do. ;-)
Logged
Christopher Hauser
[email=chris@hauser-p

Slough

  • Guest
Sensor to Sensor
« Reply #6 on: November 17, 2009, 10:18:06 am »

The D3 image is a bit brighter, which might explain the greater tendency to burn out the white curtains. Not saying it will though.
Logged

CBarrett

  • Guest
Sensor to Sensor
« Reply #7 on: November 17, 2009, 10:18:08 am »

Yeah the P65+, even with Sensor+ can't touch the D3's high ISO stuff.  The extra sensor real estate, though, is a big deal to me for shooting architecture wide and the flexibility of the files, I push mine pretty hard and need all the tonal width I can get.  All that and the desire to make big a** prints of my personal work will keep me from parting with the P65+.

-c
Logged

BobDavid

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3307
Sensor to Sensor
« Reply #8 on: November 17, 2009, 10:25:21 am »

One thing you can do with an MF file as opposed to a 5DII file is run it through much more aggressive curves and other adjustments without the image falling apart.
Logged

uaiomex

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1211
    • http://www.eduardocervantes.com
Sensor to Sensor
« Reply #9 on: November 17, 2009, 12:30:39 pm »

I'm no test expert, but for me this test doesn't say it all. First, both cameras should be showing the same captured space, the Nikon is closer, making objects larger. Second, the smaller file should be interpolated to the size of the bigger file. Then compare.
My 2 centavos
Eduardo
Logged

John Collins

  • Newbie
  • *
  • Offline Offline
  • Posts: 33
Sensor to Sensor
« Reply #10 on: November 17, 2009, 01:00:11 pm »

I realize that any response to this question will be theoretical but the comparison (thank you Chris) between the Top Nikon DSLR and the P65+ back has got me thinking -

 How would a P45+ back stack up to a DSLR using the latest technology and a full frame (24X36) sensor with about 32MP (the rumored 1Ds Mk IV)?

I'd like to make detailed prints, I realize that the maximum size can't really be known without the files being available to work with, but in this economy I need to consider the cost of my next move.

I should probably say that I'm new to this big time digital after years using film.

Thanks,
John
Logged

Professional

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 309
Sensor to Sensor
« Reply #11 on: November 17, 2009, 01:11:44 pm »

I really would like to have the point of all these tests comparisons between cameras, cameras are there, some can go with expensive and some even can't afford the cheap, i am sure in the future there will be tests between 100mp MF and 60mp Canon/Nikon cameras, and at the end some people will keep saying it is not a big difference there and not worthy that extra doubles prices of MFDB over digital 35mm/APCs, and sure there are people going with 80 and 100mp MFDB, so i really don't know the end of this war.
I have H3DII-39, but i feel i am not happy with all my cameras if people have doubles of MPs of our 35mm and MF and we have old camera or less resolution than they have, we can't afford it but we really REALLY want to print so large, what is the reason for people to go with 40+ up to 60 or more then? too bad we can't afford high top resolution cameras at least to do tests and to do prints.
Logged

Doug Peterson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4210
    • http://www.doug-peterson.com
Sensor to Sensor
« Reply #12 on: November 17, 2009, 01:15:17 pm »

Quote from: John Collins
I realize that any response to this question will be theoretical but the comparison (thank you Chris) between the Top Nikon DSLR and the P65+ back has got me thinking -

 How would a P45+ back stack up to a DSLR using the latest technology and a full frame (24X36) sensor with about 32MP (the rumored 1Ds Mk IV)?

I'd like to make detailed prints, I realize that the maximum size can't really be known without the files being available to work with, but in this economy I need to consider the cost of my next move.

I should probably say that I'm new to this big time digital after years using film.

Thanks,
John

Perhaps you should rent/buy/borrow/steal a 1Ds II and a 16 megapixel digital back (e.g. P20) or a 1Ds III and a P25+ to see the differences in platforms beyond the "megapixel" number that is used most frequently as an inappropriate proxy for "quality".

Do I know how the 1DsIV will perform? Heck no. All I know is for four generations you have heard "the next Canon will end medium format" and yet here we are with P40+/P65+ selling extremely well.

Doug Peterson
__________________
Head of Technical Services, Capture Integration
Phase One, Leaf, Cambo, Canon, Apple, Profoto, Eizo & More
National: 877.217.9870  |  Cell: 740.707.2183
Newsletter: Read Latest or Sign Up
RSS Feed: Subscribe
Buy Capture One at 10% off

asf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 510
    • http://www.adamfriedberg.com
Sensor to Sensor
« Reply #13 on: November 17, 2009, 01:20:34 pm »

Quote from: John-S
So the next test should be putting the P65+ and D3 files through the same aggressive curves, then print them in a magazine or other CMYK work.

I'm with you here. That the mfdb files hold up better to aggressive post work was most likely true without exception before the D3x/1ds3. Is it really still true?
Logged

Wayne Fox

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4237
    • waynefox.com
Sensor to Sensor
« Reply #14 on: November 17, 2009, 01:42:55 pm »

Quote from: uaiomex
I'm no test expert, but for me this test doesn't say it all. First, both cameras should be showing the same captured space, the Nikon is closer, making objects larger. Second, the smaller file should be interpolated to the size of the bigger file. Then compare.
My 2 centavos
Eduardo


Thought the same thing.  The test is meaningless in terms of resolution, since the scene doesn't really lend itself for that purpose and the image was captured in a way so the detail was recorded at a very close pixel ratio ... the 50% crop delivers the same field of view. I would assume the comment "sensor test" by the OP also implies resolution wasn't a factor and those reading this into the test shouldn't.  The difference in resolving power of a 60mp back vs a dSLR is substantial and visible.  Just try taking a group shot of 30 or 40 people with each and print out a 30x40.

As far as other qualities, pointless trying  to draw any conclusions at all from a 8bit jpeg image in a web browser.

This appears to be a test simply about the ability to handle extreme dynamic range. If it is about dynamic range, I think perhaps those not using MF misunderstand what the files can really deliver.  It isn't just about the extremes. it's also about the tonalities the back can capture through the entire images tonal range.  This is something you just cannot demonstrate with 8bit jpegs and can't even see on a computer monitor.  Before drawing any conclusions, you might want to go shoot with one for a while, testing the limits of the camera and and print some nice large prints using a current inkjet printer.  You might just discover what many others have ... there is a difference.

Whether the additional price is justified, that is a question that applies only to oneself.  Just because you can't justify it financially or because you just don't need it for what you do doesn't mean everyone else is like you.  There are some who cannot get the quality they want out of a dSLR and in fact many MF shooters frequently stitch because single captures still aren't delivering the quality they want for the size they want to print.
Logged

CBarrett

  • Guest
Sensor to Sensor
« Reply #15 on: November 17, 2009, 01:44:32 pm »

Quote from: GBPhoto
You might have missed that both cameras are shooting through the same lens, from the same place.  The D3 is only showing a "crop" of the P65 sensor area.


Thanks Alan.  The point was not to really compare the same shot produced different ways, but to remove all variables except the sensors themselves.  The point may be moot when you consider that I am giving the Nikon an advantage of the Rodenstock HR that it's not going to be married to in everyone else's dslr kits.

I'm curious too, I'll try to pull detail into the curtains and see how the files respond.  How far you can push the raw is a big deal to me, since I often shoot with people which you can't bracket, I'll take one raw, process +1.5, 0, -1.5 and then HDR it (which provides greater range than CO HDR sliders).

-C
Logged

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA
Sensor to Sensor
« Reply #16 on: November 17, 2009, 01:51:37 pm »

Thanks for posting these images - it takes a lot of courage to do this now as shortly the 35mm HDR armchair hypothetical stitch piranhas will swarm in to geek everything up.

There are regions in the image where the P65+ seems to have a noticeable advantage in dynamic range and tonality (and obviously resolution) and I think that will come through in even small prints as a more life like feel.  At least I might notice it.   That said the D3 is not too shabby!  Wonderful times to have these choices available.
« Last Edit: November 17, 2009, 01:56:36 pm by EricWHiss »
Logged
Rolleiflex USA

Professional

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 309
Sensor to Sensor
« Reply #17 on: November 17, 2009, 01:53:30 pm »

Quote from: Wayne Fox
Thought the same thing.  The test is meaningless in terms of resolution, since the scene doesn't really lend itself for that purpose and the image was captured in a way so the detail was recorded at a very close pixel ratio ... the 50% crop delivers the same field of view. I would assume the comment "sensor test" by the OP also implies resolution wasn't a factor and those reading this into the test shouldn't.  The difference in resolving power of a 60mp back vs a dSLR is substantial and visible.  Just try taking a group shot of 30 or 40 people with each and print out a 30x40.

As far as other qualities, pointless trying  to draw any conclusions at all from a 8bit jpeg image in a web browser.

This appears to be a test simply about the ability to handle extreme dynamic range. If it is about dynamic range, I think perhaps those not using MF misunderstand what the files can really deliver.  It isn't just about the extremes. it's also about the tonalities the back can capture through the entire images tonal range.  This is something you just cannot demonstrate with 8bit jpegs and can't even see on a computer monitor.  Before drawing any conclusions, you might want to go shoot with one for a while, testing the limits of the camera and and print some nice large prints using a current inkjet printer.  You might just discover what many others have ... there is a difference.

Whether the additional price is justified, that is a question that applies only to oneself.  Just because you can't justify it financially or because you just don't need it for what you do doesn't mean everyone else is like you.  There are some who cannot get the quality they want out of a dSLR and in fact many MF shooters frequently stitch because single captures still aren't delivering the quality they want for the size they want to print.

When to make a LFDB? In fact i am also not that guy who check all factors in one shot like tonality, DR, color rendition,..... i just look at many photos at 100% on different monitor [i have 30" monitor calibrated], also i print some print at 17x22, when i want to see larger prints easily i go to the lab to print something between 24x36 and 40x60 [even billboard if necessary], here i can see the real comparison, those tests don't give me anything i really want or worthy to read or see, but at the end as you said, if i want something bigger for prints than i should go with larger sensors which you said even they are not sufficient sometimes as well, so what to do if we can't afford cameras that cost about $50k up to $100k for a certain job?
Logged

gdwhalen

  • Full Member
  • ***
  • Offline Offline
  • Posts: 173
    • http://www.gdwhalen.com
Sensor to Sensor
« Reply #18 on: November 17, 2009, 01:54:13 pm »

This test makes no sense to me at all.  The image areas are completely different.  It's like comparing a telephoto to a wide angle.  The balls of yarn are from two entirely different proportions of the image area.  What is the point of this?

CBarrett

  • Guest
Sensor to Sensor
« Reply #19 on: November 17, 2009, 02:12:25 pm »

Quote from: gdwhalen
This test makes no sense to me at all.  The image areas are completely different.  It's like comparing a telephoto to a wide angle.  The balls of yarn are from two entirely different proportions of the image area.  What is the point of this?


Maybe this will clarify it a bit.  The subject to lens relationship is unaltered between the two shots.  The Nikon is mounted on the back of the Arca (just like the P65+) shooting through the same lens at the same position at the same aperture.  It's sort of like shooting something on 4x5 film, then putting a roll film back on and shooting another film.  Nothing changes but the film.. or in this case the sensor.  It's a pixel to pixel comparison of how the two sensors handle detail, tonality and color with all variables removed.

I admit a more real world test would be trying to recreate the shot using the dslr with a T/S lens, but that's something else altogether... and it's been done before and I doubt the D3 would perform as well.
Logged
Pages: [1] 2 3   Go Up