Ad
Ad
Ad
Pages: « 1 2 [3] 4 5 »   Bottom of Page
Print
Author Topic: 8x10" vs 4x5" vs IQ180 vs ... Great test by Tim Parkin et al.  (Read 25457 times)
Fine_Art
Sr. Member
****
Offline Offline

Posts: 1099


« Reply #40 on: December 26, 2011, 10:13:38 PM »
ReplyReply

Hi,

Here is some great research done by Tim Parkin and enthusiastic colleagues: http://www.landscapegb.com/2011/12/big-camera-comparison/


This part of the test may be most informative: http://static.timparkin.co.uk/static/tmp/cameratest-2/large.html

If you want to compare the lesser cameras the slightly downsized image may be of interest: http://static.timparkin.co.uk/static/tmp/cameratest-2/800px.html

The test here goes into greater detail than I have seen before, definitively worth reading.

My own observations:

- Tim's tests indicates that Phase One P45 pretty well matches 4x5" Velvia

- The Sony A900 is clearly superior to Mamiya 7 with Portra 400. Pity that they have not tested with Velvia.

- 8x10" beats IQ180 weather Velvia or Portra 400

This is essentially in good agreement with some previous testing that indicated P45 equalling 4x5" and 20 MP DSLRs matching 67 on Velvia.

The bridge on top of ridge is worth checking out visible on Velvia but not really on IQ180, see enclosed images!
Best regards
Erik

Film really should be run through a noise reduction package. Digital does the same thing from RAW.

« Last Edit: December 26, 2011, 10:29:22 PM by Fine_Art » Logged
timparkin
Jr. Member
**
Offline Offline

Posts: 56


« Reply #41 on: December 27, 2011, 04:15:48 AM »
ReplyReply

Film really should be run through a noise reduction package. Digital does the same thing from RAW.



Noise reduction is a balancing act. Any more nr here would destroy low contrast details. However, scanning on a drum scanner with a larger aperture will help a lot. I will write an addendum about this over his holiday.

Portra areas overexposed do suffer from noise when scanned at smaller apertures. Time to play some more.
Logged
imagico
Newbie
*
Offline Offline

Posts: 47


WWW
« Reply #42 on: December 27, 2011, 02:23:37 PM »
ReplyReply

First of all - thanks to Tim and all others involved for this test, which - to my knowledge - is by far the most extensive and thorough comparison on the matter done until now.

Reading it (and especially Tim's Editorís Commentary) made me realize is that, depending on how fast pixel densities will rise, the most serious limiting factor of 'small' formats will probably soon be lens quality.  There is most likely no 35mm lens available that is diffraction limited by f/4 up to the corners, especially no wide angle.  Situation might be a bit better for MF with lenses achieving maximum quality at f/5.6 existing but to reach up to the realm of 8x10" this would need to improve to at least f/2.8 (35mm) and f/4 (MF).

I remember Michael Reichmann citing Leica people saying their camera systems are currently not lens limited - but this could change soon.  Alternatives would be stitching (obviously) and larger than MF solid state sensors (which would be prohibitively expensive).  Significantly improving lens resolution OTOH will require developing new design and manufacturing techniques.
Logged

Christoph Hormann
photolog / artificial images / other stuff
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7680


WWW
« Reply #43 on: December 27, 2011, 03:37:03 PM »
ReplyReply

Hi,

I'm not really sure about lens being the limiting factor. Very clearly, this is not the case on IQ180. Even with 5x8 or 8x10 it seems to me that film may matter a lot.

I'm actually a bit confused:

- Michael Reichmann's testing has clearly showed that Canon 1DsII was superior to Velvia on Pentax 67.

- My own testing has at least indicated that 24 MP Digital (Sony Alpha 900) was competitive to Pentax 67. Digital was mostly better but not always.

- In Tim's test it seems that film on Mamiya 67 is in many aspects ahead of 20+ MP digital.

- Charly Cramer and Joseph Holmes found Phase One P45 (or P65?) to be on par width 4x5" Velvia, but it's not the case in Tim's tests.

This has clearly to do with both equipment and evaluation. Tim scans at higher resolution. Tim's testing is also more demanding, as he sets the ribbon at a very high definition scan from a 8x10". Digital images are very smooth and they are good at resolving some detail, but resolution ends abruptly at the Nyquist limit. Upscaling digital images gives a lot of ugly artifacts. High res scans may scale better due to oversampling.

Best regards
Erik




First of all - thanks to Tim and all others involved for this test, which - to my knowledge - is by far the most extensive and thorough comparison on the matter done until now.

Reading it (and especially Tim's Editorís Commentary) made me realize is that, depending on how fast pixel densities will rise, the most serious limiting factor of 'small' formats will probably soon be lens quality.  There is most likely no 35mm lens available that is diffraction limited by f/4 up to the corners, especially no wide angle.  Situation might be a bit better for MF with lenses achieving maximum quality at f/5.6 existing but to reach up to the realm of 8x10" this would need to improve to at least f/2.8 (35mm) and f/4 (MF).

I remember Michael Reichmann citing Leica people saying their camera systems are currently not lens limited - but this could change soon.  Alternatives would be stitching (obviously) and larger than MF solid state sensors (which would be prohibitively expensive).  Significantly improving lens resolution OTOH will require developing new design and manufacturing techniques.
« Last Edit: December 27, 2011, 03:52:12 PM by ErikKaffehr » Logged

ced
Sr. Member
****
Offline Offline

Posts: 267


« Reply #44 on: December 28, 2011, 09:16:07 AM »
ReplyReply

(downsized to 2000dpi from 4000dpi)
Tim are you able to upload a crop of just the house done at 4000dpi please?  Thanks!
Thanks again for the fun I had reading all about the superb tests your team presented to the photo community.
Logged
timparkin
Jr. Member
**
Offline Offline

Posts: 56


« Reply #45 on: December 28, 2011, 04:21:00 PM »
ReplyReply

(downsized to 2000dpi from 4000dpi)
Tim are you able to upload a crop of just the house done at 4000dpi please?  Thanks!
Thanks again for the fun I had reading all about the superb tests your team presented to the photo community.

which house, which film? Most of the scans in the 'large' section are shown at 4000dpi for the 8x10. If you look at the '800px' section then all of the 4x5 in here are shown at 4000dpi
Logged
ced
Sr. Member
****
Offline Offline

Posts: 267


« Reply #46 on: December 28, 2011, 04:26:17 PM »
ReplyReply

Sorry I thought the "house" was obvious, the one in your reply no:32
Logged
douglasboyd
Newbie
*
Offline Offline

Posts: 31


WWW
« Reply #47 on: December 29, 2011, 11:07:06 AM »
ReplyReply

The test is great with a lot of useful information.  I have cameras that cover the test range and my main question is when to use each type.  I print at 20x30 (Costco $8.95) and larger.  I know the experts on this forum will have a lot of valuable advice on this subject, so I will start of with a brief summary of how I decide what to shoot with:

1.  APS-C digital:  This size was not covered in the test but is important with Sony A65, A77 and Nex-7 at 24mp.   I use this when I need low size and weight and/or high depth-of-field.  I use the Nex with my Contax G lenses.

2. 35mm full frame DSLR:  I use A900 24mp when I need versatility with speed, and a high range of focal lengths.  For higher resolution I stitch.

3.  MF Film:  I use Mamiya 6 for shorter lenses (50mm and 75mm) and Mamiya RZ67 for longer lenses.  But actually, not using this system much these days.

4.  MF Digital:  I use Hasselblad H3DII-39 when I want shallow depth of field, e.g. portraits.  This system easily gives DOF of 2cm or less in typical head shots.  With the wider lenses, this system also works for some landscapes.

5.  4x5 film:  This system has the advantage of movements that can sometimes be helpful to get greater depth of field in landscapes and perspective control.  But I don't use it much these days. 

6.  LF film:  I don't have an 8x10 view camera, but have a Fuji 6 x17 camera which comes close to LF resolution in the long direction.  I use this to get the "Peter Lik" look in panoramas when I want to print large, and cannot use stitching.  I have been using Velvia 50 and 100, but based on this article, I am going to experiment more with negative film to exploit dynamic range.

The biggest cost issue for me is not MFDB cost or processing cost, but the cost of framing large prints!  Anyway, that's where I am today, and I love the comments in this thread.

==Doug
Logged
marfa.tx
Newbie
*
Offline Offline

Posts: 27


« Reply #48 on: December 29, 2011, 01:03:14 PM »
ReplyReply

as mentioned to Tim before the "test"... two paths to a digital image....
take your pick, while you can.
Logged

-------
richard
jsch
Sr. Member
****
Offline Offline

Posts: 426


« Reply #49 on: December 29, 2011, 03:55:34 PM »
ReplyReply

Hi Tim,

on your website you write about scanning 8x10 inch film:

[...  This would suggest that if you donít have access to a drum scanner, you may be better off shooting negative film? The 8◊10 Portra 400 scan was hardly degraded at all however; the results showed a similar level of sharpness of that of the drum scan with very little evidence of the colour fringing and halation that flatbed scanners are renowned for (all Epson V750 scans were made at 4800dpi and down sampled to reduce noise). ...]

Did you do this also with b&w film? Did you find there similar results?

Thank you,
Johannes
Logged
lowep
Sr. Member
****
Offline Offline

Posts: 425


WWW
« Reply #50 on: December 29, 2011, 04:14:10 PM »
ReplyReply

Thanks for posting this very interesting test.

How about versus 1.8 gigapixels?  Tongue
« Last Edit: December 29, 2011, 04:18:09 PM by lowep » Logged
Doug Peterson
Sr. Member
****
Offline Offline

Posts: 2838


WWW
« Reply #51 on: December 29, 2011, 04:30:30 PM »
ReplyReply

6.  LF film:  I don't have an 8x10 view camera, but have a Fuji 6 x17 camera which comes close to LF resolution in the long direction.  I use this to get the "Peter Lik" look in panoramas when I want to print large, and cannot use stitching.  I have been using Velvia 50 and 100, but based on this article, I am going to experiment more with negative film to exploit dynamic range.

Notably Peter Lik now uses a Phase One digital back to get his "Peter Lik" look :-) We sold him his first unit.

Doug Peterson (e-mail Me)
__________________

Head of Technical Services, Capture Integration
Phase One Partner of the Year
Leaf, Leica, Cambo, Arca Swiss, Canon, Apple, Profoto, Broncolor, Eizo & More

National: 877.217.9870  |  Cell: 740.707.2183
Newsletter | RSS Feed
Buy Capture One 6 at 10% off
« Last Edit: December 29, 2011, 04:33:32 PM by dougpetersonci » Logged

DOUG PETERSON (dep@digitaltransitions.com), Digital Transitions
Dealer for Phase One, Mamiya Leaf, Arca-Swiss, Cambo, Profoto
Office: 877.367.8537
Cell: 740.707.2183
Phase One IQ250 FAQ
Fine_Art
Sr. Member
****
Offline Offline

Posts: 1099


« Reply #52 on: December 29, 2011, 08:54:41 PM »
ReplyReply

Noise reduction is a balancing act. Any more nr here would destroy low contrast details. However, scanning on a drum scanner with a larger aperture will help a lot. I will write an addendum about this over his holiday.

Portra areas overexposed do suffer from noise when scanned at smaller apertures. Time to play some more.

I use these settings mostly for chroma noise reduction. There is no noticeable reduction in fine detail.

Edit- look at the top left corner to see the "after" difference.
Logged
timparkin
Jr. Member
**
Offline Offline

Posts: 56


« Reply #53 on: December 31, 2011, 03:09:36 AM »
ReplyReply

on your website you write about scanning 8x10 inch film:

[...  This would suggest that if you donít have access to a drum scanner, you may be better off shooting negative film? The 8◊10 Portra 400 scan was hardly degraded at all however; the results showed a similar level of sharpness of that of the drum scan with very little evidence of the colour fringing and halation that flatbed scanners are renowned for (all Epson V750 scans were made at 4800dpi and down sampled to reduce noise). ...]

Did you do this also with b&w film? Did you find there similar results?


Yes we did - the black and white didn't suffer from high contrast edge flare anywhere near as much as the transparency film but it was a little worse than the neg film.
Logged
timparkin
Jr. Member
**
Offline Offline

Posts: 56


« Reply #54 on: December 31, 2011, 03:15:43 AM »
ReplyReply

The test is great with a lot of useful information.  I have cameras that cover the test range and my main question is when to use each type.  I print at 20x30 (Costco $8.95) and larger.  I know the experts on this forum will have a lot of valuable advice on this subject, so I will start of with a brief summary of how I decide what to shoot with:

1.  APS-C digital:  This size was not covered in the test but is important with Sony A65, A77 and Nex-7 at 24mp.   I use this when I need low size and weight and/or high depth-of-field.  I use the Nex with my Contax G lenses.

2. 35mm full frame DSLR:  I use A900 24mp when I need versatility with speed, and a high range of focal lengths.  For higher resolution I stitch.

3.  MF Film:  I use Mamiya 6 for shorter lenses (50mm and 75mm) and Mamiya RZ67 for longer lenses.  But actually, not using this system much these days.

4.  MF Digital:  I use Hasselblad H3DII-39 when I want shallow depth of field, e.g. portraits.  This system easily gives DOF of 2cm or less in typical head shots.  With the wider lenses, this system also works for some landscapes.

5.  4x5 film:  This system has the advantage of movements that can sometimes be helpful to get greater depth of field in landscapes and perspective control.  But I don't use it much these days. 

6.  LF film:  I don't have an 8x10 view camera, but have a Fuji 6 x17 camera which comes close to LF resolution in the long direction.  I use this to get the "Peter Lik" look in panoramas when I want to print large, and cannot use stitching.  I have been using Velvia 50 and 100, but based on this article, I am going to experiment more with negative film to exploit dynamic range.

The biggest cost issue for me is not MFDB cost or processing cost, but the cost of framing large prints!  Anyway, that's where I am today, and I love the comments in this thread.

==Doug


From my experience running the test, you can blow up a 24Mp DSLR with a good lens to 20x24 and people will accept it although in a side by side with a higher res camera it will show (just). I imagine the Nex7 will be close but possibly 16x20 may be the limit. (I'll be testing this when I get rid of my 5Dmk2)

MF film is happy up to just over 20x24 and will look good up to 30x40 although grain will be showing unless treated and then it could look a bit 'plasticky' depending on NR

MFDB should be OK to 30x40 for landscapes although with Portraits I imagine you could get larger with acceptable results

LF 4x5 should be good for 50x40 or maybe larger if you accept some noise.

8x10 you could get to 60x80 and show no grain..

6x17 is interesting, I've found this produces very sharp 120's that easily exceed 4000dpi (with good technique of course) and you should get a 25x70 print from a 4000dpi scan.

Tim

Logged
timparkin
Jr. Member
**
Offline Offline

Posts: 56


« Reply #55 on: December 31, 2011, 03:21:03 AM »
ReplyReply

Sorry I thought the "house" was obvious, the one in your reply no:32

There were quite a few 'houses' in the original subject of the thread (in the landscape view). Get what you mean now though - let me have a look for it..

Tim
Logged
theguywitha645d
Sr. Member
****
Offline Offline

Posts: 970


« Reply #56 on: December 31, 2011, 10:26:00 AM »
ReplyReply

Funny, when you understand the concepts behind viewing distance and human perception, print size has nothing to do with format nor the number of pixels. And when you print large, as I do, you see the underlying theory works really well. (I have made 16x20s from m4/3 cameras, which are beautiful and can go larger.)
« Last Edit: December 31, 2011, 10:32:10 AM by theguywitha645d » Logged
ced
Sr. Member
****
Offline Offline

Posts: 267


« Reply #57 on: December 31, 2011, 10:43:41 AM »
ReplyReply

As I understand the ideal viewing distance is minimum 2.5X the diagonal of the image in question.
Logged
theguywitha645d
Sr. Member
****
Offline Offline

Posts: 970


« Reply #58 on: December 31, 2011, 10:54:11 AM »
ReplyReply

Viewing distance is equal to the diagonal of the image.
Logged
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7680


WWW
« Reply #59 on: December 31, 2011, 11:15:26 AM »
ReplyReply

Hi,

Viewing distance is the distance the viewer chooses to view the image, IMHO.

Norman Koren discusses the issue here: http://www.imatest.com/docs/sqf/

Best regards
Erik


Viewing distance is equal to the diagonal of the image.
Logged

Pages: « 1 2 [3] 4 5 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad