Ad
Ad
Ad
Pages: « 1 ... 3 4 [5]   Bottom of Page
Print
Author Topic: My journey into MF digital, starting with a P45+ on a Hasselblad 555ELD  (Read 12179 times)
EricWHiss
Sr. Member
****
Offline Offline

Posts: 2305



WWW
« Reply #80 on: July 14, 2013, 03:08:29 AM »
ReplyReply

Bart,
Transmissive step wedges (as opposed to printed step wedges) cover more range (over 13 stops)  so are therefore more useful for DR testing of current digital cameras. I'm 'sorry' that you have been confused.  They are available from Stouffer for about $35 but perhaps other sources as well.   
Eric



Hi Eric,

Sorry, but shooting a step-wedge is not more accurate, although may be closer to variable conditions real live shooting situations (partly due to veiling glare), depending on the lens and lighting situation used. For a fair, unbiased review one usually tries to avoid as many variables (such as lens / lighting conditions used) as possible. Ignoring things like these variables is exactly what makes comparisons controversial.

The DxO Mark evaluations seem to be pretty accurate though.

Cheers,
Bart
Logged

Authorized Rolleiflex Dealer:
Find product information, download user manuals, or purchase online - Rolleiflex USA
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 6922


WWW
« Reply #81 on: July 14, 2013, 04:32:34 AM »
ReplyReply

Hi,

I have a Stouffer wedge 41 steps, but I didn't get around to shooting it. You need to eliminate all surrounding light and flare. It is not easy.

Best regards
Erik
Logged

BartvanderWolf
Sr. Member
****
Online Online

Posts: 3015


« Reply #82 on: July 14, 2013, 06:04:12 AM »
ReplyReply

I have a Stouffer wedge 41 steps, but I didn't get around to shooting it. You need to eliminate all surrounding light and flare. It is not easy.

That's correct, I know from personal experience (I have various Stouffer step wedges, reflection and transmission versions) that the biggest variable can come from the shooting conditions (just like in real life, but unwanted for an objective test).

Veiling glare is the proverbial elephant in the room, and surface reflections can also spoil the quality of the test.

Cheers,
Bart
Logged
EricWHiss
Sr. Member
****
Offline Offline

Posts: 2305



WWW
« Reply #83 on: July 14, 2013, 01:09:12 PM »
ReplyReply

Hi,

I have a Stouffer wedge 41 steps, but I didn't get around to shooting it. You need to eliminate all surrounding light and flare. It is not easy.

Best regards
Erik

Just cut a small window out of a black matte board or something similar and tape the transparency wedge to the back from the edges.  Use a large enough matte board to cover your light source.  
« Last Edit: July 14, 2013, 01:11:50 PM by EricWHiss » Logged

Authorized Rolleiflex Dealer:
Find product information, download user manuals, or purchase online - Rolleiflex USA
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 6922


WWW
« Reply #84 on: July 14, 2013, 01:50:56 PM »
ReplyReply

Update: samples reshot

A series of samples were shot and checked with RawDigger, the most ETTR exposures without clipping were chosen.



Perhaps some of this is useful: http://echophoto.dnsalias.net/ekr/Articles/MFDJourney/RawImages/Stouffer/

Best regards
Erik
« Last Edit: July 14, 2013, 04:47:10 PM by ErikKaffehr » Logged

ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 6922


WWW
« Reply #85 on: July 15, 2013, 01:36:13 PM »
ReplyReply

Hi,

Some comments about the Stouffer wedge shots.

1) Tried to shield external light as much as possible.
2) I guess flare limits DR on both Hasselblad and Sony. Macro lenses used on both. Planar 120/4 and Minolta 100/2.8 AF

Here are the evaluations by Imatest:


So the Sony Alpha 99 has about one EV advantage. This is much less than DxO measurement on both cameras but DxO may use a better test target? http://www.dxomark.com/index.php/About/In-depth-measurements/DxOMark-testing-protocols/Noise-dynamic-range
« Last Edit: July 15, 2013, 01:45:01 PM by ErikKaffehr » Logged

EricWHiss
Sr. Member
****
Offline Offline

Posts: 2305



WWW
« Reply #86 on: July 16, 2013, 11:50:50 AM »
ReplyReply

Your imatest numbers are definitely less than mine for testing similar backs (p20, CF 22, CF 528, CF 39, etc) however I can't say why yours are lower and I haven't run your files through imatest myself.  In my own testing, the imatest low values come closer to DXO's values and also fit with the sensor manufacturers stated DR.    But does DXO test camera and lenses or just the sensor?  Lenses may limit contrast and therefore DR - so they may have some effect.  In fact as a side discussion, I wonder what the DR limits of lenses are?  I've read somewhere that the most we can expect through good real life lenses is about 13-14 stops. Anyone have any info on that?  Certainly we are at the point where the lenses can make the difference.

Since I have tested many cameras, the thing I find most useful is the ratio of drop from low to high values in imatest.  DSLR's of past used to drop quite a lot from low to high - ie about 4-5 stops.  Current generation DSLR's like the Nikon d800 don't drop as much between the ratios.  I find the Imatest High value of most use to evaluate how an image will hold up through the editing process.
Logged

Authorized Rolleiflex Dealer:
Find product information, download user manuals, or purchase online - Rolleiflex USA
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 6922


WWW
« Reply #87 on: July 16, 2013, 02:28:00 PM »
ReplyReply

Hi Eric,

In this conversion I have used Imatest internal conversion, based on DCRaw. I can get much better results using LR4, for instance.

The first test I have done was with the 24-70/2.8 ZA lens but the second one was made with the Macro 100/2.8. The Macro 100/2.8 gave much better results which may depend on less internal flare.

The DR you can get trough a lens depends much the size of the brightest patch. According Norman Koren's test most good lenses have about 0.5% flare (or so), that would limit DR to about 8 stops. If the bright patch is small it would be possible to have a much higher DR.

In practice I can see a DR of about 9-10 EV in my raw data, using RawDigger. I have found some files with wider DR but those typically included the sun (like setting sun).

Of the about 65000 images I have on my computer I guess that a few dozens are limited by DR. I have just found a couple of images that really got better by multiple exposure HDR, but in many cases some significant tone mapping effort has been necessary to make a good image.

http://echophoto.dnsalias.net/ekr/index.php/photoarticles/61-hdr-tone-mapping-on-ordinary-image

Best regards
Erik



Your imatest numbers are definitely less than mine for testing similar backs (p20, CF 22, CF 528, CF 39, etc) however I can't say why yours are lower and I haven't run your files through imatest myself.  In my own testing, the imatest low values come closer to DXO's values and also fit with the sensor manufacturers stated DR.    But does DXO test camera and lenses or just the sensor?  Lenses may limit contrast and therefore DR - so they may have some effect.  In fact as a side discussion, I wonder what the DR limits of lenses are?  I've read somewhere that the most we can expect through good real life lenses is about 13-14 stops. Anyone have any info on that?  Certainly we are at the point where the lenses can make the difference.

Since I have tested many cameras, the thing I find most useful is the ratio of drop from low to high values in imatest.  DSLR's of past used to drop quite a lot from low to high - ie about 4-5 stops.  Current generation DSLR's like the Nikon d800 don't drop as much between the ratios.  I find the Imatest High value of most use to evaluate how an image will hold up through the editing process.
Logged

ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 6922


WWW
« Reply #88 on: July 16, 2013, 08:12:09 PM »
ReplyReply

Hi,

This was converted by C1 (linear conversion, default settings).

Note higher SNR in medium and high tonal range. This reflects higher photon count, probably, it is expected. Liking C1 a bit better, learning all the time ;-)

Note: in my samples, the right and of the wedge is blackend out by a credit card on top of the wedge. There is a small light leak giving a false step in some developments. This was cropped out in the latest conversion. Might be a small disadvantage for Sony, as some data were cropped away.


Best regards
Erik




« Last Edit: July 16, 2013, 08:28:39 PM by ErikKaffehr » Logged

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

Ad
Ad
Ad