Pages: 1 [2] 3 4 ... 9   Go Down

Author Topic: expose to the right?  (Read 62074 times)

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
expose to the right?
« Reply #20 on: November 08, 2007, 09:47:13 am »

Quote
btw - I don't use the profile you made for me for my HP printer. The stock HP profile gave more realistic looking greens, yellows and reds.
[a href=\"index.php?act=findpost&pid=151239\"][{POST_SNAPBACK}][/a]

As you would see on my site, I offer a money back guarantee on profiles if you're not happy.

I have absolutely no record of building a profile for you BTW. But you can contact me off list about this since apparently you feel that this has something to do with your disagreements about ETTR which clearly it does not.

Then you can explain what "realistic looking" is colorimetrically <g>. I'm having a difficult time understanding your use of language in a few areas here.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA
expose to the right?
« Reply #21 on: November 08, 2007, 11:17:08 am »

Quote
I have absolutely no record of building a profile for you BTW. But you can contact me off list about this since apparently you feel that this has something to do with your disagreements about ETTR which clearly it does not.

No need to take this off topic here. Got the profile a couple years back.  You probably don't remember meeting me at SFdig either.
Logged
Rolleiflex USA

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
expose to the right?
« Reply #22 on: November 08, 2007, 11:21:14 am »

Quote
No need to take this off topic here. Got the profile a couple years back.  You probably don't remember meeting me at SFdig either.
[a href=\"index.php?act=findpost&pid=151314\"][{POST_SNAPBACK}][/a]

No recollection that is correct.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA
expose to the right?
« Reply #23 on: November 08, 2007, 11:22:27 am »

Quote
I'm not sure it would be wise to write off the advice of Thomas Knoll, Andrew Rodney, Bruce Frasier and a host of others who know a great deal more about the subject than you do.  While they may not shoot for a living they did have a huge hand in creating the software that drives the industry.  ...

Definitely I'm not writing off Knoll or Frasier, or Margulis or Schewe....did I forget anybody?    But  what I am saying is I (me personally) like the look of the file better without ETTR.   I am not looking at 1's and 0's....I am looking at an image.
Logged
Rolleiflex USA

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
expose to the right?
« Reply #24 on: November 08, 2007, 01:58:47 pm »

Quote
Definitely I'm not writing off Knoll or Frasier, or Margulis or Schewe....did I forget anybody?    But  what I am saying is I (me personally) like the look of the file better without ETTR.   I am not looking at 1's and 0's....I am looking at an image.
[a href=\"index.php?act=findpost&pid=151316\"][{POST_SNAPBACK}][/a]

I did the experiment that Eric suggested and am somewhat surprised by the results. I had some exposures of a Stouffer step wedge. The first was exposed so that step 1 has a pixel value of 250 when converted with ACR at default settings into aRGB. This is exposed to the right as far as possible. A second exposure placed the same step at 221, and this step was brought back to 250 in ACR, which required an exposure compensation of +1.05 EV.

If you compare the resultant images in Photoshop using the difference blending mode, the differences are in the mid tones and are too dark to make out on the screen, but can be detected with the eye dropper.



Imatest plots show the tonal response curves more clearly and they are different.As expected, the under exposure costs dynamic range, but when the ACR exposure compensation is carried out, the TRCs are different as shown. Comments are invited.

Logged

Wayne Fox

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4237
    • waynefox.com
expose to the right?
« Reply #25 on: November 08, 2007, 02:31:46 pm »

Quote
I did the experiment that Eric suggested and am somewhat surprised by the results. I had some exposures of a Stouffer step wedge. The first was exposed so that step 1 has a pixel value of 250 when converted with ACR at default settings into aRGB. This is exposed to the right as far as possible. A second exposure placed the same step at 221, and this step was brought back to 250 in ACR, which required an exposure compensation of +1.05 EV.

Curious as to which camera/back and which RAW converter was used for the test.

(noticed the nikon d200 after I posted and ACR after I posted. Sorry)
« Last Edit: November 08, 2007, 02:34:03 pm by Wayne Fox »
Logged

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA
expose to the right?
« Reply #26 on: November 08, 2007, 03:43:49 pm »

Quote
I did the experiment that Eric suggested and am somewhat surprised by the results.
.
.
.

Imatest plots show the tonal response curves more clearly and they are different.As expected, the under exposure costs dynamic range, but when the ACR exposure compensation is carried out, the TRCs are different as shown. Comments are invited.

Bill,
Thanks very much for your work here. I really appreciate seeing the imatest chart.  
Eric
Logged
Rolleiflex USA

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
expose to the right?
« Reply #27 on: November 08, 2007, 03:57:42 pm »

Quote
Bill,
Thanks very much for your work here. I really appreciate seeing the imatest chart. 
Eric
[a href=\"index.php?act=findpost&pid=151369\"][{POST_SNAPBACK}][/a]

So this told you what exactly? That they are different and you lose dynamic range from under exposing (as I'd expect). And what about increased noise in the shadows? Also, ISO plays a pretty big role here!
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA
expose to the right?
« Reply #28 on: November 08, 2007, 04:06:09 pm »

Quote
Bill,
Thanks very much for your work here. I really appreciate seeing the imatest chart. 
Eric
[a href=\"index.php?act=findpost&pid=151369\"][{POST_SNAPBACK}][/a]


Since I have noticed the differences in the look of the images depending on exposure, I have often wondered if the sensors response is not actually completely linear near saturation or if the software handles the conversion in a non-linear way?  Bill just wondering how you had the default curve set in ACR for your tests?  If you used the default settings, then probably a medium curve was used and not a linear conversion? Just wondering if this might account for some of the differences in the TCRs?
Logged
Rolleiflex USA

EricWHiss

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2639
    • Rolleiflex USA
expose to the right?
« Reply #29 on: November 08, 2007, 04:13:35 pm »

Quote
what about increased noise in the shadows? Also, ISO plays a pretty big role here!
[a href=\"index.php?act=findpost&pid=151371\"][{POST_SNAPBACK}][/a]


As I have stated previously, I'm talking about low ISO shots where the DR is not bigger than the range of the camera's sensor, ie. clipping the shadows.  My P20 isn't doing well even with ETTR over ISO 400 but I'm mostly talking about studio use with strobes and I normally am shooting ISO 50.


I care a lot more about how the overall image looks than about some noise in the shadows which most likely will not show up in the prints anyhow.  This was not the case when I shot with my 1D, 1Ds because with those cameras even if I only pushed up the exposure by 1/2 a stop I'd get blotches and banding and other ugly things. But with the MFDB I don't see those things and now can look at the overall image.  That's when I noticed the difference in the normal exposure vs ETTR.
Logged
Rolleiflex USA

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
expose to the right?
« Reply #30 on: November 08, 2007, 04:25:05 pm »

Quote
As I have stated previously, I'm talking about low ISO shots where the DR is not bigger than the range of the camera's sensor, ie. clipping the shadows.

Well if the time comes that reading my piece isn't to laborious for you, you'll see that lower ISO's produce less benefits from ETTR than higher but the results in better data is still there (its just going to be hard to see).

I can't speak again to your converter but with the proper tools, there should be little reason why you can't produce a rendering you prefer with better data. Its pretty easy to add noise to shadows if that's what you like. Removing noise, that aint so easy. Same said of dynamic range. Pretty easy to clip the blacks IF that's the look you want. Extending the range? Not going to happen.

Again, for the last time (and you did start this by saying it might be a stupid question so you brought this onto yourself), ETTR isn't about over exposure, its about the best exposure for providing the best possible data to shape and tone. One could under expose a print paper and let it lay in the developer a lot longer and you might actually prefer that print. But I doubt many here would call that best practices. At the very least, its useful to see how to produce the best possible data for rendering, then if so desire, go ahead and break the rules. But its kind of pointless to argue that under exposure is going to produce a better Raw file for rendering than one exposed correctly.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
expose to the right?
« Reply #31 on: November 08, 2007, 06:00:19 pm »

Quote
So this told you what exactly? That they are different and you lose dynamic range from under exposing (as I'd expect). And what about increased noise in the shadows? Also, ISO plays a pretty big role here!
[a href=\"index.php?act=findpost&pid=151371\"][{POST_SNAPBACK}][/a]

Yes, there is 1 f/stop less dynamic range and 1.4 times more luminance noise with the reduced exposure and 1 stop push, as one would expect. On average the noise follows a Poisson distribution with a factor of sqrt(2). However, in the deep shadows where read noise comes into play, the noise doubles.
Logged

AndreNapier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 422
    • Andre Napier Photography
expose to the right?
« Reply #32 on: November 08, 2007, 06:39:25 pm »

The amount of knowledge that you guys have about these charts and numbers as well as the amount of energy that you spend on it is amazing. I must be the dumbest photographer here. My approach is so unscientific and simple based only on what is pleasing to an eye and aesthetically working in an image. I can spend 20 hours post processing a single pic and never look at any digits. I guess I consider myself lucky as certainly numbers were not what attracted me to photography in the first place.
andre
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
expose to the right?
« Reply #33 on: November 08, 2007, 06:47:12 pm »

Quote
The amount of knowledge that you guys have about these charts and numbers as well as the amount of energy that you spend on it is amazing. I must be the dumbest photographer here. My approach is so unscientific and simple based only on what is pleasing to an eye and aesthetically working in an image. I can spend 20 hours post processing a single pic and never look at any digits. I guess I consider myself lucky as certainly numbers were not what attracted me to photography in the first place.
andre
[a href=\"index.php?act=findpost&pid=151404\"][{POST_SNAPBACK}][/a]

One can use decent empirical testing methods AND just look at images.

Least you forget, all those lovely images you're viewing on screen are just a big pile of 1's and zero's.

You can go back to analog photography (I don't know how old you are) and get all caught up in chemistry if you want. Numbers, chemicals, they don't make good images but without them, you have NO images.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

jjj

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4728
    • http://www.futtfuttfuttphotography.com
expose to the right?
« Reply #34 on: November 08, 2007, 08:10:13 pm »

Quote
The amount of knowledge that you guys have about these charts and numbers as well as the amount of energy that you spend on it is amazing. I must be the dumbest photographer here. My approach is so unscientific and simple based only on what is pleasing to an eye and aesthetically working in an image. I can spend 20 hours post processing a single pic and never look at any digits. I guess I consider myself lucky as certainly numbers were not what attracted me to photography in the first place.
andre
[a href=\"index.php?act=findpost&pid=151404\"][{POST_SNAPBACK}][/a]
Shame on you Andre, judging simply by eye is simply not on. It doesn't count unles you pixel peeep. Next, you'll be calling yourself an artist!  

I'd also observe that I've rarely seen any interesting/creative images produced by those overly concerned with the very techy aspects of photography. There are the odd exceptions.
Logged
Tradition is the Backbone of the Spinele

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
expose to the right?
« Reply #35 on: November 08, 2007, 08:14:00 pm »

Quote
I'd also observe that I've rarely seen any interesting/creative images produced by those overly concerned with the very techy aspects of photography. There are the odd exceptions.
[a href=\"index.php?act=findpost&pid=151413\"][{POST_SNAPBACK}][/a]

Like Ansel Adam’s? Or Pete Turner? Or Greg Gorman? Or Douglas Dubler (to name just a few).
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
expose to the right?
« Reply #36 on: November 08, 2007, 08:16:29 pm »

Quote
Since I have noticed the differences in the look of the images depending on exposure, I have often wondered if the sensors response is not actually completely linear near saturation or if the software handles the conversion in a non-linear way?  Bill just wondering how you had the default curve set in ACR for your tests?  If you used the default settings, then probably a medium curve was used and not a linear conversion? Just wondering if this might account for some of the differences in the TCRs?
[a href=\"index.php?act=findpost&pid=151376\"][{POST_SNAPBACK}][/a]

Yes, the default ACR has a medium curve. If you make the ACR conversions completely linear by setting brightness and contrast to zero and setting the point curve to linear, the two results become very similar. In this situation, ACR does not vary the TRC according to the content of the image.

[attachment=3762:attachment]
Logged

AndreNapier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 422
    • Andre Napier Photography
expose to the right?
« Reply #37 on: November 08, 2007, 08:17:18 pm »

Quote
One can use decent empirical testing methods AND just look at images.

Least you forget, all those lovely images you're viewing on screen are just a big pile of 1's and zero's.

You can go back to analog photography (I don't know how old you are) and get all caught up in chemistry if you want. Numbers, chemicals, they don't make good images but without them, you have NO images.
[a href=\"index.php?act=findpost&pid=151405\"][{POST_SNAPBACK}][/a]

Andrew,
With all do respect zero's and one's has as much to do with good images as an ink has to do with good book. You right, without an ink there is no book.
I shot analog for almost 20 years. I shoot strictly MF digital for last 8 years. I operate PS since the PS2.
I consider Photoshop an artistic tool and know it as good as the other guy who spends 6-8 hours a day post processing. When working in darkroom I never cared to know chemical composition of my developers but sure knew how to use them and get the best of each one combo. I knew it because I looked at the results - the prints. Knowing how many atoms of sulphur it takes was left to others, others who look at beautiful fully commercially marketable image and check for signs of banning in the shadows.

Andre
Logged

Wayne Fox

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4237
    • waynefox.com
expose to the right?
« Reply #38 on: November 08, 2007, 08:19:10 pm »

Quote
I did the experiment that Eric suggested and am somewhat surprised by the results. I had some exposures of a Stouffer step wedge. The first was exposed so that step 1 has a pixel value of 250 when converted with ACR at default settings into aRGB. This is exposed to the right as far as possible. A second exposure placed the same step at 221, and this step was brought back to 250 in ACR, which required an exposure compensation of +1.05 EV.

OK, I guess I'm missing something here.  To me you seem to be testing opposite of what I do with ETTR.

I  expose the image to push all values as high as possible without blowing pixels, and typically use a -0.5 to -1.5 EV adjustment to pull the exposure into normal range.  This sounds like you are trying to push the exposure of the normally exposed image to match the ETTR image.

I would assume the only purpose to do a test like this is to confirm that using ETTR and pulling down the exposure in a RAW converter would be a very linear process ... if it isn't very linear then ETTR doesn't work.  I guess you can assume it is linear in either direction, but seems logical that you would pull down the over exposed image to make the test rather than push up the normal one, which would duplicate what ETTR does.  May make no difference what so ever .. I have no clue.

What am I missing?
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
expose to the right?
« Reply #39 on: November 08, 2007, 08:32:21 pm »

Quote
With all do respect zero's and one's has as much to do with good images as an ink has to do with good book. You right, without an ink there is no book.
[a href=\"index.php?act=findpost&pid=151416\"][{POST_SNAPBACK}][/a]

Exactly! You may not care about how the ink or the chemicals were made nor should you have to assuming someone else has taken the care and time to do this. As we both agree, without ink, there's no book, without numbers, there is no digital image. But you now control a great deal more of the process than before. You create the numbers, you don't create the film stock or the chemicals. Exposing for digital simply means you are aware of your actions when building numbers, just as exposing for the shadows, developing for the highlights (even if the development wasn't something you personally did) played a role in the quality of your image.

Yes, you'd be better off artistically being half a stop under and capturing a killer image than nailing your exposure and shooting something that isn't at all interesting. But you'd be better off exposing properly while making an artistically beautiful image. The two don't have to be mutually exclusive. I sometimes find this is a straw man argument from those would prefer to ignore best practices in any field (and I'm NOT talking about you specifically). When someone discusses proper technique, others will dismiss this as not being part of the artistic process. Well it is, to some degree.

So having art enter this particular technical discussion is just a distraction (for those of you who are going into that direction). Let's examine the original question here. Its about proper exposure and the belief that medium format is somehow different from DSLR's in this respect. I don't buy it because of the 1's and Zero's. The math is undeniable.

One can say, I don't care, I prefer the way I'm exposing by not using ETTR. That's cool. But as I've tried to point out to others lurking here, you should fully understand what's going on under the hood and probably, like me, test this yourself. I didn't write a fluff piece about how wonderful ETTR is, it clearly has issues (which I'll add, many but not all could be fixed by the manufacturers).
« Last Edit: November 08, 2007, 08:34:18 pm by digitaldog »
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".
Pages: 1 [2] 3 4 ... 9   Go Up