Ad
Ad
Ad
Pages: « 1 [2]   Bottom of Page
Print
Author Topic: Phase One vs. Adobe Camera RAW comparison  (Read 6337 times)
teddillard
Sr. Member
****
Offline Offline

Posts: 664


WWW
« Reply #20 on: March 28, 2010, 06:57:50 PM »
ReplyReply

Quote from: Schewe
So, if you don't like the term "profile" what would YOU suggest?

I'd suggest getting a grip.  
Logged

Ted Dillard
Schewe
Sr. Member
****
Offline Offline

Posts: 5544


WWW
« Reply #21 on: March 28, 2010, 07:07:16 PM »
ReplyReply

Quote from: teddillard
I'd suggest getting a grip.


Well, if you are going to engage in professional software review, it would be useful if you got the nomenclature correct when you write about it, ya know?
Logged
teddillard
Sr. Member
****
Offline Offline

Posts: 664


WWW
« Reply #22 on: March 29, 2010, 04:01:15 AM »
ReplyReply



My nomenclature is correct.  "Camera Profile" is also the nomenclature used in every color management software I've ever used.  "Camera Profile" has been used as long as I can remember in ACR, yet it's not a conventional ICC profile.  

Maybe to Adobe this doesn't seem confusing.  To me (and the people I work with, my students, clients, photographers and readers), it is, and has been since day one, confusing and unnecessary especially in the context of other RAW processors that do, in fact, allow ICC profile integration in the processing.  

Phase One C1Pro uses the nomenclature "ICC Profile".  Very clear.  



The purpose of nomenclature is to clarify, not obfuscate.  

ya know?
« Last Edit: March 29, 2010, 05:45:50 AM by teddillard » Logged

Ted Dillard
imagetone
Jr. Member
**
Offline Offline

Posts: 53


WWW
« Reply #23 on: April 09, 2010, 03:08:25 PM »
ReplyReply

In the section on sharpening the C1 Pro screenshot with the tree branch shows some CA which can be analysed and removed simply in C1 Pro. I presume this wasn't done.

Tony May

Logged

teddillard
Sr. Member
****
Offline Offline

Posts: 664


WWW
« Reply #24 on: April 09, 2010, 03:22:07 PM »
ReplyReply

Quote from: imagetone
In the section on sharpening the C1 Pro screenshot with the tree branch shows some CA which can be analysed and removed simply in C1 Pro. I presume this wasn't done.

Tony May

That's correct, Tony, it was not done.
Logged

Ted Dillard
Christopher
Sr. Member
****
Offline Offline

Posts: 944


WWW
« Reply #25 on: April 10, 2010, 04:58:10 PM »
ReplyReply

Quote
Both were used at their "default" levels of sharpening. Actually, everything else was at the default level too...

Sorry but that alone makes the whole review senseless. It never makes sense to compare defaults.
Logged

Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 7126


WWW
« Reply #26 on: April 10, 2010, 05:06:28 PM »
ReplyReply

Quote from: Christopher
Sorry but that alone makes the whole review senseless. It never makes sense to compare defaults.

Indeed, as I mentioned in post #7.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
teddillard
Sr. Member
****
Offline Offline

Posts: 664


WWW
« Reply #27 on: April 11, 2010, 05:44:53 AM »
ReplyReply

Quote from: Christopher
Sorry but that alone makes the whole review senseless. It never makes sense to compare defaults.

I'd be interested in hearing what you feel makes sense, then.  I feel like I've discussed our strategy pretty thoroughly above, in response to Mark's comments, and although I certainly feel that more can (and will) be done in comparing these processors more completely, the "defaults" review is at the very least a good starting point to comparing the two packages.  Far from "senseless"... and very much appreciated by many people who have contacted me directly.  

Considering the number of controls, and thus variables, in any processing system I could see doing a complete Head-2-Head on every feature- say, sharpening, color controls, scaling, speed, "Iron Chef" IQ, and on and on...  Then there would be the assorted combinations of the controls- let's compare scaling plus sharpening too...  and don't forget that every camera's files are handled slightly differently.

Then, of course, if anyone actually reads the entire thing (I figure it's about a year's worth of work...), there'd be the inevitable questions...  "why didn't you do the whole series of tests at ISO 400?  Why didn't you compare my camera?"  And after a year, I'd have to hear Jeff say that the software was old...  

 

...sorry, without any additional contribution, your comment is just is sounding a little silly to me at this point.
« Last Edit: April 11, 2010, 05:48:25 AM by teddillard » Logged

Ted Dillard
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 7126


WWW
« Reply #28 on: April 11, 2010, 08:38:58 AM »
ReplyReply

Quote from: teddillard
...sorry, without any additional contribution, your comment is just is sounding a little silly to me at this point.

Ted, it might sound silly to you, but not to me. "Defaults" are manufacturers' best estimates of what they think their clients may find "pleasing" out of the box. As judgments about this can well vary between manufacturers, they are not a reliable basis for an apples-to apple-comparison, and for serious users, defaults have little operational significance. I'm not saying that doing an apples-to-apples comparison for this kind of thing is easy - I know it isn't - but respecting scientific procedure, one should explicitly and consciously normalize as many variables as possible to make the comparison valid and meaningul in respect of the performance parameters one is trying to evaluate.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Jack Flesher
Sr. Member
****
Offline Offline

Posts: 2595



WWW
« Reply #29 on: April 11, 2010, 09:51:17 AM »
ReplyReply

My FWIW .02 --- and the scientist/engineers won't like it because of uncontrolled variables.  But IMHO the only way you can really compare is to have users facile with both programs produce the "best" file possible from the same image and compare the output at 100% screen and on print.  The capture should contain a minimum a color checker and detail at various frequencies like a real world capture might. (Even could add a Koren wedge for the pixelists...)  Finally, make the raw file available so anybody who thinks they can do "best" with their favorite program -- even one that isn't yet official -- can try and back their statement up that with shared results.  Also, making the raw available should help eliminate all complaints of improper processing techniques, as the complainers can easily post their own "improved upon" conversions for critique...

Cheers and good luck,
Logged

Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 7126


WWW
« Reply #30 on: April 11, 2010, 10:41:02 AM »
ReplyReply

Quote from: Jack Flesher
My FWIW .02 --- and the scientist/engineers won't like it because of uncontrolled variables.  But IMHO the only way you can really compare is to have users facile with both programs produce the "best" file possible from the same image and compare the output at 100% screen and on print.  The capture should contain a minimum a color checker and detail at various frequencies like a real world capture might. (Even could add a Koren wedge for the pixelists...)  Finally, make the raw file available so anybody who thinks they can do "best" with their favorite program -- even one that isn't yet official -- can try and back their statement up that with shared results.  Also, making the raw available should help eliminate all complaints of improper processing techniques, as the complainers can easily post their own "improved upon" conversions for critique...

Cheers and good luck,

Jack, this is basically the approach (but with a "real world photograph") which Andrew Rodney used in his "Iron Chef" panels at PPE, on one of which I participated several years ago as a panelist persenting some findings on aspects of colour correction with Camera Raw. I think it's a fully valid exercise - it included Capture, ACR, Aperture and Bibble, and essentially showed that in the right hands any of these programs can do a fully satisfactory job of converting a raw file to the taste of the user. So if that's what one wants to know, it's been done and the results are known. Of course as time marches on there are new versions and new tricks to tickle the urge for re-testing. In this approach, of course, any concept of defaults is also meaningless, because the program operators ignore them and make whatever adjustments the programs allow and they think are needed to optimize the image.

The other approach - "scientific" or let us say "normalizing variables" serves a somewhat different purpose - to compare how the various programs respond to similar adjustments of given variables from comparable (as feasible) base points.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
teddillard
Sr. Member
****
Offline Offline

Posts: 664


WWW
« Reply #31 on: April 11, 2010, 06:42:23 PM »
ReplyReply

Quote from: Mark D Segal
Ted, it might sound silly to you, but not to me. "Defaults" are manufacturers' best estimates of what they think their clients may find "pleasing" out of the box. As judgments about this can well vary between manufacturers, they are not a reliable basis for an apples-to apple-comparison, and for serious users, defaults have little operational significance. I'm not saying that doing an apples-to-apples comparison for this kind of thing is easy - I know it isn't - but respecting scientific procedure, one should explicitly and consciously normalize as many variables as possible to make the comparison valid and meaningul in respect of the performance parameters one is trying to evaluate.

Point taken, as it has been earlier...  and I appreciate your opinion.  What sounded silly to me was simply Christopher's comment, without any further elaboration...  as I said.  Not your opinion, or the considerable effort you've made to communicate it.  (I still find it impractical, however...   )
« Last Edit: April 11, 2010, 06:44:35 PM by teddillard » Logged

Ted Dillard
Schewe
Sr. Member
****
Offline Offline

Posts: 5544


WWW
« Reply #32 on: April 11, 2010, 09:09:25 PM »
ReplyReply

Quote from: Mark D Segal
The other approach - "scientific" or let us say "normalizing variables" serves a somewhat different purpose - to compare how the various programs respond to similar adjustments of given variables from comparable (as feasible) base points.

And it would be useful to be comparing current "state of the art", not stuff that Adobe will be NOT selling as of Mon April 12th...Camera Raw 6 and Lightroom 3 will throw down the gauntlet with regards to Image Quality (IQ) so it makes sense to do objective non-Default types of comparisons early not late (or at the very, very of of) in the product life...comparing Capture One 5.1 to Camera Raw 5.6/Lightroom 2.6 is pretty much a day late and a dollar short...cause as of now (Mon the 12th), the measuring stick has been changed...
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 7126


WWW
« Reply #33 on: April 11, 2010, 09:35:50 PM »
ReplyReply

Quote from: Schewe
And it would be useful to be comparing current "state of the art", not stuff that Adobe will be NOT selling as of Mon April 12th...Camera Raw 6 and Lightroom 3 will throw down the gauntlet with regards to Image Quality (IQ) so it makes sense to do objective non-Default types of comparisons early not late (or at the very, very of of) in the product life...comparing Capture One 5.1 to Camera Raw 5.6/Lightroom 2.6 is pretty much a day late and a dollar short...cause as of now (Mon the 12th), the measuring stick has been changed...

Yes, and the timing is good from the Capture-1 perspective as well, insofar as that program was up-dated quite recently; hence, both applications are in pretty good sync with respect to the proximity in time of their latest versions.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
teddillard
Sr. Member
****
Offline Offline

Posts: 664


WWW
« Reply #34 on: April 12, 2010, 04:44:09 AM »
ReplyReply

Quote from: Schewe
And it would be useful to be comparing current "state of the art", not stuff that Adobe will be NOT selling as of Mon April 12th...Camera Raw 6 and Lightroom 3 will throw down the gauntlet with regards to Image Quality (IQ) so it makes sense to do objective non-Default types of comparisons early not late (or at the very, very of of) in the product life...comparing Capture One 5.1 to Camera Raw 5.6/Lightroom 2.6 is pretty much a day late and a dollar short...cause as of now (Mon the 12th), the measuring stick has been changed...

Good job poking your head up on the launch date, Jeff...  I'm sure we all needed the reminder.  (Can you invoice Adobe now?  )

Logged

Ted Dillard
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 7126


WWW
« Reply #35 on: April 12, 2010, 07:29:03 AM »
ReplyReply

Ted, you diminish yourself with remarks like this.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
teddillard
Sr. Member
****
Offline Offline

Posts: 664


WWW
« Reply #36 on: April 12, 2010, 07:39:00 AM »
ReplyReply

Quote from: Mark D Segal
Ted, you diminish yourself with remarks like this.

Sorry, just too easy.
Logged

Ted Dillard
bjanes
Sr. Member
****
Offline Offline

Posts: 2882



« Reply #37 on: April 12, 2010, 10:46:20 AM »
ReplyReply

Quote from: Schewe
And it would be useful to be comparing current "state of the art", not stuff that Adobe will be NOT selling as of Mon April 12th...Camera Raw 6 and Lightroom 3 will throw down the gauntlet with regards to Image Quality (IQ) so it makes sense to do objective non-Default types of comparisons early not late (or at the very, very of of) in the product life...comparing Capture One 5.1 to Camera Raw 5.6/Lightroom 2.6 is pretty much a day late and a dollar short...cause as of now (Mon the 12th), the measuring stick has been changed...
One can usefully compare the new versus the old algorithms to determine how much has changed. Shown are Imatest analyses of an image taken with the Nikon D3 and the 50 mm f/2.8 AFS lens with no sharpening using ACR 5.6 and LR 3 beta 2. There is virtually no difference. Since the D3 uses a blur filter, sharpening is absolutely necessary to obtain the best results, and optimal sharpening would require optimal settings and experimentation. Without sharpening MTF 50 is only 0.274 cycles per pixel, slightly more than half of the Nyquist limit of 0.5 cy/px.

[attachment=21460:03_ACR_LRB3.png]

With this image, I got the best sharpening results using deconvolution image restoration with Focus Magic and the result of 0.398 cy/px ableit with some increase in aliasing near Nyquist is reasonable. I did not test sharpening in LR 3 B2, since I am not that familiar with that program.

[attachment=21461:003_ACR_...BR12_cpp.png]

Here are the images with ACR 5.6 and no sharpening, ACR default sharpening, and the Focus Magic 1 pixel and 100%. The ACR default sharpening is likely not the optimum setting, but this exercise demonstrates the difficulties inherent in such testing. For a really valid test one would have to have a "shoot out" as Jack Flesher has proposed.

[attachment=21462:CompositeShFMb.png]






Logged
Schewe
Sr. Member
****
Offline Offline

Posts: 5544


WWW
« Reply #38 on: April 12, 2010, 10:57:41 AM »
ReplyReply

Quote from: teddillard
Good job poking your head up on the launch date, Jeff...  I'm sure we all needed the reminder.  (Can you invoice Adobe now?  )

That the best ya got bud? Really?

Go back and redo your review next month when CS5 and Camera Raw 6 ships...
Logged
teddillard
Sr. Member
****
Offline Offline

Posts: 664


WWW
« Reply #39 on: April 12, 2010, 12:00:49 PM »
ReplyReply

Quote from: Schewe
That the best ya got bud? Really?

Go back and redo your review next month when CS5 and Camera Raw 6 ships...

That's the best I'm gonna give you, bud...  

But seriously folks...  I am looking forward to getting my hands on the thing and whatever Phase has by the time I can get it, and doing a comparison of not only them, but comparing them with the previous versions.  'Course, I woulda have to have done the previous versions- which was part of the purpose of the exercise.
« Last Edit: April 12, 2010, 12:02:01 PM by teddillard » Logged

Ted Dillard
Pages: « 1 [2]   Top of Page
Print
Jump to:  

Ad
Ad
Ad