Adobe Raw converter HELP?

Stephen,

First, the camera JPG's are wrong, in the "accuracy" sense. Second, they are "more wrong" in BB. They are a little "less wrong" in Iview. (and, don't ask me why,...I've never put in the time to find out why) Now, I use to shoot Raw and Matrix 3, so that I could have contacts (from the jpgs) that "pop". But, soon enough, that got me in trouble,...especially if the AD took those contacts with them and did direct comparisons to the contract proof. (but, let's not get into that CMYK stuff here)
Now, this could be totally the wrong way to do this, but this is what I do;

Set camera to Matrix 5, closest preset (daylight, tungsten, whatever),
Custom tone curve for 1D, (to take out the green), Standard tone curve for
1Ds.

Overexpose by 2/3-1 stop. (well, not really,...but histogramic hump is well
over to the right on midtones, so if you use matrix meter, it will read
overexposed)

Set ACR2 converter to camera default as such;

TEMP to 5000K (will adjust this as needed for finals,...then batch that
group)

TINT to -1

EXPOSURE to 0 (only other one you would adjust before batching)

SHADOWS to 6

BRIGHTNESS to +55

CONTRAST to +44

SATURATION to -23

SHADOW TINT to 0

RED HUE to +43

RED SAT to 0

GREEN HUE to -5

GREEN SAT to +37

BOTH BLUES to 0

ACR2 output will pretty closely match JPG's viewed in IVIEW, and contacts.....with adjustments to TEMP and EXPOSURE ONLY, whether you use gels, filters, late evening light, etc.

CAVEAT: THIS IS ONLY PER MY EYE AND SKINTONES AS PER MY "NEUTRAL" STARTING
POINT, AND HAS NO RELEVANCE TO THE WORD "ACCURACY". YMMV.

View presets the same as you would "fixed" temp film.

Whether you like this recipe or not is not the point. With no filters or gels, first come up with your own "recipe" that matches one of your JPG's, (the ones without the filters or gels for a neutral starting point) Also, use a typical "skintone" image, not a Macbeth chart. (i.e., I am with you on the "pleasing" as opposed to "accurate" stance)

One other quick note,....C1, ACR1 or 2, and other converters do NOT read Kelvin temperature acurately from cameras. So, trust your eyes. (on your profiled monitor, of course) And don't trust those numbers in the little box.

Regards
Mastrianni
 
And for some reason I cannot open your file, probably a PC issue, I
change the extension and still nothing, is it compressed somehow?
any ideas?
try this: http://www.johnmaclean.com/DFES2820.TIF.zip

I had used Aladdin Stuffit to compress but it probably only works on Mac's, but their zipping works cross platform.

Be warned, my assistant holding the card has no make up or hair styling! ;-)

Best,
John
--
http://www.johnmaclean.com
http://www.nmdigital.com
 
So far that is the closest starting ponit I have seen, thanks you I will start working from there and see where it gets me.

I have noticed over the past few days that 1DS files seem to convert closer to the originals than 1D files, has anyone else had that experience? I had posted a bunch of 1D files for the size of them online, maybe this is more of an issue with 1D files than 1DS.

I will set out to shoot some test samples next week and make some camera settings and see what they get me. One question why shoot in matrix 5?

Stephen
Stephen,

First, the camera JPG's are wrong, in the "accuracy" sense. Second,
they are "more wrong" in BB. They are a little "less wrong" in
Iview. (and, don't ask me why,...I've never put in the time to find
out why) Now, I use to shoot Raw and Matrix 3, so that I could have
contacts (from the jpgs) that "pop". But, soon enough, that got me
in trouble,...especially if the AD took those contacts with them
and did direct comparisons to the contract proof. (but, let's not
get into that CMYK stuff here)
Now, this could be totally the wrong way to do this, but this is
what I do;

Set camera to Matrix 5, closest preset (daylight, tungsten, whatever),
Custom tone curve for 1D, (to take out the green), Standard tone
curve for
1Ds.

Overexpose by 2/3-1 stop. (well, not really,...but histogramic hump
is well
over to the right on midtones, so if you use matrix meter, it will
read
overexposed)

Set ACR2 converter to camera default as such;

TEMP to 5000K (will adjust this as needed for finals,...then batch
that
group)

TINT to -1

EXPOSURE to 0 (only other one you would adjust before batching)

SHADOWS to 6

BRIGHTNESS to +55

CONTRAST to +44

SATURATION to -23

SHADOW TINT to 0

RED HUE to +43

RED SAT to 0

GREEN HUE to -5

GREEN SAT to +37

BOTH BLUES to 0

ACR2 output will pretty closely match JPG's viewed in IVIEW, and
contacts.....with adjustments to TEMP and EXPOSURE ONLY, whether
you use gels, filters, late evening light, etc.

CAVEAT: THIS IS ONLY PER MY EYE AND SKINTONES AS PER MY "NEUTRAL"
STARTING
POINT, AND HAS NO RELEVANCE TO THE WORD "ACCURACY". YMMV.

View presets the same as you would "fixed" temp film.

Whether you like this recipe or not is not the point. With no
filters or gels, first come up with your own "recipe" that matches
one of your JPG's, (the ones without the filters or gels for a
neutral starting point) Also, use a typical "skintone" image, not a
Macbeth chart. (i.e., I am with you on the "pleasing" as opposed to
"accurate" stance)
One other quick note,....C1, ACR1 or 2, and other converters do NOT
read Kelvin temperature acurately from cameras. So, trust your
eyes. (on your profiled monitor, of course) And don't trust those
numbers in the little box.

Regards
Mastrianni
--
Stephen Eastwood
http://www.nyphotographics.com
 
First, the camera JPG's are wrong, in the "accuracy" sense.
Federico,

OK this is where you lose me. The jpgs are wrong? By what standard? The standard of a color chart set on a profiled monitor, because I know you and you are not going to tell me that you ever shoot a color chart.

Trust our eyes, cool that's what we do, but where do my eyes go. The lcd, Iview, BB, PS, C1, tft, crt, plasma? Where? I am not talking conversion, printing or purposing the file in cmyk, rgb or whatever, I am talking what do we trust our eyes to?

I want a perfect lcd that looks like the "film". Period. I want the contacts to match the lcd which looks like the film. I want . . . .

I am capable of moving sliders and setting numbers and yes trusting my eyes and my heart, but I would like to NOT live next to a computer. I would love to hand someone a cd at the end of the day that when they popped it up on their computer it looked like my vision and not some jpeg, or raw file that needs 22 color correction settings, three secret sharpening methods and a blend layer of grain.

I will admit ACR II, or CS or 8 or whatever the heck it is called is a lifesaver compared to anyone else's offerings, but still, there is no harm in making these things look good straight out of the camera.

Someday, somebody is going to make the digital film camera that allows you to shoot a lcd polaroid and tweak a knob on the back until the lcd looks the way you want.

Bam.

Shoot, deliver, chat up the models, go to dinner.

Best,

James Russell
 
First, the camera JPG's are wrong, in the "accuracy" sense.
Federico,

OK this is where you lose me. The jpgs are wrong? By what
standard? The standard of a color chart set on a profiled monitor,
because I know you and you are not going to tell me that you ever
shoot a color chart.
When it comes to people, I don't know any jpgs that are all that accurate,...uh, or rather, pleasing. (except maybe the S2 when all the stars are aligned)
Trust our eyes, cool that's what we do, but where do my eyes go.
The lcd, Iview, BB, PS, C1, tft, crt, plasma? Where? I am not
talking conversion, printing or purposing the file in cmyk, rgb or
whatever, I am talking what do we trust our eyes to?
Well, if your monitor matches your output, and you like what you see on your monitor, then all is well and the world is good.
I want a perfect lcd that looks like the "film". Period. I want
the contacts to match the lcd which looks like the film. I want .
. . .
I just want it to look good,...I don't care if it "looks like film". And what film, pray tell, do you speak of? Because Astia doesn't look like Velvia, or Provia, and E100s doesn't look like E100sw or EPN,...so,...
I am capable of moving sliders and setting numbers and yes trusting
my eyes and my heart, but I would like to NOT live next to a
computer. I would love to hand someone a cd at the end of the day
that when they popped it up on their computer it looked like my
vision and not some jpeg, or raw file that needs 22 color
correction settings, three secret sharpening methods and a blend
layer of grain.
Well, I don't know other peoples vision of what they deem good. I do know you at one time were happy with just the jpgs right out of the S2. In reality, we could all shoot jpgs and be quite happy,...assuming we had a definitive idea before we ever clicked the shutter, and accomplished that idea with traditional photographic skills. Remeber that Sisley thing Mark sent us? Well, there's a guy who obsesses about the "right" color, tones, etc., (as many do), yet raved about those images. In reality, the color on those images is waaaay too yellow,...not even pleasing yellow,...yet, he overlooked that on someone elses work. Why? Because when most people see images, they either feel an impact from them or not. They don't actually start comparing them to other ones that may or may not have better skin tones. Color is actually only a very small component of imaging, and I just fear that we, (including "I"), may give it more time, rework and consideration than it is worth.
I will admit ACR II, or CS or 8 or whatever the heck it is called
is a lifesaver compared to anyone else's offerings, but still,
there is no harm in making these things look good straight out of
the camera.
I've sent clients stuff from BB, and it was perfectly fine,....even got compliments from them on it. And for $35 (what I paid a few years ago) and endless free upgrades, it is quite awesome. So I can understand some peoples concerns with the various programs, especially high priced ones, falling just a bit short. But, I don't ever remember doing any gallery shows where someone asked me what converter I used. We all need to keep our eyes on the prize,...the final image. Not the pixels, or software, or what brand equipment we use.
Someday, somebody is going to make the digital film camera that
allows you to shoot a lcd polaroid and tweak a knob on the back
until the lcd looks the way you want.
What's a "digital film" camera? Isn't that an oxymoron?
Bam.

Shoot, deliver, chat up the models, go to dinner.
I'll drink to that. Chivas, anyone?

Mastrianni
 
camera settings and see what they get me. One question why shoot
in matrix 5?
Note the desaturation number in ACR,......this matches more closely with the low chroma of Matrix 5. (a very subdued color matrix,...especially in the reds and yellows,...but you can make up for it with a higher kelvin shooting temp.

Regards
Mastrianni
 
Even though shooting Nikon myself, I'm jumping in since this is all about ARC (2). I've tried Thomas Knoll's adjustment suggestions on your files(which looked great BTW after converting) and on some of my own. Personally I find those much more useable than those of Federico's since Thomas' suggestions don't include overexposure and are more simple. I just will never buy it that a RAW convertor needs so many adjustments to come close to the first JPEG. Since Thomas' adjustments work on your files and my Nikon files too, my conclusion is that ARC2 is just off and needs default settings adjusted in the way Thomas suggested, since then you don't have to adjust and save your settings as personal ones. It will work for everyone from scratch. It wouldn't hurt if Adobe would admit the default settings are wrong and change it in an upgrade. I've seen many suggestions about using ARC2, but only after Thomas' suggestions, I'm now able to use the eyedropper for adjusting WB, even on black and white, not only on grey. And get natural colours, sometimes even better (personal thing ofcourse) than what Nikon Capture gives me.
Thanks for starting this thread, it's been most helpful
Kind regards,
Bart
I have noticed over the past few days that 1DS files seem to
convert closer to the originals than 1D files, has anyone else had
that experience? I had posted a bunch of 1D files for the size of
them online, maybe this is more of an issue with 1D files than 1DS.

I will set out to shoot some test samples next week and make some
camera settings and see what they get me. One question why shoot
in matrix 5?

Stephen
Stephen,

First, the camera JPG's are wrong, in the "accuracy" sense. Second,
they are "more wrong" in BB. They are a little "less wrong" in
Iview. (and, don't ask me why,...I've never put in the time to find
out why) Now, I use to shoot Raw and Matrix 3, so that I could have
contacts (from the jpgs) that "pop". But, soon enough, that got me
in trouble,...especially if the AD took those contacts with them
and did direct comparisons to the contract proof. (but, let's not
get into that CMYK stuff here)
Now, this could be totally the wrong way to do this, but this is
what I do;

Set camera to Matrix 5, closest preset (daylight, tungsten, whatever),
Custom tone curve for 1D, (to take out the green), Standard tone
curve for
1Ds.

Overexpose by 2/3-1 stop. (well, not really,...but histogramic hump
is well
over to the right on midtones, so if you use matrix meter, it will
read
overexposed)

Set ACR2 converter to camera default as such;

TEMP to 5000K (will adjust this as needed for finals,...then batch
that
group)

TINT to -1

EXPOSURE to 0 (only other one you would adjust before batching)

SHADOWS to 6

BRIGHTNESS to +55

CONTRAST to +44

SATURATION to -23

SHADOW TINT to 0

RED HUE to +43

RED SAT to 0

GREEN HUE to -5

GREEN SAT to +37

BOTH BLUES to 0

ACR2 output will pretty closely match JPG's viewed in IVIEW, and
contacts.....with adjustments to TEMP and EXPOSURE ONLY, whether
you use gels, filters, late evening light, etc.

CAVEAT: THIS IS ONLY PER MY EYE AND SKINTONES AS PER MY "NEUTRAL"
STARTING
POINT, AND HAS NO RELEVANCE TO THE WORD "ACCURACY". YMMV.

View presets the same as you would "fixed" temp film.

Whether you like this recipe or not is not the point. With no
filters or gels, first come up with your own "recipe" that matches
one of your JPG's, (the ones without the filters or gels for a
neutral starting point) Also, use a typical "skintone" image, not a
Macbeth chart. (i.e., I am with you on the "pleasing" as opposed to
"accurate" stance)
One other quick note,....C1, ACR1 or 2, and other converters do NOT
read Kelvin temperature acurately from cameras. So, trust your
eyes. (on your profiled monitor, of course) And don't trust those
numbers in the little box.

Regards
Mastrianni
--
Stephen Eastwood
http://www.nyphotographics.com
 
OK well after playing with Adobe Raw I am coming to the conclusion that it descriminates against 1D files, at least in the default settings. The default settings for a 1D are way off, for a 1DS they are slightly off, for a d60 almost right on, for a Kodak 14N again almost right on (bump up the shadows to 5 and its near perfect. Unfortunately, I am not switching to a 14N yet so I will continue to slowly tweak little by little and maybe wait for an improved default, at the moment the best default I have is from Mastrianni, I thank you again for that, it is the closest starting pioint of all I have played with.

If anyone from Adobe is still reading I have a ton of ideas that I would love to share to make this whole system work great feel free to email or call me anytime. And one specific question, is there some way to build into the raw converter a color picker that can adjust for the individual patches on a Macbeth color chart? For that matter is there a way to build into photoshop in general a color picker that would allow you to click on the various squares on the Macbeth color chart and have photoshop adjust curves to the known values in the current color space? While this does not help my original problem of pleasing color vs accurate color this would certainly make accurate color easier to attain. Just a thought.
OK, I have not used the adobe raw converter yet and so I finally
had some time and figured I would give it a try and get use to it.
So I opened some older files and grabbed a raw file to convert, WOW
was it disgusting! So I tried another and again disgusting. So I
was wondering what do I have to do to get the file as close to as
shot as possible? I converted the same file in Breezebrowser
(which I always use) and also Phase one all set to default as shot
Adobe RGB output, all brought into Photoshop and assigned the
proper colorspace and then I placed them on a new canvas to show
you how off they were. Is there any way to get the default to be
as close to as shot as possible? Are there any settings that I
missed? I am not looking to play with every file to get it right,
I get it the way I want it in camera most of the time using gels on
the strobes to achieve the color I want. Here are the samples,
they are named and following that are the files as shot processed
in breeze with matrix 2 (which I usually prefer and only wish
someone would develop a canon in camera curve that can simulate
this on an Adobe RGB shot)

Any help from those of you who are using adobe raw would be
appreciated.

All these are straight as shot samples from the camera.













--
Stephen Eastwood
http://www.nyphotographics.com
--
Stephen Eastwood
http://www.nyphotographics.com
 

Keyboard shortcuts

Back
Top