Which Gamut is Largest....

[johnorzechowski]

Hello, My scanner is calibrated using the IT-8 profile. I use Silverfast Ai and would like to select the largest practical color profile to embed with my scanned image. Which color profile is the largest/best one to use, 1. the scanners IT-8 color profile, 2. Adobe1998 or 3. ProPhotoRGB?

Is the scanner's IT-8 icc color profile/gamut larger than ProPhotoRGB? Adobe1998?

[LSI_Morales]

Dear John,

You can only embed either Adobe RGB or ProPhoto RGB. The IT8 profile is only used by the scanner itself to digitize the image. The resulting file will then be written in one of the profiles you choose to have for internal convertion.

Is the scanner's IT-8 icc color profile/gamut larger than ProPhotoRGB? Adobe1998?

No, the scanner profile has a smaller gamut than PhoPhoto RGB or Adobe RGB.

Cheers

[johnorzechowski]

Thanks Alejandro....

So, if one scans a 48 bit HDR negative with an IT-8 calibrated scanner profile and converts the resulting scan/file to a larger color space in an editing application, which Silverfast CMS ICC profile embedding option will render a right approach:

1. Embed the Scanner's IT-8 ICC profile and convert to the larger working space in the editing application, or
2. Do not Embed an ICC profile, import the file into the editing application and convert to the larger working space

I ask because in trying both ways with the end goal of both having the same ICC profile between the two, results in slightly different color appearances and histograms between the two files. Not embedding appears to render a more neutral/blue appearance while embedding appears to be a bit warmer/red.

Appreciate any guidance/thoughts....

john

[johnorzechowski]

Hi,

I am preparing to scan 35mm color negatives in 48-bit HDR ("RAW") in Silverfast Ai. Read many threads but need some guidance.

Under Options->CMS, I am able to create an IT-8 calibrated scanner profile for Reflectives. No Problem. BTW-my "Profiles for ColorSync: Internal" is Adobe1998.

However, what should be used for Options->CMS->"Profiles for ColorSynch: Scanner (Transparent)"? If I set "Profiles for ColorSynch: Scanner (Transparent)" to <NONE>, I receive an error message that the setting is missing. Does the user need to create one or download--is so, where can I find more info?

In addition, should a ICC profile be embedded and if so, which one? It appears the Options->CMS->Profiles for ColorSynch->Scanner (Transparent) selection has a bearing on the Options->CMS->Embedded ICC Profiles. Can/Should an ICC profile be embedded?

Thanks,
john

[LSI_Morales]

Dear John,

When you scan in 48 bit HDR mode you are creating an unmodified RAW file (I believe that is what you mean by "negative"). It means the file contains the entire information delivered by the scanner.

If you do not embed the scanner profile, the editing application will not be capable of correctly interpret the file and will assign any color profile which does not accurately match the image.

So the right approach is to always embed the ICC scanner profile into the RAW file.

Cheers

[johnorzechowski]

Hello Alejandro,

Thank You. May I ask for some clarification as follows:

When you scan in 48 bit HDR mode you are creating an unmodified RAW file (I believe that is what you mean by "negative"). It means the file contains the entire information delivered by the scanner.

Yes, I am scanning in 48 bit HDR to create an unmodified "RAW" file--I'll call that a "linear scan". Moreover, I am scanning a color 35mm negative in NEGATIVE mode.

If you do not embed the scanner profile, the editing application will not be capable of correctly interpret the file and will assign any color profile which does not accurately match the image.

Two questions:
1. I have read in the forums that profiles, in particular the SF_T [scanner specific] profile, can not be assigned to color NEGATIVE (HDR) scans--that the current SF_T (scanner) profile should apply only to POSITIVE scans, accurate?
2. If not true and the SF_T profile can be assigned to NEGATIVE mode linear scans, I am interested to better understand the advantage of an embedded SF_T [scanner specific] profile vs. just delivering a "RAW" linear scan with no embedded profile to the editing application and applying a wide gamut profile there. What does the SF_T [scanner specific] profile do to the linear scan data that the editing application doesn't--isn't it just one more step?

So the right approach is to always embed the ICC scanner profile into the RAW file.

Even for a color negative scanned in HDR negative mode?

Cheers[/quote]
Hi Alejandro...

[LSI_Morales]

Hi John

1. I have read in the forums that profiles, in particular the SF_T [scanner specific] profile, can not be assigned to color NEGATIVE (HDR) scans--that the current SF_T (scanner) profile should apply only to POSITIVE scans, accurate?

It is accurate, the SF_T scanner profile is the profile for positive transparent material.

I am interested to better understand the advantage of an embedded SF_T [scanner specific] profile vs. just delivering a "RAW" linear scan with no embedded profile

The profile is not applied to negative material but it tells SilverFast HDR which scanner was used to digitze the picture, so you can select the proper NegaFix profile to do the conversion to positive.

If you are not doing the conversion to positive using SilverFast HDR but other editing applications, you might leave the pictures without embedding any profile.

Cheers

[johnorzechowski]

If you are not doing the conversion to positive using SilverFast HDR but other editing applications, you might leave the pictures without embedding any profile.

Thank You, Alejandro...

When scanning color 35mm negatives in Silverfast Ai 48-bit HDR (RAW) for archival, is your statement above applicable and preferable to BOTH NEGATIVE and POSITIVE mode scans? Other than Silverfast HDR editing, is there any reason to embed the SF_T scanner profile?

[LSI_Morales]

Hi John

is your statement above applicable and preferable to BOTH NEGATIVE and POSITIVE mode scans? Other than Silverfast HDR editing, is there any reason to embed the SF_T scanner profile?

I never stated POSITIVE material, as a matter of fact I specifically asked and stated only negatives.

The SF_T scanner profile should always be embedded when scanning positive transparent material (i.e. slides).

Cheers

[johnorzechowski]

Thank You.....yes, you are correct about never stating positive material. The question was referring only to negative material scanned in HDR 48 bit color, either negative and/or positive mode. Here's why....

When scanning color negative material in either a positive or negative mode, AND the latter assuming that "auto film mask detection" is not invoked in negative mode, the resulting scan should deliver a full RGB tonal range (0-255). The net result is that there does not appear to be any discernible differences between images scanned from negative material in either POS or NEG (No Auto Film Mask Detection) mode.

If scanning for archival reasons, given storage space is not a limitation, and scanning HDR with a full tonal range (in POS or NEG mode) would appear to be the preferred workflow. Is there any reason to apply the "orange mask" auto film mask detection when scanning negative material in negative mode (notwithstanding larger files and storage)?

Embedding the SFprof_T profile in either mode for color negative material does appear to render a "native" color space with HDR scans. Embedding that profile with the HDR scans appears a particularly good workflow implementation; especially when Silverfast HDR is used as the post-process editor.

So, full circle to gamuts....when re-assigning a wider gamut color space (such as ProPhoto) in other editing applications, there appears to be much more magenta/red in the resulting images of the target, wide gamuts than SFprof_T; particularly in skin tones.

Is there a standard gamut which closely aligns with the native SFprofT profile when the image is ready to converting yet maintaining a standard system color space?

[LSI_Morales]

Hi John,

Is there any reason to apply the "orange mask" auto film mask detection when scanning negative material in negative mode (notwithstanding larger files and storage)?

Not really, as I already mentioned somewhere else in this thread:

When you scan in 48 bit HDR mode you are creating an unmodified RAW file .... It means the file contains the entire information delivered by the scanner.

It means that if you use the orange mask detection during scanning it will be used only for viewing purposes, no correction will be applied.

Embedding the SFprof_T profile in either mode for color negative material does appear to render a "native" color space with HDR scans. Embedding that profile with the HDR scans appears a particularly good workflow implementation; especially when Silverfast HDR is used as the post-process editor.

It is correct!

So, full circle to gamuts....when re-assigning a wider gamut color space (such as ProPhoto) in other editing applications, there appears to be much more magenta/red in the resulting images of the target, wide gamuts than SFprof_T; particularly in skin tones.

Probably other applications use the wider gamut color space in a different way than SilverFast. It also depends on the rendering intent.

Is there a standard gamut which closely aligns with the native SFprofT profile when the image is ready to converting yet maintaining a standard system color space?

Not really, you should try embedding the color space (although not used with the negative scans), and choose the wide gamut color profile as your internal color profile in SilverFast (i.e. PhoPhoto).

Cheers