Best Order for Workflow

[mcraige65]

For the sake of simplicity, let's say I'm using LRC and Photo AI. Which is better:

Open in LRC, use File --> Plug-in Extras --> Photo AI (to edit as raw), send to LRC as DNG, crop, edit colors, etc.
-or-
Open in LRC, crop, edit colors, etc., use Photo --> Edit in --> Photo AI (sends to Photo AI as TIFF), save back to LRC

In other words, should I edit raw to crop and fix colors first, then edit in Photo AI as a TIFF or edit in Photo AI as raw, then crop and fix colors in LRC as DNG?

 

[Ernie Misner]

When Photo AI first came out, I think that Topaz recommended the first approach, where Photo AI gets the raw file to work with. A small amount of functionality gets lost when it comes back into Lr as a DNG, but the results could be worth that.

 

[Ab S]

I used Topaz Sharpen and Photo AI as plug-ins to Capture One. Since Topaz may introduce a color shift I always (also) apply C1 editing on the DNG after the Topaz application. Since I am over to DxO with their excellent NR and sharpening in PL6 Elite, I apply Topaz mostly to correct parts which have motion blur and I perform the Topaz correction on the full size jpeg resulting from PL6. If color shifted I use Topaz as a plug in to PL6 (exporting from Topaz back to PL6 as DNG) and do final corrections in PL6.

Cheers,
Ab

 

[archer]

For the sake of simplicity, let's say I'm using LRC and Photo AI. Which is better:

Open in LRC, use File --> Plug-in Extras --> Photo AI (to edit as raw), send to LRC as DNG, crop, edit colors, etc.
-or-
Open in LRC, crop, edit colors, etc., use Photo --> Edit in --> Photo AI (sends to Photo AI as TIFF), save back to LRC

In other words, should I edit raw to crop and fix colors first, then edit in Photo AI as a TIFF or edit in Photo AI as raw, then crop and fix colors in LRC as DNG?

I do everything in LrC first, than send to Photo AI or Topaz Sharpen or Topaz Denoise as TIFF. Works for me.

 

[coldcoffee]

Noise reduction should always be done first on the raw file firstly before any editing IMHO

 

[Ab S]

Noise reduction should always be done first on the raw file firstly before any editing IMHO

Yes, I agree, first noise reduction then the rest, that prevents sharpen algorithms to see noise as objects to be sharpened causing artefacts.

 

[Painted with Light]

I now have prints matching my display, after looking for a solution to this problem for fifteen years.

My workflow:

1. Calibrated 99%Adobe RGB monitor with 5000k colour temperature;
2. Using eciRGB v2* as my working colour space;
3. Matching my editing black and white points to the minimum black and maximum white able to be printed with my printer/paper combination.
- My minimum black producible is 44/256 and maximum white is 244/256. These figures I set as my black and white points throughout editing as sometimes they increase/decrease with edits.

eciRGB v2*:

RGB colour model (AdobeRGB and sRGB colour spaces) is the colour model usually used for sending data to inkjet printing for photos;
CYMK is the usual colour model used by the majority of printers;
eciRGB v2 is the colour space developed, in Europe, to get RGB and CMYK models aligned at the beginning of raw image data processing, for better printing outcomes in an industrial setting and for inkjet printing.

As most people don't work in the CMYK color model, and the majority of industrial printing and inkjet printers do, the European Color Initiative (ECI) developed eciRGB v2 to solve issues with the computerisation of colour data. The issues are/were:

1. AdobeRGB and sRGB colour spaces are subtractive in how they produces millions of colours from Red, Green and Blue;
2. CMYK colour model is additive in how it produces millions of colours from Cyan, Magenta, Yellow and Key(aka Black); and
3. Adobe RGB and sRGB colour spaces are referenced to a 6,500 Kelvin light source, whereas CMYK Gamut colours are referenced to a 5,000 Kelvin light source.

The European Color Initiative worked out the math's, of the RGB and CMYK three dimensional colour models, to convert (render) a subtractive colour model in a 6,500 Kelvin light space, into an additive colour model in a 5,000 Kelvin light space.

When I demosaic raw files I immediately save the file with the eciRGB v2 profile before doing any other processing; and
I send the processed file to printer with eciRGB v2 profile

A printing guru told me eciRGB v2 was irrelevant!