Calculate G7 curves with Excel

[cmcfarling]

I've attached an Excel doc that I've been working on that will calculate G7 curves. Before proceeding I'd like to verify that it produces accurate results. Currently the only way to generate G7 curves is to use the expensive ($1200) IDEALink Curve software or the cumbersome (but cheap) fan graph paper, plotting everything manually. I assume this Excel doc would be more accurate that the fan graph paper but less accurate than the Curve software. I'm wondering if someone that has access to the Curve software could run this sample data set through it for comparison.

Currently this will only calculate CMY RIP curves. In addition the max neutral density has to be between 1.28 - 1.56. Attached is a sample data set consisting of neutral density values from column 5 of a P2P target (CMY column). Also included in that file are the calculated RIP curve data points based on a target curve with a max neutral density of 1.28. I'm curious to see how close these are to the RIP curves generated by the Curve software.

If all is well I intend to post the final version here.

Chris McFarling

 

[eran]

Nice Work! I look forward to experimenting with this tomorrow. I attempted to figure this out some time ago but found a harsh barrier within my lack of Excel skills. Cool stuff. If I recall correctly, SpotOn started as an excel spreadsheet as well...

 

[meddington]

I'm inferring that your spreadsheet also assumes gray balance has been achieved, as it only reports a single "curve" of rip correction, instead of separate CMY curves to achieve gray balance and tonality? From a density perspective, the spreadsheet appears to correctly calculate the needed TVI adjustment...i/e the correct input percentage that results in the desired output density, but again, assuming gray balance. So if you can verify that your data is achieving correct gray balance based on your paper's white point, than this might work for you, but you might want to double check with a manual plot before implementing.

I'm wondering if someone that has access to the Curve software could run this sample data set through it for comparison.

Sample data would need to be full P2P target data in CIELab or spectral for Idealink Curve to utilize, so this isn't possible with the data provided.

 

[cmcfarling]

Yes, currently this spreadsheet assumes gray balance has already been achieved.

Here are some new attachments:

G7 P2P Analysis_beta0.2.zip - updated Excel doc that allows selection of the CMY144 curve to allow it to work with the data in the other attachments

CMY Vis Den.txt - CMY density values from column 5 of P2P target for pasting into Excel doc

P2P_NOUV.txt - full P2P target data saved from ColorPort in CGATS format for use with IDEALink Curve

 

[meddington]

Ran your P2P data through Idealink Curve 1.1 (not latest) and it yielded results very close to your Excel sheet, though approaching 2% differences in some areas (see attachment). Though if you turn on gray balance, the recommended CMY curves are larger (see other attachment). Whether or not you'd want to force gray via curves, via SID adjustment, or at all, would be a personal choice at this point.

 

[cmcfarling]

Thanks for taking the time to test this out. That revealed a error I had in some calculations. See attachment for a new comparison.

I'm basically "scraping" the FanGraph jpg file for the curve data points. So it will only be as accurate as the values that can be derived from a 225ppi graph image, which would also be the case with manual plotting. However this has the advantage of repeatability and speed.

If a special (spectro-readable) grayfinder target was used, I believe this could be configured to calculate separate CMY curves as well in a semi-automated manner.