google / fonts

Font files available from Google Fonts, and a public issue tracker for all things Google Fonts
https://fonts.google.com
18.15k stars 2.61k forks source link

Update Kumbh Sans with contrast axis #4696

Closed RosaWagner closed 1 year ago

RosaWagner commented 2 years ago

upstream repo: https://github.com/xconsau/KumbhSans

p1

RosaWagner commented 2 years ago

I need confirmation that we don't need to register a new axis for the font, but would use the YOPQ axis (https://github.com/googlefonts/axisregistry/blob/main/Lib/axisregistry/data/y_opaque.textproto)

cc @davelab6 @chrissimpkins @xconsau

xconsau commented 2 years ago

Thanks for looking into this, Rosalie. The current (proposed) axis is Contrast Geometric (CTGO) with range 0-100. If we use YOPQ (which also seems almost close to what results we have with the current axis), then the range need to be changed from -2000 to 1000 with 166 as default.

vv-monsalve commented 2 years ago

HI @xconsau, From what I see on KumbhSans[CTGO,wght].ttf, the range used in the font (8--72) is already covered by the YOPQ - Thin Stroke parametric axis (-1000--2000), so it wouldn't be an issue. Otoh, the default_value (166) could be overridden in the METADATA.pb file of the font. So YOPQ axis seems to work fine for Kumbh Sans.

xconsau commented 2 years ago

Thanks for the heads up, @vv-monsalve. So, am I required to only rename the axis in source file as YOPQ and generate the var ttf as KumbhSans[YOPQ,wght].ttf ?

Also, thank you for the information about METADATA.pb file. I was not aware of this method and was searching across Fonttools documentation for a way for overriding axes and ranges post production.

davelab6 commented 2 years ago

Sounds good everyone

vv-monsalve commented 2 years ago

Also, thank you for the information about METADATA.pb file. I was not aware of this method and was searching across Fonttools documentation for a way for overriding axes and ranges post production.

The default_value is the only one that could be overridden for the time being.

xconsau commented 2 years ago

Thanks @davelab6 and @vv-monsalve .

@RosaWagner, should I change the axis name in source file and generate new ttf? Or it will be handled at your end.

RosaWagner commented 2 years ago

Yes @xconsau you can rename the axis and generate ttf, I'll QA and onboard ;)

RosaWagner commented 1 year ago

@xconsau just tell me when I can onboard ;)

xconsau commented 1 year ago

Sorry for the delay, @RosaWagner . Apart from the axis name change, it was required to add some new glyphs as per latest GF Latin Core glyph requirements. I have added new glyphs and renamed the axis. The VF is here

Fontbakery shows some errors related to YOPQ range defaults. Current range is 8-72 mapped as 0-100. The default value is the maximum value only.

xconsau commented 1 year ago

@RosaWagner Any update on the QA please?

RosaWagner commented 1 year ago

AH sorry, I didn't see your previous notification, will look at it today

RosaWagner commented 1 year ago

@dberlow we would need confirmation from you on this project. We decided to go with YOPQ axis to avoid registering a new axis that would do almost the same thing, although the axis is not fully parametric in this project. Indeed, whilst YOPQ decreases here, it also slightly tighten the glyph width to compensate the lost of "black". Is it something that you would accept as part of the YOPQ axis?

https://user-images.githubusercontent.com/12222436/205296701-a119122d-8dfd-43eb-9035-cc1a77bfdff1.mov

dberlow commented 1 year ago

@RosaWagner maintaining a glyph's side bearings to compensate for the lost of "black" is what I would expect as part of the YOPQ axis, and this changes the width of such glyphs. So I would accept this as an YOPQ axis if it's consistently applied to all the glyphs.

RosaWagner commented 1 year ago

@dberlow great, thank you!

dberlow commented 1 year ago

Not a problem, cheers!

On Fri, Dec 2, 2022 at 8:53 AM Rosalie Wagner @.***> wrote:

@dberlow https://github.com/dberlow great, thank you!

— Reply to this email directly, view it on GitHub https://github.com/google/fonts/issues/4696#issuecomment-1335261184, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAO5VDXZCG2BHDGFHNHLUFDWLH5NXANCNFSM5WMR4O6Q . You are receiving this because you were mentioned.Message ID: @.***>

xconsau commented 1 year ago

Thanks very much, Sir @dberlow for your time and feedback on the issue.

Thank you @RosaWagner for the technical QA and moving the update forward.