googlefonts / roboto-flex

SIL Open Font License 1.1
473 stars 32 forks source link

Cyrillic wght: 100-400 | wdth: 100 | slnt: 0 | GRAD: -50 - +50 | opsz: 144 #204

Open vikagrabowska opened 2 years ago

vikagrabowska commented 2 years ago

Here are my observations for opsz: 144 for weights 100 and 400 (I will follow with wght 900 next). Unless mentioned differently the comments apply to whole range of grad -50 - +50.


wght: 100 | wdth: 100 | slnt: 0 | GRAD: -50 - +50 | opsz: 144

  1. шщж - too wide
  2. ПЦ - slightly too wide
  3. Ж - could just a touch narrower
  4. ШЩ - too wide
  5. ф - too narrow
  6. Ы - right stem rendering issue - appears too light or disappears, while in fact the weight is correct.

    wght: 400 | wdth: 100 | slnt: 0 | GRAD: -50 - +50 | opsz: 144

GRAD: -50 - +50

  1. This starts in grad -50, but is most visible in grad 50. БВРЯЗЪЬ - seem pretty narrow in titlecase setting, while Д and especially Л give the opposite feeling. Since БВРЯЗ widths came from relation to Latin letterforms, maybe ДЛ could be narrowed? Breve.cyr.case - slightly dark and could use more contrast Breve.cyr - too wide and slightly dark - can use more contrast.
  2. пн can be slightly wider. This starts in grad -50, but is most visible in grad 50.
  3. Ф - appears wide
  4. Ч - slightly too wide
  5. б (grad 0-50) - can the tail be touch darker?
  6. ШЩ - too wide
  7. ПЦ - slightly too wide
  8. ж - slightly dark
  9. Д - left descender going a little further to the left in grad 0-50, than in grad -50.

Grad 50:

  1. ИЙ appear a little darker in text than the rest
  2. к - feels touch dark
dberlow commented 2 years ago

6 "Ы - right stem rendering issue - appears too light or disappears, while in fact the weight is correct."

I found no rendering issue on the right stem.

SO, contour of 042C, "Soft" appears to be reversed.

I will be going over the rest of the review later. However, I can respond to all issues of widths, as per this illustration of agreed upon as correct application and removal of weight to make a grade axis:

Screen Shot 2022-02-06 at 8 56 36 AM

So, the suggestion in the second "6" listed, that we make the Ш narrower, e.g. is not desirable, as grades are intended to restore the appearance or a style that would otherwise appear heavier or lighter than intended. Grade is not intended to present another style.

vikagrabowska commented 2 years ago

Hi David,

SO, contour of 042C, "Soft" appears to be reversed.

Thank you for checking! That would explain the issue in Ы.

So, the suggestion in the second "6" listed, that we make the Ш narrower, e.g. is not desirable, as grades are intended to restore the appearance or a style that would otherwise appear heavier or lighter than intended. Grade is not intended to present another style.

So changing the width of letterform would only make sense if made across whole range of grad, do I understand correctly? My intention was to suggest narrowing ШЩ in whole range of grad. I'm sorry if my description GRAD: -50 - +50 was confusing, I meant from -50 to 50. We've been running tests in range -50; 0; 50 grad, but I realise that actual range of grad is larger.

Also this header had wght and wdth values swapped, apologies.

wght: 100 | wdth: 400 | slnt: 0 | GRAD: -50 - +50 | opsz: 144

It should be like below, I will edit it. wght: 400 | wdth: 100 | slnt: 0 | GRAD: -50 - +50 | opsz: 144

Grad -50: Ч - slightly too wide

I realise that wouldn't make sense to propose in a single grad -50. I would just suggest slightly narrower proportions for Ч across whole range of grad.

dberlow commented 2 years ago

Also, the grade range is broader, but only -50 to 50 are used for typographic purposes, the rest is for animation per Google request.

On Feb 6, 2022, at 2:50 PM, vikagrabowska @.***> wrote:

 Hi David,

SO, contour of 042C, "Soft" appears to be reversed.

Thank you for checking! That would explain the issue in Ы.

So, the suggestion in the second "6" listed, that we make the Ш narrower, e.g. is not desirable, as grades are intended to restore the appearance or a style that would otherwise appear heavier or lighter than intended. Grade is not intended to present another style.

So changing the width of letterform would only make sense if made across whole range of grad, do I understand correctly? My intention was to suggest narrowing ШЩ in whole range of grad. I'm sorry if my description GRAD: -50 - +50 was confusing, I meant from -50 to 50. We've been running tests in range -50; 0; 50 grad, but I realise that actual range of grad is larger.

Also this header had wght and wdth values swapped, apologies.

wght: 100 | wdth: 400 | slnt: 0 | GRAD: -50 - +50 | opsz: 144

It should be like below, I will edit it. wght: 400 | wdth: 100 | slnt: 0 | GRAD: -50 - +50 | opsz: 144

Grad -50: Ч - slightly too wide

I realise that wouldn't make sense to propose in a single grad -50. I would just suggest slightly narrower proportions for Ч across whole range of grad.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.

vikagrabowska commented 2 years ago

Also, the grade range is broader, but only -50 to 50 are used for typographic purposes, the rest is for animation per Google request.

Thanks for clarification!