I just wanted to do a quick survey to see how many people are using higher UPM than the standard 1000. Is anyone here regularly creating fonts with 1000+ UPM, or are people generally sticking to 1000 to avoid any kind of issues? I understand it’s no longer a problem to step outside the 1000UPM standard, but I’m unsure how well graphics programs support non-1000UPM fonts. Thanks.
I enlarge my UPM if the design needs it. A connecting Spencerian script benefits greatly from 2000. I should also note that I don’t do floating point fonts; everything to the grid.
2 Likes
Most of my work is 2048. I find I need the resolution to get the curves that I like. I do agree with, George, use what the font requires.
I know some that decrease the UPM. A coarser grid makes it easer to get thing consistent.
Higher UPM also gives smoother variable font interpolation, especially if it’s sensitive to details (yep, it does snip to the grid along the way)
1 Like
Are you sure with the grid snapping?
It’s really visible in low UPM. But not sure about technical aspect of it, maybe it’s tied to something else like rounding?
Thanks for the replies. I think I will do some testing to see how well it works with interpolation etc. It would be fairly simple for me to upsample some existing projects from 1000 upm to 2000 upm so that may be an option.
I almost never use 1000 UPM. Although it is the standard for CFF fonts, I consider it inappropriate for design work. Either, it’s too fine to capture the type design, and you end up fiddling with irrelevant 1-unit-nudges. Or, it is too coarse to be considered practically stepless.
I usually use UPM 250 (JAF Bernini Sans is completely designed at 250 UPM), which helps me focus on the design without getting lost in imperceptible modifications. Believe me, it is enough for your type design as well.
Or, I use e.g. 3000 UPM, then it feels practically stepless, and I don’t have to worry about rounding errors that much (being a self-confessed rounding-error-phobe).
2 Likes