4 ms·
"You can see the low legibility of Helvetica if you type texts in a small size and make them blur. Some texts become blended and hard to decipher. They say that
by a_e_k 11y ago
"You can see the low legibility of Helvetica if you type texts in a small size and make them blur. Some texts become blended and hard to decipher. They say that Apple developed San Francisco fonts in order to make small size texts in Apple Watch more legible."
Funny thing is, this is exactly the problem that font hinting was meant to solve [1][2]. Apple developed the tech for TrueType font hinting nearly 25 years ago but then ditched most of it in their antialiasing rasterizer for OS X. If they still had hinting in their rasterizer, then instead of creating a whole new font with flatter segments they could have hinted Helvetica to flatten out at smaller sizes while still curving in the larger sizes. Then they would have had the best of both. Granted, that would have meant admitting that there's something to respecting the pixel grid [3], but this seems like a tacit admission of that anyway.
[1] https://developer.apple.com/fonts/TrueType-Reference-Manual/RM03/Chap3.html https://developer.apple.com/fonts/TrueType-Reference-Manual/...
[2] http://www.rastertragedy.com/ http://www.rastertragedy.com/
[3] http://blog.codinghorror.com/font-rendering-respecting-the-pixel-grid/ http://blog.codinghorror.com/font-rendering-respecting-the-p...
- theoh 11y agoThat wouldn't be Helvetica, so they would probably give it a new name. Given the ambiguous ubiquity of Helvetica, it's possible they'd have chosen a name like San Francisco anyway. Whether the feature is provided by hinting or a pair of fonts is moot.
- pavlov 11y agoHinted fonts did look great on low-DPI screens. To my taste, Microsoft's Tahoma, the Win95 / NT4 system font designed by Matthew Carter, is probably the best font that you can squeeze into a grid about 9 pixels high: https://www.microsoft.com/typography/fonts/family.aspx?FID=19 https://www.microsoft.com/typography/fonts/family.aspx?FID=1... However, nicely hinted fonts were designed from scratch for that purpose. You can see it in Tahoma's letter shapes at larger sizes, the construction is very grid-like overall. A well-hinted Helvetica really wouldn't look any different from Arial back in 1995. The low resolution makes Helvetica lose its character. I think Apple made a good choice in designing a more contemporary font.
- polpo 11y agoI'm not sure how font hinting would solve this problem. The resolution of the Apple Watch is 326ppi, so the pixel grid doesn't matter so much. Seems like the different San Francisco variants were designed to solve problems due to the physical dimensions of the letterforms at small sizes, not anything to do with a pixel grid.
- a_e_k 11y agoThe DPI may be high, yes, but the number of actual pixels is still quite low. Looking it up, I see that the smaller watch has a 272x340 display and the larger has a 312x390. That's either 92k or 122k pixels to work with -- about 1/3 VGA -- over the entire display. At those scales, hinting is absolutely relevant unless perhaps you're only displaying a few relatively large words.
- jameshart 11y agoIt's less than an inch across, and viewed typically at arm's length. Yes, you're only displaying a few relatively large words. They're not trying to display 80 column text on a watchface.
- etangent 11y agoAs others pointed out, font hinting is solving a completely different problem from the one that Apple faces or that you describe. You describe the problem of displaying small text on a low-resolution grid. Font hinting is indeed a passable solution to that problem. However Apple's problem is displaying small text, period (Retina displays have plenty of resolution). The first sentence of the top link you give, "A quality outline font is one that provides legibility at small sizes on low resolution devices" tells us why this is not relevant. Nobody cares about low-resolution devices anymore. Apple set out to make things legible assuming infinite resolution.
- esrauch 11y agoMaybe we should have a hinting equivalent for physical size then (rather than pixel hinting or an entirely different font).
- jacobolus 11y agoAdobe used to have this feature in their typeface format and type design tools in the 1990s. The format was called “Multiple Master”, see https://en.wikipedia.org/wiki/Multiple_master_fonts https://en.wikipedia.org/wiki/Multiple_master_fonts and http://blog.typekit.com/2014/07/30/the-adobe-originals-silver-anniversary-story-how-the-originals-endured-in-an-ever-changing-industry/ http://blog.typekit.com/2014/07/30/the-adobe-originals-silve.... (Metafont also has this feature.) In practice, type designers found that making such fonts took dramatically more work, and typographers/publishers weren’t willing to pay commensurately more money. So it didn’t see much adoption outside Adobe. As a result, the format died out, and we stuck with descendants of the original PostScript font format. More recently, many companies (but especially Adobe) ship “pro” typefaces with separate fonts for various “optical sizes”. Again, producing these takes a lot of extra work, so most type designers don’t bother with it, and because it is designed to work with standard tools (which don’t have special knowledge of the feature), typographers/designers must explicitly choose which optical size they want to use in each context. So in theory it’s inferior to the Multiple Master format, but in practice it is backwards compatible, which ends up being more important. Apple can solve this problem for their UI fonts specifically, because they have full control. Unfortunately, they haven’t offered these features for use with other typefaces.
- derefr 11y agoI'm not sure OSX can respect the pixel grid, at least any more. People think "Retina" means "one virtual pixel is 2x2 physical pixels", and on iOS it does mean that; on OSX, though, it means "we render to a back-buffer that's at least 1.5x as large, using our unscaled 2x assets, and then downsample[1] to the physical display size." That means that the physical pixel grid is truly a black-box abstraction from the text renderer's perspective. Applications have no idea how many physical pixels there are (unless they draw to an OpenGL buffer instead of going through Quartz.) [1] This settings panel (http://i.imgur.com/TRtR17g.png http://i.imgur.com/TRtR17g.png), despite looking like a DPI adjuster, is actually to choose what size backbuffer gets rendered-onto-and-downsampled. Note the shitty blurry edges of the text in that just-taken screenshot (no, it's not a JPEG): the text is obviously sitting on non-integral pixel positions after downsampling. It actually looks fine, when it's shrunk down onto your 15" screen and you're sitting two feet from it; it just doesn't look terribly clean at examinable 1:1 scale.
- amake 11y agoIt doesn't even mean that on iOS since the advent of the iPhone 6 Plus.