January 2, 2008 at 01:32Font smoothing, and subpixel rendering
An excellent article by Joel Spolsky on the differences in rendering text on computers running Mac OS X and Windows, as well as on the rejection by Windows users of how Safari renders texts on the screen. It is written in simple language, accessible and without loading details.
Original
Algortm
Apple and Microsoft have always had disagreements about how to display fonts on a computer screen. Today, both companies use “sub-pixel” rendering to achieve sharper-looking fonts on traditionally low-resolution screens. Where they differ is in philosophy.
Apple believes that the purpose of the algorithm is to keep the font design as much as possible, even due to a slight blur.
Microsoft believes that the shape of each letter should be within the boundaries set by pixels to prevent blurring and improve readability, even due to incomplete matching of the style.
Now that Safari for Windows has come out, and I had to try hard to use Apple's text rendering algorithms on Windows, you can compare the differences in company philosophy on your monitor, and you’ll understand what I mean. I think you will notice the difference. Apple fonts are actually a little blurry, with blurry edges, but with a small font size, there are noticeably more variations between different font families, because their rendering is closer to how the font would look printed in high resolution.
(Attention: To correctly see the illustration, you need an LCD monitor with the order of pixels R, G, B, like mine. Otherwise, everything will look different and wrong.)
Translation of text from top to bottom:
• Poppy
- The difference between bold and less bold text corresponding to the print
- Beautifully curled tail of the letter “g”
- Blurred “i”
- ... and “e”
• Windows
- low-fat text is lighter than print to fit in the pixel grid
- the tail of “g” is aligned, to fit in the pixel grid
- “i” is clear
- the jumper in “e” is moved higher to fit in the pixel grid
The differences are due to Apple's legacy in desktop printing and graphic design. The nice thing about the Apple algorithm is that you create a page of text to print, and on the screen you have a version very close to the final product. This is especially useful if it matters to you how dark a block of text looks. Microsoft's mechanism of “dropping” fonts into pixels means that they really don't mind using thinner lines to avoid blurry edges, even when it makes an entire paragraph easier than it would look when printed.
The advantage of the Microsoft method is that it works better for screen reading. Microsoft pragmatically decided that the design of the style is not so inviolable, and that clear text on the screen that is convenient to read is more important than the font designer’s idea of how light or dark a whole block of text should feel. Microsoft really designed screen fonts, such as Georgia and Verdana, based on pixel boundaries. They are very beautiful on the screen, but not impressive when printed.
Traditionally, Apple chooses a stylish way, putting art above practicality, because Steve Jobs has a taste, and Microsoft chooses a convenient way, a measurable pragmatic way that has not a gram of style. It could be said differently, if Apple were Target, then Microsoft would be Wal-Mart. (Note lane. - Target and Wal-Mart are supermarket chains in the USA. Target is beautiful, stylish stores, while Wal-Mart is a mass market, inexpensive goods, and the corresponding design of stores)
Now, regarding the issue of preferences
. Apple users prefer the Apple method, while Windows users prefer the method developed by Microsoft. This is not ordinary fanaticism at all; it reflects the fact that when you ask people to choose a style or design that they like, if they are not prepared, they usually choose the one that looks most familiar. In matters of taste, when you do a survey of preferences, you can find that most people do not know what to choose, and therefore choose what seems most familiar. This applies to everything from cutlery (people choose a pattern that looks like a cutlery from their childhood) to fonts and graphic design: if people are not trained to know what to look for, they will choose what they want familiar.
That's why Apple engineers probably think that they are doing a great service to the Windows community using their “superior” font rendering technology to pagans, and this explains why Windows users will most likely think that Safari's font rendering is blurry and weird, and they don’t know why this is so, but they just don't like it. They rather think ... “Oppa ... It's different. I do not like when it is different. Why don't I like these fonts? Yeah, when I look closer, they look blurry. That's why."
Translation: alexmak.net .
PS
Sorry, maybe you saw this topic, which I posted, but I still did not fully understand the principle of the Habr engine. :(
Original
Algortm
Apple and Microsoft have always had disagreements about how to display fonts on a computer screen. Today, both companies use “sub-pixel” rendering to achieve sharper-looking fonts on traditionally low-resolution screens. Where they differ is in philosophy.
Apple believes that the purpose of the algorithm is to keep the font design as much as possible, even due to a slight blur.
Microsoft believes that the shape of each letter should be within the boundaries set by pixels to prevent blurring and improve readability, even due to incomplete matching of the style.
Now that Safari for Windows has come out, and I had to try hard to use Apple's text rendering algorithms on Windows, you can compare the differences in company philosophy on your monitor, and you’ll understand what I mean. I think you will notice the difference. Apple fonts are actually a little blurry, with blurry edges, but with a small font size, there are noticeably more variations between different font families, because their rendering is closer to how the font would look printed in high resolution.
(Attention: To correctly see the illustration, you need an LCD monitor with the order of pixels R, G, B, like mine. Otherwise, everything will look different and wrong.)
Translation of text from top to bottom:
• Poppy
- The difference between bold and less bold text corresponding to the print
- Beautifully curled tail of the letter “g”
- Blurred “i”
- ... and “e”
• Windows
- low-fat text is lighter than print to fit in the pixel grid
- the tail of “g” is aligned, to fit in the pixel grid
- “i” is clear
- the jumper in “e” is moved higher to fit in the pixel grid
The differences are due to Apple's legacy in desktop printing and graphic design. The nice thing about the Apple algorithm is that you create a page of text to print, and on the screen you have a version very close to the final product. This is especially useful if it matters to you how dark a block of text looks. Microsoft's mechanism of “dropping” fonts into pixels means that they really don't mind using thinner lines to avoid blurry edges, even when it makes an entire paragraph easier than it would look when printed.
The advantage of the Microsoft method is that it works better for screen reading. Microsoft pragmatically decided that the design of the style is not so inviolable, and that clear text on the screen that is convenient to read is more important than the font designer’s idea of how light or dark a whole block of text should feel. Microsoft really designed screen fonts, such as Georgia and Verdana, based on pixel boundaries. They are very beautiful on the screen, but not impressive when printed.
Traditionally, Apple chooses a stylish way, putting art above practicality, because Steve Jobs has a taste, and Microsoft chooses a convenient way, a measurable pragmatic way that has not a gram of style. It could be said differently, if Apple were Target, then Microsoft would be Wal-Mart. (Note lane. - Target and Wal-Mart are supermarket chains in the USA. Target is beautiful, stylish stores, while Wal-Mart is a mass market, inexpensive goods, and the corresponding design of stores)
Now, regarding the issue of preferences
. Apple users prefer the Apple method, while Windows users prefer the method developed by Microsoft. This is not ordinary fanaticism at all; it reflects the fact that when you ask people to choose a style or design that they like, if they are not prepared, they usually choose the one that looks most familiar. In matters of taste, when you do a survey of preferences, you can find that most people do not know what to choose, and therefore choose what seems most familiar. This applies to everything from cutlery (people choose a pattern that looks like a cutlery from their childhood) to fonts and graphic design: if people are not trained to know what to look for, they will choose what they want familiar.
That's why Apple engineers probably think that they are doing a great service to the Windows community using their “superior” font rendering technology to pagans, and this explains why Windows users will most likely think that Safari's font rendering is blurry and weird, and they don’t know why this is so, but they just don't like it. They rather think ... “Oppa ... It's different. I do not like when it is different. Why don't I like these fonts? Yeah, when I look closer, they look blurry. That's why."
Translation: alexmak.net .
PS
Sorry, maybe you saw this topic, which I posted, but I still did not fully understand the principle of the Habr engine. :(