Type 1, TrueType, OpenType
Tom Arah inspects the Type 1, TrueType and OpenType standards as Adobe, Apple and Microsoft battle it out for font supremacy. .
All publishing and design is concerned with the transmission of information and the vast majority of that information takes the form of text. Bizarrely however, we pay almost no attention to the system by which this information is transferred: the humble font. In a way this seems quite natural. What we are concerned with when reading is meaning. As such, the fundamental units for us are the letters which can be joined together in any number of arrangements to make the real unit of meaning - words. It's a fantastically efficient system built, in the case of Roman languages, on just the twenty-six letters of the alphabet.
For the computer it is all very different. The computer has no sense of meaning and so no interest in words or letters. As its name implies, the only unit that the computer can deal with is numbers. Of course we also need the computer to handle text for us, so the solution is to encode each letter as a number, turning the alphabet into a character set. Storing each character as a single byte, as the standardised ASCII (american standard code for information interchange) does, gives 256 character codes to play with that any application on any platform immediately knows how to interpret and manipulate.
For this data stored on the computer to be given back its meaning and for the all-important transmission of information to take place, the encoding system has to be reversed. In other words the character codes have to be translated back into semantic letters. And much the most common and efficient way that this is achieved is visually - the printed or onscreen text is read. This is such an everyday activity that we take it for granted and hardly give it a thought - we think that we are directly reading the meaning. In fact we are recognising character shapes on the page or screen, called "glyphs". In other words the computer translates numbers into graphics and it's actually these graphics that we translate into letters and then into words and meaning.
Whereas we think of letters and computers handle character codes what we are actually looking at are graphical glyphs.
Most of us might not give much thought to the essentially graphical nature of writing and reading, but since Gutenberg's invention of re-usable metal type, the art of typography has quietly evolved into a fully developed system. To begin with the core 26 letters of the Roman alphabet have been expanded to include their upper case equivalents, the various punctuation marks and italicized and emboldened versions, all designed to aid the interpretation of meaning and the efficiency of reading as a means of communication.
More importantly, over these 500 years, different graphical interpretations of each of these sets of glyphs have been painstakingly designed. The result is that today there are literally thousands of different typefaces on offer from the earliest old face serifs, such as Goudy, through to the self-consciously modern sans, such as Futura. Looking at the individual glyphs it's possible to break them down to see how each distinctive typeface is built - sloping counters, large x-height, compressed descenders and so on - but what really matters is the effect that each face produces in action. The most useful analogy is to music. At one level it's the notes, the letters, that are important but the same notes can be played on any number of instruments and the typographer's skill comes in choosing the right instruments and mix of instruments to suit the piece.
Nowadays we take this vast choice of typefaces for granted but just a few decades ago the situation was completely different - and little different from the Middle Ages! Until very recently the only tool at the typographer's disposal was essentially the same reusable type that Gutenberg had invented. It's worth thinking about just what that meant in practice: every single glyph had to be individually cast in metal. In fact each glyph set at every necessary point size had to be cast, and often subtly redesigned, to produce each individual font (the common blurring between the terms "face" and "font" is a modern luxury).
What broke the mould - literally - was the invention of the computer and its use for text handling. For the end user to see anything it had to be possible to present glyphs on screen. The earliest screen fonts weren't anything to write home about but crucially they were based on the principle of creating the glyphs by switching pixels on and off on a grid. From here it was a relatively simple step to translate the idea of bitmapped fonts to paper output on the appropriately named dot matrix printer.
The use of in-built or downloaded bitmapped fonts was a huge step forward, but even when translated to the 300 dpi laser printer, the output quality couldn't compare to crisp metal type and the sheer amount of data meant it simply wasn't practical to use bitmapped fonts with true high resolution devices such as imagesetters. It looked like an impasse, but a truly revolutionary solution was around the corner.
The Type 1 breakthrough was that fonts were described as scalable outlines.
Back in the early-1980's Adobe was a small company founded to develop and promote the use of PostScript. PostScript's great strength is that it is a page description language (PDL) so that, rather than storing objects as fixed resolution bitmaps, each element of the page layout is stored programmatically - to describe a square for example you only need to define the four corner co-ordinates. Run the PostScript program for the page as a whole and it will be reliably recreated at the resolution of whatever PostScript device you output it on.
All that was needed was a way of handling fonts, but in fact that was simplicity itself. Each glyph in the typeface can simply be defined as an outline, as a programmatic mathematical description, using PostScript's own system of cubic Bézier curves. This so-called Type 1 outline is then mapped to the grid of the output device and turned into dots or "rasterized". Essentially if the centre of each grid square falls within or on the glyph outline it is switched on, otherwise it is switched off. Crucially this means that the outline can be scaled to any size before rasterizing. In other words each Type 1 face can produce type at any size and the distinction between "face" and "font" disappears.
The outline is rasterized according to the output grid, the higher the resolution the higher the quality.
Best of all, because of its vector nature, each Type 1 face is resolution-independent. That means that exactly the same font can be output on a PostScript-based imagesetter as on a local PostScript-based laser printer, such as Apple's breakthrough 1985 LaserWriter, the only difference is the resolution. This access to high quality final output and reliable roundtripping made the computer the ideal design platform, kick-started the DTP revolution and rendered Gutenberg's metal type a distant memory.
But something was missing. Type 1 and PostScript took care of print, but what about the screen? With the shift from MS-DOS to Windows, graphical user interfaces were now standard but screen type was still handled as awkward individual bitmapped fonts. Clearly the same scalable outline system used for print needed to be translated to the screen and the obvious candidate was Adobe's Type 1 format - especially as it had developed Display PostScript for exactly this purpose.
Unfortunately things are never that simple. Back in the mid-1980's Adobe was not the open-standards company it is today and was milking PostScript and Type 1 licences for every penny that it could (I know; I spent hundreds of pounds on Type 1 fonts that are given away today!). In the circumstances neither Apple nor Microsoft was willing to write the blank cheque that Adobe was demanding to incorporate Type 1 technology into their OS's and so they formed an alliance to fight back with their own solution.
The end result was the development of the TrueType font format which, by one of those quirks of fate, is always credited to Microsoft. In fact Microsoft's share of the deal was to develop TrueImage, a PostScript killer, and it was Apple that developed the font technology. Apple was first to incorporate the system in early 1991 with Microsoft following in 1992. Again it's now completely taken for granted, but it was this incorporation of scalable TrueType technology that made Windows 3.1 such a breakthrough release. Effectively Apple gift-wrapped Microsoft its dominance of computing in general - and no, Microsoft never did come up with TrueImage.
Perhaps because of the mistaken association with Microsoft, the TrueType format was viewed with considerable suspicion. I remember scaling up a 10-point body copy font to A4 size in Corel Draw and being seriously unimpressed with its crude quality. Even worse, if you look at a set of TrueType body copy point sizes onscreen you'll see a sudden jump in boldness, say between 12 and 13 point, and even a change in the actual letter shapes! While it was nice to have scalable onscreen fonts there was no way that I was going to trust TrueType for high quality print.
In fact that was a complete mistake. TrueType's quadratic B-splines are simpler than Type 1's cubic Bézier curves - a circle takes eight rather than twelve points to define for example - but they are just as capable of defining any glyph and are therefore more efficient. TrueType has other advantages. Its character set support is greater and more extensible for other writing systems and easier to handle as a simple look-up table. Each face also includes its own font metrics - the information on glyph widths, kerning pairs and so on - so that managing fonts is much simpler.
Ironically what really sets TrueType apart from Type 1 is its quality. This isn't down to the core maths but rather to the system of hinting. Hinting is the way in which the outline of any given glyph can be massaged to give the best rasterised results. This is particularly important for smaller point sizes for print and even more so for screen use where the number of pixels is barely sufficient to hold the character shape let alone the finer typographic nuances. In these circumstances it's essential to prevent font features such as serifs and stems from breaking up or varying in width just because of the way the outline happens to fall over the grid. With hinting you can intelligently fine-tune the font to the output grid.
Hinting itself wasn't new as the Type 1 font format provided it - in fact it was really hinting that Adobe was licensing as the non-hinted Type 3 format was always made available as an open standard. However the Type 1 hinting is only aimed at print and only allows a core set of features, such as horizontal and vertical stems, to be marked up and a pixel threshold set for each at which point the PostScript interpreter kicks in to process the hints to produce the best results it can.
By comparison the TrueType hinting is aimed at both print and screen and is completely extensible so that any and all features can be marked up and controlled at any size. The format even allows each glyph's defining control points to be repositioned at any given size so that the font can change shape depending on its output size. This dynamic adaptability is built into the font itself rather than the interpreter and effectively enables the developer to hand-craft each size of font much as they did in Gutenberg's day.
The changing weight and shape of the TrueType Verdana at different sizes is actually its great strength and thanks to advanced hinting.
Bearing this in mind it's no surprise that my blown-up TrueType gylph should seem comparatively crude as its simpler shape would actually have worked better at the original intended size. Likewise there's no way of getting around a jump in boldness in any screen font when you move from strokes being represented as one pixel wide to two pixels wide, and that's far preferable to randomly varying stroke widths. And finally the radical change in the shape of onscreen glyphs is actually what makes optimized TrueType fonts such as Verdana so readable and so well suited for onscreen use. In other words what I originally took to be signs of TrueType's inferior quality were actually the signs of its great strength.
So if TrueType was the superior technology both for print and now onscreen, why didn't it take over from Type 1? Misconceptions like mine certainly didn't help and there were other practical teething problems such as Windows 3.1's 16-bit limitations and slight differences between the Microsoft and Apple implementations. However, the real reason was that while TrueType promised higher quality it didn't guarantee it and so the mass-market appeal of Windows soon led to a flood of cheap and even free fonts with absolutely no hinting and so terrible quality. Not surprisingly the serious type foundries weren't going to cut their own throats by trying to compete at this level and refused to convert their fonts to TrueType.
Bizarrely then the expensive, inferior, print-only Type 1 format survived by providing a niche solution for high-end designers! This inverted state of affairs couldn't last and Adobe fought back by opening up the Type 1 format and by making its Display PostScript concept available as a utility in the form of its Adobe Type Manager application. At last Type 1 fonts were affordable and could be seen and handled on-screen. The result was an uneasy truce in which designers generally preferred to work with Type 1 fonts for print design and TrueType fonts for screen and Web design - and just had to accept the occasional conflict when for example font names clashed or fonts with slightly different metrics were substituted at the printer!
Apple meanwhile was licking its wounds and, recognizing that Microsoft was never going to deliver TrueImage, began developing its own rival to the PostScript/Type 1 combination in the form of QuickDraw GX and its associated GX-Font technology. Essentially GX-Fonts would support both existing Type 1 and TrueType fonts by encapsulating them in a GX wrapper. It would also extend TrueType's character set support to make available expert sets with old style figures, ligatures and so on. What would prove truly radical was the integration of state-of-the-art typographic composition, including advanced features such as the alignment of type based on the glyph's actual shape rather than on fixed metrics. Most importantly, all this power would be available to any application whether professional DTP application or humble word processor.
And that was the format's fatal flaw. For Adobe to support GX would be cutting its own throat. And without Adobe's support and the professional design market it represented there was no future for GX so eventually Apple was forced to stop development. Adobe and Microsoft meanwhile had recognized the threat and resolved their differences and in 1996 signed a joint development agreement in which they licensed the Type 1 and TrueType font technologies to each other. The first results were native support for TrueType fonts under PostScript 3 and native support for Type 1 under Windows 2000.
More significant was the joint Adobe-Microsoft development of a totally new and advanced font format, OpenType, to run alongside Type 1 and TrueType. Actually "totally new" is rather overstating the case as is "advanced". OpenType is effectively a cross between the two font technologies, TrueType and GX, both of which were actually developed by Apple! At least they had the decency to make it an open standard.
OpenType is the third and hopefully final major font format.
Effectively the OpenType font is either a Type 1 or TrueType font in a TrueType wrapper. Otherwise the big differences are all ideas lifted from GX with the main selling points being a wider range of glyphs based on Unicode's 64,000 character set - more than enough for every character in every language - and typographically advanced features such as expert sets and automatic glyph substitution for example of ligatures. Crucially though these advanced typographical features aren't made available at the OS level so you need an advanced application to take advantage of them - Adobe InDesign springs to mind.
So which company has won the font wars? Well you have to say that Microsoft has spent very little of its own forces, benefited enormously from its alliances and has come out with much the most territory. In a way though it's easier to say who's lost. As always it's the civilians who are the major casualties and it's the end user who is now being made to pay again - quite possibly buying exactly the same typeface for the third time!
Ultimately however, if it means that the font wars are truly over for good, it's a price worth paying. After all, we might not give fonts much thought, but without them there would be no design.
Hopefully you've found the information you were looking for. For further information please click here.
For free trials and special offers please click the following recommended links:
For further information on the following design applications and subjects please click on the links below:
[3D], [3ds max], [Adobe], [Acrobat], [Cinema 4D], [Corel], [CorelDRAW], [Creative Suite], [Digital Image], [Dreamweaver], [Director], [Fireworks], [Flash], [FreeHand], [FrameMaker], [FrontPage], [GoLive], [Graphic Design], [HTML/CSS], [Illustrator], [InDesign], [Macromedia], [Macromedia Studio], [Microsoft], [NetObjects Fusion], [PageMaker], [Paint Shop Pro], [Painter], [Photo Editing], [PhotoImpact], [Photoshop], [Photoshop Elements], [Publisher], [QuarkXPress], [Web Design]
To continue your search on the designer-info.com site and beyond please use the Google and Amazon search boxes below:
|designer-info.com: independent, informed, intelligent, incisive, in-depth...|
All the work on the site (over 250 reviews, over 100 articles and tutorials) has been written by me, Tom Arah It's also me who maintains the site, answers your emails etc. The site is very popular and from your feedback I know it's a useful resource - but it takes a lot to keep it up.
You can help keep the site running, independent and free by Bookmarking the site (if you don't you might never find it again), telling others about it and by coming back (new content is added every month). Even better you can make a donation eg $5 the typical cost of just one issue of a print magazine or buy anything via Amazon.com or Amazon.co.uk (now or next time you feel like shopping) using these links or the designer-info.com shop - it's a great way of quickly finding the best buys, it costs you nothing and I gain a small but much-appreciated commission.
Thanks very much, Tom Arah
[DTP/Publishing] [Vector Drawing] [Bitmap/Photo] [Web] [3D]
[Articles/Tutorials] [Reviews/Archive] [Shop] [Home/What's New]