The Licensing and Price of Web FontFonts

RahimSnow's picture

I can't tell you how excited I am that WOFF is a real standard, supported by a real browser (Firefox 3.6), and pursued aggressively by a talented and prolific company like FontFont that just released more than 30 of their most successful families.

I am even more excited that my favorite FontFont family Milo and Milo Serif is among them. I have licensed many weights of Milo and the entire family of Milo Serif. And I have been wanting to use them on the web without using any hacks (Cufon, SIFR) or have them hosted somewhere (Typekit). I want the WOFF version so I can host it myself. So I have been very excited to learn that the Milo family was released in WOFF format.

And yet my excitement came to a sudden halt due to 2 things:

Licensing

Reading About Web FontFonts, I find this statement:

While standard desktop fonts are licensed by the number of users or workstations using the fonts, Web FontFonts are licensed by the average pageviews per month of all the domains within the licensing organization. There are three simple license levels: up to 500,000; up to 5 million; and up to 50 million pageviews per month.

And what exactly is a pageview? In the blog comments, Ivan Bettger writes:

A pageview in this case is defined as a request to load a single page on any of your sites that use the licensed Web FontFonts. A refresh or a clickthrough would count as a pageview.

Depending on the architecture of your site, that could add up very quickly. I thought pageview was a metric that might be necessary if FontFont was hosting these fonts and needed to charge more to those who are requesting the fonts more (e.g. putting a strain on their servers and bandwidth).

If I am licensing the WOFF fonts and have them stored on my own server, how does pageview even come into play? If I was licensing the OTF versions for traditional publishing purposes, this would be like charging me less if my newspaper has a small subscriber count (a local zine) and charging me more if I'm the New York Times.

Pricing

As I said, I have licensed the entire FF Milo Serif family. It wasn't cheap, but I am happy to pay for the quality and artistry of Mike Abbink's work.

Now I simply want to use the web versions of this same family but there is no explicit path for me to extend my existing license onto the web. If there is a path, please let me know.

Due to the pageview-based licensing, the price is unexpectedly high.

Keep in mind: each of the 12 fonts in the Milo Serif OTF family are sold for $65 for a total of $780.

Now if I want to license FF Milo Serif Web (WOFF) for

personal (500,000 pageviews/month), the cost is $583

business (5,000,000 pageviews/month), the cost is $2332.00

professional (50,000,000 pageviews/month), the cost is $8745.00

So for me, a previous license holder of the OTF version, that would be $583 on top of $780. Wow. Yes, I know a lot of work went into making the web versions. I know you have your own overhead. I'm not asking for it to be free. But $583? 3/4 the price of the OTF?

Keep in mind that these are all WOFF fonts that will sit on my server. Neither FontFont nor Typekit have to serve them for me. I don't take up their bandwidth or server space or anything.

So with my OTF family, I can print 50,000,000 issues of a newspaper, and the licensing will cost me just what it does if I use it on my one computer. But with the WOFF family, I have to pay depending on how popular my site is. Now, before you go and tell me that 50,000,000 pageviews would take X number of years to achieve and what not, I just want you to think about the principle of it.

Another way to think about is this: suppose every time you fire up Photoshop and use your favorite FF font, there is a counter that counts how many times you accessed that font. And you have to agree that you will stay inside a specified number of times that you are allowed to access that font before they hit you with overage charges. (just like the minutes on your cellphone)

OTF licensing is usually per computer in a single organization. WOFF licensing is per pageview. If we changed OTF licensing to reflect how many readers you have, would that make any sense? Why does it suddenly make sense for the web? I am not repeatedly accessing a service on your side. I am accessing a digital asset that I have licensed from you and it will live on my side.

Bottom line: the current Web FontFont license is metering my use of their fonts. This is a showstopper for me. As much as I wanted to use FF Milo Serif Web on my websites, I can't do it due to the reasons given above.

I would like to request that FontFont withdraw this pageview-based licensing scheme and give those of us who have already invested in the OTF version a generous and reasonable license extension. We are your customers. We enjoy your work. We want to continue working with you. Please work with us on this.

Thanks for listening.

Rahim

Chris Dean's picture

What happens if someone sets up a script that hits your site millions of times bumping you over the "personal" into the "business" level?

eriks's picture

Aaron:
the hardback book is the .otf font and the paperback the .woff. You pay separately for those two types of book which have the same content, as you will have to do for a print font and a webfont. If you design a brochure for a client, you need the font and you can send a pdf to the printers. As soon as the client or the printer makes any changes to the document, they also need a license. And if a client has chosen a typeface for their communication, it is very likely that they will want to use it in-house, for Powerpoint, for Word documents, even for design work. Over the life of a typeface, a license fee of a few hundred $, £ or € is negligible. One image from a professional photographer will cost more and can only be used once.

I also buy fonts because I like them but have no instant use for them. Like I buy books that I’ll read later (or not), have more than one bicycle although I can only ride one at the same time. (Disclaimer: we have access to fonts on our server by special arrangement with several foundries, but as soon as we use one of them for client work, we buy a license.) If type is a tool that you use, you need to buy it. If you build websites for less than 10k of whatever currency, then you cannot be expected to invest and will have to do with system fonts. But then those brands cannot expect to look different from others. An expensive way to avoid paying for adequate typography.

John Hudson's picture

Aaron: Just so we're on the same page, can you explain for me your view/understanding of the 'different piece of font software' that the designer would be paying extra for?

In addition to what others have written:

The level of technical differentiation of web fonts from desktop fonts will vary. Obviously if a desktop font is in TrueType format and has been designed and produced with screen readability in mind, then it is going to be more directly converted to an .EOT or .WOFF web-servable font. If a font was originally licensed as a PostScript font, it will need to have been converted to TrueType format, both in order to be made to work as a .EOT -- due to a bug in IE's web font support -- and in order to improve screen rendering on Windows in both .EOT and .WOFF.

I've not seen the FontShop web fonts, so can't comment on the degree of work that would have been required to make them as distinct from the desktop versions of the same typefaces. But in terms of constituting a separate product under a separate license, it doesn't really matter.

jacobsievers's picture

What happens if someone sets up a script that hits your site millions of times...?

I doubt any lawyers would interpret average in the statistical sense. Perhaps mode is the more accurate term.

Stephen Coles's picture

> Just saw Typotheque's pricing for web fonts:

Keep in mind that Typotheque's product is hosted on their network, so it could be considered more of a service than FontShop's product which delivers font files that you host on your own. This may or may not be appropriate for your needs, just as the Typekit option may or may not be more appropriate.

Richard Fink's picture

@johnhudson
>it will need to have been converted to TrueType format, both in order to be made to
>work as a .EOT -- due to a bug in IE's web font support

Unsure of what you mean by "due to a bug in IE's web font support".
I'm not aware of any bugs in IE's @font-face support. Only deliberately chosen features.
Could you explain?

Aaron Moodie's picture

Erik:

I'm not sure if the book analogy really works for me. As you pointed out before, a font has no wear and tear. Its digital not physical. But I understand what you're getting at.

What concerns me is how type is perceived to be 'used' on a website. It's true that the physical type file needs to be present in order to use @font-face, but if the site is hosted on my server (or even the .woff for that matter), is the client able to produce another site with that typeface? No more than then they are able to scan in the printed typeface on their flyer to use in a poster.

If you transfered this thinking to the print world, then I would be buying a new license for a typeface every time I used it on a different flyer, business card or poster, and then I would have to pay additional if greater than x amount of people were going to see that product.

Should clients purchase a license to use a typeface in Powerpoint, Word or other documents? Yes, as they can produce material with that typeface.

From my perspective (and in my situation), clients and public would have no more access to my .woff files than they do any other typefaces on my machine. If a font file is duplicated, by all means, buy a new license. But If I'm just using the one file as I would for print, then I don't see how these two situations are different.

Aaron Moodie's picture

Erik said: If you build websites for less than 10k of whatever currency, then you cannot be expected to invest and will have to do with system fonts.

Is this how you believe the printed world should work as well? Even if i've invested in a typeface myself, if the client can't afford to cover an additional license for their 1000 print run program and 300 print run poster, then I will have to stick with system fonts?

Richard Fink's picture

@sii
The other reason for additional fees is to cover the engineering investment needed for Web-ready fonts - primarily hinting, but also additional engineering and testing work.

sii, I didn't know you wrote fiction!
The plot needs some work. Too clichéd.
I have seen claims of fonts "optimized" for screen but I have yet to see a single side-by-side before-and-after comparison. Are you aware of any? I'm not. When I see it with my own eyes I'll believe it.
This sounds like a recycling of the usual BS you hear from all kinds of companies in all kinds of industries about how so much of the purchase price is going to be re-channeled into great products for the future and wait, you'll see. Just wait.

I've heard and sensed nothing but reluctance from font designers about this. And this skepticism has a foundation.
Here is what one fairly high-profile font designer had to say on the issue:

...TTF versions of fonts is complicated, labor intensive and expensive. It's that there are very few people who are knowledgeable about the antiquated craft of TrueType hinting. It's that the demand for hand-hinted TTF outlines is low despite the current "Web Fonts Now!" meme. It's that there is no reason to rush because most of the browser makers are too bored and/or antagonistic to make up their minds on a web font format.
...this is one of the main reasons why I think WOFF is a good format. It's a clean start. Microsoft will switch it's type rendering in the next IE in Windows 7 to DirectWrite. Firefox is working on the same thing. DirectWrite renders CFF fonts much, much better than GDI. Foundries can pick which format (TTF or CFF) best suits each font by weighing lots of variables—file size, intended point size range, browser targets, financial feasibility of hand hinting and so on.

Nobody is obliged to put in the work necessary to provide fonts suitable for use in browsers. I don't care if they do or they don't. The demand will be filled.
But publicly declining to do so is a bit of a public relations problem because nobody wants to appear as a naysayer or an obstructionist, or as out of step with the times.

Engage with the web or not, just don't piss on my leg and tell me it's raining, is all I ask.
Or feed me pablum.

Si_Daniels's picture

>sii, I didn't know you wrote fiction!

I write checks (often quite big ones) for font hinting, engineering and testing services on a regular basis. Fiction, not so much.

kari.patila's picture

What happens if someone sets up a script that hits your site millions of times...?

Assuming you don't run your own datacenter, your host will eventually stop and report the incident. I can't imagine being a victim of a DDoS attac, for instance, would bump you up to the next bracket.

Frode Bo Helland's picture

*Tracking*

Ray Larabie's picture

I like this pricing model because it makes other pricing models seem more appealing.

Richard Fink's picture

@sii
I am very appreciative of MSFT's commitment to screen readability as a part of its corporate mission - no firm has done more. But you seemed to be speaking on behalf of or in defense of others.
Anyway, touché.

@typodermic
You made me laugh.

DrDoc's picture

Tracking this thread because all the cool kids are doing it.

Frode Bo Helland's picture

I too would love to see some real life comparisons of what the hinting involves. Typoteque’s fonts are very variable in preformance in different systems and browsers.

Ivo's picture

I really like John’s comments on this discussion. I second them, especially his first one.

What we (FSI) already have learned from web designers is that most of them won’t buy licenses for a complete family like FF Milo anyway, since they are way familiar to work with just the standard weights Regular, Italic, Bold, and Bold Italic. That’s why we offer Basic Sets consisting of these weights (if they are available) and single weights only to allow for the greatest possible flexibility. Of course you can also license the entire family, but we don’t expect many Web FontFont customers who need complete super families. FF Milo Web Basic Set is $142 only, which should be okay for a website with less than 500k pv/m for an unlimited time (You can even use it on all of your websites). It’s not by chance that Web FontFonts are less expensive than their OpenType companions. Actually we discussed our prices also with a lot of web designers and already licensed some in the last months. Believe it or not, all of them considered our pricing scheme fair and comprehensible. Far from it! Some expected way more expensive fees.

Rob O. Font's picture

>I too would love to see some real life comparisons of what the hinting involves.

Can you be more precise? Are you talking about the daily life of a hinter, VTT use or...

Here is Georgia, followed by Reogia (pronounced “Re-Org Ya”), which is Georgia with the instruction portion removed from the glyph table, (pronounced “hint-stripped”).

The specimen was set on a Mac with the smoothing preference in the Appearence control panel set to “12”.

I assume this is approximately the same difference a windows xp user would see in default Appearance mode (aliased) with Georgia vs. Reogia, but I’ve been accused of tampering with specimens to make windows rendering look worse than it is, so I only show those live now. I also assume there would be a less obvious, but clear difference a windows vista/7 user would see in default Appearance mode (antialised) with the same comparison, but I have no time to provide live video of samples being made there.

On the contrary, there is no significant difference on the Mac, Georgia vs. Reogia, when antialised, i.e. above 12 or whatever value the user sets on a Mac for a smoothing preference in the Appearence control panel, it just works.

If you are having trouble getting fonts to look good across all windows preferences and versions, it's likely you have too many fonts anyway.

Cheers!

nina's picture

> You can even use it on all of your websites

Pardon me if this is slightly stupid, but I'm still confused.
Is it true that: The client (organization that runs the web site in question) licenses the fonts, and then the fonts can be used on all domains this client owns/runs; but I, as a web designer in this case, do not have to license the fonts in my name as well in order to do work on the web sites on these domains?
And if the client uses the font on multiple domains, the average page views from each domain are just added together?

Ivo's picture

Is it true that: The client (organization that runs the web site in question) licenses the fonts, and then the fonts can be used on all domains this client owns/runs; but I, as a web designer in this case, do not have to license the fonts in my name as well in order to do work on the web sites on these domains?

This is true (as it is also for ‘print fonts’ when you have to do a project with your client’s corporate font), but you as a designer may transfer your license to your client after you have finished the work on the website (provided that you delete all your copies of the fonts and the transferee accepts and agrees in writing [with copy to FSI] to be bound by the EULA).

And if the client uses the font on multiple domains, the average page views from each domain are just added together?

Yes.

Richard Fink's picture

@frode frank

>I too would love to see some real life comparisons of what the hinting involves.
>Typoteque’s fonts are very variable in preformance in different systems and browsers.

We're moving OT here, but....

That's a question you should put to Paul B, of Typotheque. (Or maybe I will.)
In Windows, at least, it seems TrueType hinting can homogenize - there are still differences, but they are less apparent, from browser to browser.
It's interesting that the variations in font rendering within browsers has only now fallen under scrutiny with @font-face. It happens with the "web safe" fonts, too, you know.
Perhaps when you are powerless and have to accept a thing you cannot change, there is a willing suspension of discernment.
What we have managed to milk out of the web safe fonts for fifteen years is unbelievable.
@font-face is one way to tackle the lack of variety. But there are technical and learning barriers to overcome. The other way would be to add to the web safe fonts.
In 2005, Joe Clark wrote a blog post urging Apple to license the ClearType fonts from Microsoft for inclusion on the Mac.
http://blog.fawny.org/2005/01/21/ccccccm/
This would have, at least, doubled the typographic palette that web designers had to work with and, considering how much mileage has been gotten out of Verdana, Georgia, et al, this would have been no small thing. It's not too late for this, either.
Here's a great survey question:
Where do the web safe fonts on the Mac come from?
1) They've been there since the beginning. (circa 1995, when the world began)
2) Steve Jobs won the rights to use them in a card game with Bill Gates.
3) Steve Jobs invented them.
4) Apple licenses them from Microsoft.
5) The web safe fonts are public domain, free for anybody to use.

Apple users would most probably pick number three (just joking, I'm buying a Mac in a couple of months) but I'm betting number 5 would get selected the most.

Lastly - when you get charged monthly by the number of page views, what happens if you fall behind in your payments? Does the font company stop by like the electric company does and give you one last chance to cough up before turning off the fonts?

"Hey man, don't turn off the fonts, please. I got kids."

Later frode,

rich

nina's picture

Got it Ivo, thanks.

Richard Fink's picture

@davidberlow
On the contrary, there is no significant difference on the Mac, Georgia vs. Reogia, when antialised, i.e. above 12 or whatever value the user sets on a Mac for a smoothing preference in the Appearence control panel, it just works.

But works how well or how poorly, in comparison to Windows?
That Windows needs the hints for the font to "work" we know.

Hmmm...?

Rich

Frode Bo Helland's picture

David: I’m not directing this question to you specifically, but what I would love to see in the marketing for web fonts is the different results of auto hinting (or no hinting in case the fonts updated for web doesn’t have any) and manual hinting. Some of my fonts actually preform reasonably well without any hinting at all (at medium sizes), by the way.

Jens Kutilek's picture

Here's a comparison of the various appearances of FF Clifford Nine Roman in its OpenType CFF and TrueType incarnations. The first line of each size has no hinting at all, the second line has autohinting by FontLab Studio, and each third line is the font as offered by FSI (FF Clifford Pro Nine for the CFF sample respectively FF Clifford Web Pro Nine for the TT-based fonts).

-
CFF fonts (no difference if ClearType is on or off):

-
TrueType fonts, first with Windows Standard font smoothing:

-
TrueType (= EOT & WOFF FontFonts)with ClearType font smoothing:

-
And another one with TrueType and ClearType:

Rob O. Font's picture

>This would have, at least, doubled the typographic palette that web designers had to work

... that's because you don't know what a web font is, or you missed the long discussion between myself and others over just exactly what the CT collection represents.

Cheers!

andreas's picture

Jens,
I can't believe these hinting enhancements are worth the hassle. For these kinds of font designs we need a mutch higher monitor ppi. :-) Hopefully it looks better on Win7 and OSX.

If I remeber right some people say a true type hinting with VVT of a std font will cost between 1000 - 2000 €. Even if these are moon prices and its done for 400 €, its a lot.

Jennie Wojtulewicz's picture

I have to consider this from the perspective of a high school digital publications adviser (that's what yearbook/newspaper/lit mag are called by the kiddies now).
I read earlier by @James Puckett:
"Royalty-free digital font licensing was a mistake made in the 1980s that has been perpetuated far too long. Creators of great photos, illustrations, writing, video, music, etc. are all paid royalties. Great type should be the same way."

This makes complete sense to me, it really does!! But I wonder how you feel about programs like yearbook and literary magazines that certainly make a profit each year using designer fonts... shouldn't we be teaching our high school students typography? How can we if we can't afford to buy the licenses? I know in my experience, our publisher provides a font book to us each year, but man are they lame.... and very over-used. I encourage my kiddos to get online and find fonts they like instead. This is not to say I have a garage sale of fonts in my yearbooks, but I do go beyond the comic-sansish fonts provided to us.

What are all your thoughts?

Si_Daniels's picture

As this thread has veered a little bit off topic I hope you don't mind me posting a link to this article that mentions FireFox switch to Direct2D font rendering... http://news.cnet.com/8301-30685_3-20000104-264.html linking to this blog post http://weblogs.mozillazine.org/asa/archives/2010/02/direct2d_landed_in_f...

>What are all your thoughts?

The model for licensing fonts for print production is unlikely to change in the foreseeable future. The fact that most font EULAs do not currently allow EOT and WOFF conversion enables this and other new models.

Frode Bo Helland's picture

On Clifford: I’m not really sure which is best at medium to large sizes, autohinting or manual hinting? The real difference is in the small sizes. How would this translate to a simpler sans serif?

To steer this back on topic: this is what we’ll be paying for, and such these kind of comparisons should be a vital part of the marketing. What separates the hinting of Typotheque fonts from FSI’s?

Frode Bo Helland's picture

@ Jens: Any reason for the lining figures in the web version?

Jens Kutilek's picture

I think if you want to judge hinting quality, you have to learn to look at rendered fonts the right way. Just like you need to develop an eye for proportion and rhythm if you want to design type. It helps to know what hinting can and cannot do in order to judge its quality. Differences may not be visible at a glance, but remember that letters will be repeated over and over when fonts are used, and small faults can really become annoying over time. All the little details add up.

It's true that autohinting works well for some sizes. Unfortunately which sizes are affected is always different. And if you do the hints right in one size, that usually also fixes the other sizes.

Here are some more samples, using FF Kievit. I made these using Word 2007, which switches between ClearType and Greyscale smoothing at a certain size, so don't be confused :) I've added a grey dot to the lines that use Greyscale ("Windows Standard") smoothing.

-
FF Kievit Thin Italic. First line is Autohinting, second line FSI hinting. Look for inconsistent stem widths.

-
FF Kievit Light. Lines 1 & 3: Autohinting, 2 & 4: FSI hinting. First two lines: Autohinted CDO are flat on the bottom, vertical proportions of a are off, g has a black spot, pq have kinks in the bottom, u looks rectangular.

Second two lines: inconsistent pixel widths of stems.

-
FF Kievit BoldItalic: In the first line (Autohinting), inner stem of B is too thin. Upper right corner of PR is too dark. Stem widths are inconsistent between top and bottom of bowls in abdgop, also visible in the bigger size.

Third line: stem of B gets wider from bottom to top, bottom bowl is too dark on the right curve. Q is too thin where it touches the baseline. Spine of S is too thin.

-
FF Kievit Black: First line (autohinted) is barely readable. I've marked some more of the problems I'd correct manually. The fifth line shows some errors that are apparent in Greyscale mode but not in ClearType: The top bowl of B is farther right than the bottom one, but in the actual design it's of course the other way round. S leans to the right. Left curve of e doesn't follow through. Dot of i is misplaced.

-
I'm not so sure any of these comparisons should be used in marketing. It's too complicated, I doubt a non-typedesigner person will even have an idea what this is all about. And if you take the effort to explain this to your font-buying customers, keep in mind they are not the ones who will be reading the fonts most. Especially with regard to web fonts. Web site visitors will not care what effort went into hinting, they just see (hopefully) if a font looks good or not on the whole.

aluminum's picture

I know analogies are bad and quite flawed when comparing...but Jens post got me thinking about cup holders.

Cup holder are likely the least important aspect in terms of judging a car's quality overall, but it's what people focus on. Hinting is the core indicator of quality for on-screen type but most folks are likely just going to be looking at price or ease of implementation (ie, the cup holder) and "yea, that looks good enough".

I appreciate all the samples being posted...it's really useful to see the differences.

For those of you taking the time to do the proper hinting, do you feel it's worth it or that perhaps, if we're lucky, in short time screen resolutions will finally start creeping up again ultimately making it less of an issue?

I'd like to believe that a premium price for premium hinting would be a selling point, but, well, clients are cheap.

dezcom's picture

Great samples, Jens! Thank you!

Frode Bo Helland's picture

Thanks for posting these samples, Jens. Perhaps you are right about this being to complicated for marketing, but for me seeing the comparisons is the only way to tell if I get my moneys worth. That said, I trust FSI to deliver high quality work.

Btw, where you at Typostammtisch last week?

Richard Fink's picture

@jens

I'm not so sure any of these comparisons should be used in marketing. It's too complicated, I doubt a non-typedesigner person will even have an idea what this is all about.

This is so endearing. You're right, us non-typedesigner persons couldn't have any idea what this is all about. Thanks for climbing down off your pedestal and letting us know.

@dberlow
missed the long discussion between myself and others over just exactly what the CT collection represents.
What it represents? Ah, so deep, so lofty, so metaphysical. And such doubletalk, too! You're such a kidder. What it represents really doesn't sound worth my time, frankly. What the CT fonts are, I can see for myself, but thanks.

@sii
The model for licensing fonts for print production is unlikely to change in the foreseeable future. The fact that most font EULAs do not currently allow EOT and WOFF conversion enables this and other new models.
I don't understand the "enables this and other new models" part. What models do you mean and what's WOFF/EOT got to do with licensing for print production? Serious. Not following at all.

regards,

rich

miha's picture

You're right, us non-typedesigner persons couldn't have any idea what this is all about.” (Richard)
Looking at your blog post about hinting almost three months ago, I am not sure you really understand it.

Thomas Phinney's picture

@Rich:

I believe Si's referring to the idea that most existing desktop licenses do not allow for web font usage (for example, not allowing EOT conversion), which is what creates a "gap" allowing for new licensing models for web fonts.

Cheers,

T

Rob O. Font's picture

> What the CT fonts are, I can see for myself, but thanks.
lol. who's kidding now? so how do they look on the Mac?

cheers!

Frode Bo Helland's picture

I just grabbed this screenshot from Typekit on a Windows computer. Are there less hinting involved in the fonts offered there than those offered directly from Fontfont?

Btw, there are some jpeg artifacts (I had no professional image editing program at hand), but the appearence of the fonts isn't that far off from how it looks on this screen.

dezcom's picture

Wow! That looks pretty bad, Frode. Is this normal looking for you?

Frode Bo Helland's picture

I'm usually on a Mac. I borrowed this computer from my brother, It's a laptop running Windows Vista. The web browser is Google Chrome and Cleartype is not enabled (by default?).

John Hudson's picture

Frode, these screenshots are fully aliased, with no font smoothing at all, which is very odd as this kind of rendering is pretty uncommon these days, and at these sizes borders on the prehistoric.

Cleartype is not enabled

Options for font smoothing are ‘None’, ‘Standard’ (greyscale), and ‘ClearType’. This is ‘None’, unless these fonts have gasp table settings that disabled smoothing in ‘Standard’, which would be very unusual.

Frode Bo Helland's picture

The funny thing is that Frank Chimero's website using FF Tisa Web looks amazing.

The font smoothing on this computer is set to "None".

Richard Fink's picture

@frode
Hey man, gotta upgrade from that 386 Windows 3.0 machine.
But seriously, how do the web safe fonts like Verdana, Georgia, etc.. look in comparison? As has been noted many times, cleartype/font-smoothing can be turned on and off. Also, what kind of screen?
The baseline is: how do the intalled workhorse fonts look in comparison? You can't evaluate in a vacuum.

@miha
Ah, a faithful subscriber. Thanks for the link. Well, it's a pleasure to lock horns with someone besides Hudson or Berlow for a change. You're right, I don't understand hinting. Enlighten me.

@db
so how do they look on the Mac?
Don't rightly care at the moment.

@TP
Thanks for explaining. Frankly, I'm still having trouble understanding it. Will give it another go tomorrow. Maybe the light will go on.

Rob O. Font's picture

>The font smoothing on this computer is set to "None".
Frank that specimen of the other Frank's site looks like the Mac, where "none" is not an option after OS X.

Cheers!

Frode Bo Helland's picture

Both screenshots are from the same Windows computer running Google Chrome. I didn’t change the settings.

Santiago Orozco's picture

Let's say that the business model is on, everyone is happy, type designers get royalties and web designers use the fonts on their websites

@sii and @kari get the main problem: Failing on font rendering quality is the bigger risk, Poor rendering quality would kill every distribution model.

It actually looks horrible with Windows' ClearType –about 58% of market share–

@Don McCahhill, Perhaps the solution would be to make digital fonts dependent on the OS, so that when you change from WinXP to Win7 you need a new font library.

Kind of yes, microsoft will switched to DirectWrite on Win7, the drawback is that IE9 CANNOT be installed on WinXP, forcing the consumer to buy a license for Win7, so they have to pay to have de possibility to enjoy a better Web

I have a few websites which use typekit, and as @frode's screenshot shows, you'll get the idea of how it looks on the mortals' computers.

mike_duggan's picture

if ClearType is disabled, and you made no other changes, then its likely that that the screenshot above is rendered in Greyscale. Turning ClearType off, does not disable all font smoothing. If its a TrueType font, then the GASP table settings must be set to allow font smoothing at the lower sizes. When I zoom in on the text it is Greyscale text.

mike_duggan's picture

here is how it looks in IE9, the Regular font looks to be well hinted, but the italics and bolds on the site seems to be faked.

Syndicate content Syndicate content