More FL class based Kerning

coop's picture

Hi all,

I'm using FL 4.5, and I'm doing the kerning using classes to produce a opentype font. Everything is fine apart form this output I get when I compile the opentype features.

Start of new pair positioning subtable; some pairs may never be accessed: [Y Yacute Ycircumflex Ydieresis] [agrave aacute acircumflex atilde adieresis aring amacron abreve] [WORK:Applications:FontLab 4.5:fontlab.fea 114]

I've checked the classes to make sure there is no duplications, and the kerning for the key glyph. The kerning seems to be working fine, so I carn't work out what's causing this output.


k.l.'s picture

Hello! I think all aspects have been touched in other threads already. Spot Adam Twardoch's description here:
But these also mention things to consider when using classes:
And a few more:
Seems it's one of the favorite subjects.  ;-)

dezcom's picture

Isn't FL 4.6 a free upgrade?


k.l.'s picture

I think so. And FL Studio 5 makes some things even easier (like assigning rightside/leftside flags for kern classes right in the classes panel).

However, neither of them would free you of setting up kern classes correctly.  :)


dezcom's picture

Darn it, wish they did :-)


twardoch's picture

> Isn’t FL 4.6 a free upgrade?

Even if it was, it is not anymore. We’re not issuing any free updates for FontLab 4.5 or 4.6 anymore. Even though they did help in some matters, these versions of the software are still behind the current font format needs. Especially if you’re making OpenType fonts, I’d definitely consider upgrading to FontLab Studio 5, since previous versions of FontLab simply did not completely fulfill the "best practice" recommendations for OpenType fonts. Unfortunately, the reality of OpenType for the last 2-3 years has been that the OpenType specification has been frozen and a lot of un- or underdocumented fixes and recommendations were issued by Adobe and Microsoft. FontLab Studio 5 takes these into account.


Miguel Sousa's picture

Adding a few more references to threads related with the Start of new pair positioning subtable; some pairs may never be accessed error,

Miguel Sousa's picture

And one more,

(Can a moderator please correct the typo in the thread's title Kering -> Kerning. Thanks!)

Syndicate content Syndicate content