A new metafont (beta test)

A new metafont (beta test)

[Taco Hoekwater]

Hi all,

I just uploaded 

	http://www.elvenkind.com/~taco/elogo/elogo.zip

This zip contains the metafont source for the 'elogo' font.

Elogo is an extension of the normal Metafont logo font, roughly
covering 7bit TeX encoding. The font isCaps/Small caps, with 
the lowercase letters identical to the ones in mflogo (where
applicable).

The encoding is much like TeX Typewriter, with the following 
exceptions:

- there is no 'visible space' (nor will there be)
- there is no underscore character 
  (just discovered I forgot it, place is used for dot accent)
- there is a Double Acute/Hungarian umlaut at position 6
- there should have been an ogonek at position 7 
  (I picked the wrong location for invquestion :-))
- No greek, these locations are used for various accents
- no F ligatures (would have been pointless)

Other noteworthy things:

- The -- ligature doubles as minus sign
- Much like the original, this font is quite broad
- italic corrections are added to nearly all characters
- the italic correction for "t" has been adjusted
- all numbers/+/--/* etc. have the same width
- I may yet do an EC version, but only if there are enough 
  people are interested in it. 

Characters I personally don't like much: "&" (very hard to come 
up with a good shape that fits in), "$" (perhaps one bar would be 
better), "3" (plain ugly), "AE", "OE" (these don't scale well, 
the lowercase versions are much better), "invquest", "invexclam"
(maybe they need to move up to the baseline again).

Have fun, and let me know what you think:

	taco@elvenkind.com

Taco

Re: A new metafont (beta test)

[Taco Hoekwater]

Hi again,

Sorry for this followup to myself, but I just uploaded a new 
version that corrects some of the remarks I made tonight :-)

> - there is no underscore character

There is one now, dotaccent moved to position 5.

> - there should have been an ogonek at position 7

fixed.

> Characters I personally don't like much: "&" (very hard to come
> up with a good shape that fits in), "$" (perhaps one bar would be
> better), "3" (plain ugly), "AE", "OE" (these don't scale well,
> the lowercase versions are much better), "invquest", "invexclam"
> (maybe they need to move up to the baseline again).

Changed all of these a little and they look much better now IMO. 

There are also a number of minimal changes to make low-res look better.
I'm still interested in comments, of course.

Greetings again, Taco

Re: A new metafont (beta test)

[Hans Hagen]

At 12:34 AM 3/4/01 +0100, you wrote:

>Characters I personally don't like much: "&" (very hard to come 
>up with a good shape that fits in), "$" (perhaps one bar would be 
>better), "3" (plain ugly), "AE", "OE" (these don't scale well, 
>the lowercase versions are much better), "invquest", "invexclam"
>(maybe they need to move up to the baseline again).

Ah, so we agree on that. AE is more or less acceptable. Mabye you can turn
them into to touching chars, i.e two up/down strokes in the middle (didn't
test it). The 3 can have a straight top line and a straight slightly curved
(like v) line going from top right to the center. I didn't even recognize
the $. Make it an euro -) Concerning &, maybe an alternative et (small e
with t). The spanish stings, well, it took me some beers to recognize them
as such, but then, many fonts have funny ones, so that's ok, since one will
recognize them by their place in the sentence. 

Nice display fonts! 

Hans    
-------------------------------------------------------------------------
                                  Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                      Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
 tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------

Re: A new metafont (beta test)

[Hans Hagen]

At 12:34 AM 3/4/01 +0100, you wrote:

>	http://www.elvenkind.com/~taco/elogo/elogo.zip

Ah, bitmap fonts. 

(1) somehow ljfour is not known, what should i change add ? 
(2) when i run 'm and get 2602 files i don't get tfm's 

So, what I did was: 

mpost -ini &mfplain \dump 

move the mem to the web2cpath 

mpost &mfplain elogo10
mpost &mfplain elogobf10
mpost &mfplain elogod10
mpost &mfplain elogosl10
mpost &mfplain elogosl9
mpost &mfplain elogo8
mpost &mfplain elogo9
mpost &mfplain elogoslbf10

texexec --fig=c elogo* --pdf 

and voila, got 'm, so making type 1 should be possible since the glyphs
look okay. 

Some chars have a nice subtle curvature, i assume you went to the max in
that. Nice. The imaginary low x height was probably a strong driving force. 

I think that in general the chars are nice, but i'm not sure about the oe
ligature and the &. Also, the 3 is the only one with a small hook left/top
(straight line would be better). 

Hans

-------------------------------------------------------------------------
                                  Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                      Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
 tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------

Re: A new metafont (beta test)

[S2P development]

Hans Hagen wrote:
> 
> At 12:34 AM 3/4/01 +0100, you wrote:
> 
> >       http://www.elvenkind.com/~taco/elogo/elogo.zip
> 
> Ah, bitmap fonts.
> 
> (1) somehow ljfour is not known, what should i change add ?

Strange. modes are defined in modes.mf, and typically there
is an entry for ljfour there. perhaps ljIV or ljfive? But it
doesnt really matter, any local bitmap mode should do. Try
"localfont", which definately should do *something*

> (2) when i run 'm and get 2602 files i don't get tfm's

That's a feature. :-) Undefined modes fall back to proofing
mode.

> So, what I did was:
> [.........]
> 
> and voila, got 'm, so making type 1 should be possible since the glyphs
> look okay.

Actually it doesn't convert well at all, a side effect of using
an elliptical pen (sigh.). I'll try to convert the pen into
a path that looks the same (that's an interesting experiment,
involving redefinition of 'pickup' and 'draw').

> I think that in general the chars are nice, but i'm not sure about the oe
> ligature and the &. Also, the 3 is the only one with a small hook left/top
> (straight line would be better).

"et"  might be better than &, but under the constraints this & is the
best
I could come up with. The changes to the 3 sound good. 

I'm working on an outline font now, but I'l come back to this font
as soon as there is a beta version for that font. 

Greetings, Taco

Re: A new metafont (beta test)

[Hans Hagen]

At 10:03 AM 3/13/01 +0100, S2P development wrote:

>involving redefinition of 'pickup' and 'draw').

metafun redefined draw occasionally, so you can see the trick there: it's a
two step thing. 

Hans
-------------------------------------------------------------------------
                                  Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                      Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
 tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------

Re[2]: A new metafont (beta test)

[Giuseppe Bilotta]

Hans wrote:

HH> The spanish stings, well, it took me some beers to recognize them
HH> as such, but then, many fonts have funny ones, so that's ok, since
HH> one will recognize them by their place in the sentence.

I haven't seen the fonts so maybe I should not talk, and it probably
has been done this way. But isn't jut turning them upside down enough?

Giuseppe

Re: Re[2]: A new metafont (beta test)

[Hans Hagen]

At 01:13 AM 3/14/01 +0100, Giuseppe Bilotta wrote:
>Hans wrote:
>
>HH> The spanish stings, well, it took me some beers to recognize them
>HH> as such, but then, many fonts have funny ones, so that's ok, since
>HH> one will recognize them by their place in the sentence.
>
>I haven't seen the fonts so maybe I should not talk, and it probably
>has been done this way. But isn't jut turning them upside down enough?

Beware, they maybe real special. It's like saying that an ogonek is a
cedilla and a polish accent can be replaced by any accent. There is often a
lot of history and reason behind these things. cut and past and rotate and
scale is seldom enough. 

Hans  
-------------------------------------------------------------------------
                                  Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                      Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
 tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------

Re: A new metafont (beta test)

[Denis B. Roegel]

`Taco Hoekwater' wrote
>   
>   
>   Hi all,
>   
>   I just uploaded 
>   
>   	http://www.elvenkind.com/~taco/elogo/elogo.zip

elogo seems larger than logo. Am I right?
I thought it would be upward compatible with logo.
I made my tests with a modified mflogo.sty and ulogo.fd.

Denis

Re: A new metafont (beta test)

[S2P development]

"Denis B. Roegel" wrote:
> 
> elogo seems larger than logo. Am I right?
> I thought it would be upward compatible with logo.
> I made my tests with a modified mflogo.sty and ulogo.fd.

I should have been clearer about that. Try "metafont", instead
of "METAFONT". The lowercase letters of elogo10 are identical
to the uppercase ones in logo10. I needed to do it that way
because I thought otherwise some characters got too cramped.
(10/14 of 6pt => gives an x height of about 4.2pt. It's hard
to fit a pen with dimension .66ptx/.60pty in that)

Greetings, Taco

Re: A new metafont (beta test)

[Denis B. Roegel]

`S2P development' wrote
>   
>   I should have been clearer about that. Try "metafont", instead
>   of "METAFONT". The lowercase letters of elogo10 are identical
>   to the uppercase ones in logo10. I needed to do it that way
>   because I thought otherwise some characters got too cramped.
>   (10/14 of 6pt => gives an x height of about 4.2pt. It's hard
>   to fit a pen with dimension .66ptx/.60pty in that)
>   

Thanks.

Denis