Re: Sample inkjet printouts

[Rebecca and Rowland <rebecca@astrid.u-net.com>]

At 1:51 pm +0100 1/4/99, Berthold Horn wrote:
>At 00:53 1999-04-01 +0100, Rebecca and Rowland wrote:
>
>>It's easier to read to my eye, and looks more like the definitive versions
>>printed in Knuth's book.  You're quite right about the evenness of the
>>grey, but what's that got to do with anything?  I'm comparing the printouts
>>not with a subjective aesthetic standard, but with a definitive printed
>>standard in Knuth's books.
>
>Well, since Knuth use the MetaFont version it is a priori clear which one
>will pass your test :-).

Not at all: any half-decent rendering engine will produce identical results
at high resolution, as my test print outs demonstrated: there is no
noticable difference between the ATM and Metafont rendered versions in
large sizes, is there?  The *real* test of a rendering engine comes when
you render at low resolution: Metafont does a better job in this case than
ATM, because Metafont's results at low resolution are closer to the results
you get with both Metafont and ATM at high resolution.

I repeat: Metafont and ATM produce virtually identical results at high
resolution.  The designer's intention is what matters here: ATM and
Metafont meet the definitive appearance of the founts and shown in Knuth's
books when printing at high resolution.  ATM departs from this appearance
at low resolution.  Metafont doesn't.  ATM is not a good rendering engine.

>>> It would be nice then to use e.g.
>>> cmr10 at 10pt rather than cmr12 at 12pt, since the latter introduces
>>> a confounding difference in round-off that has little to do with
>>> "quality" (see below).  The obvious difference in x-height rounding
>>> in CMR12 at 300dpi obscures any other differences there might be.
>>
>>But this is *exactly* the sort of quality point I've been talking about!
>
>But who is to say which is a better representation of the underlying
>continuous glyph shape?  The one rounds down, the other rounds up.
>The one looks compact and squashed compared to the same font at other
>pix-per-em, while the other looks open and expanded  compared to the
>same font at  other pix-per-em.

But the Metafont version I sent you *doesn't* look as you describe to my
eyes: to my eyes, the Metafont version (12pt at 12pt at 300dpi) looks very
similar to the version printed in Knuth's books.  The ATM rendered version
looks vertically stretched to my eyes - certainly no more open, just
stretched.  That is however *irrelevant*: what matters is how close the
low-res version matches the high-res version in general appearance, and the
ATM rendered version (12pt at 12pt at 300dpi) looks very, very different to
the high res version.  You might prefer it, but that's an aesthetic
judgement and not relevant to this matter: all that matters is how close to
the designer's intention the printed result is.

>>Of *course* you'd expect ATM and Metafont to produce similar results in
>>limited cases.  When you're working at medium to low resolution, such
>>rounding problems are inevitable, and are a very serious quality issue.  If
>>ATM is such a good rendering engine, it ought to be able to handle all the
>>borderline cases at lower resolutions excellently.  Since it can't, it's
>>not as good as some people claim.
>
>How do you conclude it can't?  It is  more readable, and has more even
>grey.  One can now argue about which discrete approximation is better.

*YOU* say that it's more readable.  I disagree.  A more even grey?  Yes,
that's right.  So what?  This is utterly irrelevant: all that matters is
how close the printed result is to the fount designer's intention.  You can
see what the fount designer intended by looking at any of Knuth's books.
The 12pt at 12pt at 300dpi ATM rendered versions looks like a completely
different fount to me.  So while you might claim that it's `higher
quality', it doesn't matter: Knuth wants the fount to look very different
to the result you get when rendering the CM/PS CMR12 at 120pt and 300dpi.

>>> Note that 12pt at 300dpi is 59.775.. pixels.  So a x-height of
>>> 0.430555.. em corresponds to 21.448 pixels. Apparently the alignment
>>> zone at the x-height is causing the rasterizer to round this level
>>> upward in the Type 1 font while Metafont rounds it downward. CM at
>>> 12pt is where this difference shows up most noticably.  You don't see
>>> this with CM at 10pt.  This has been noted before, and I suspect is
>>>why you chose that particular size and resolution.
>>
>>I chose that particular size (not resolution - that's fixed) because it
>>shows up more problems than other sizes on my printer.  Of *course* I
>>showed you the worst case.
>
>:-)  It is interesting that even in this worst case, I don't see your
>point other
>than that ATM happens to round up the x-height and Metafont rounds it down
>(one pixel difference on your 300 dpi device).

<sigh>  The point is that ATM produces something that doesn't look the way
Knuth wanted it to look.

>>>I find the Type 1 version more "open" and more  readable.  It is only
>>>"wrong" if you assume that it should round off  to the underlying grid
>>>in exactly  the same way as the Metafont version does, which is not
>>>possible given the very different way the grid-fitting happens.
>>
>>Your aesthetic judgement isn't relevant: what matters is how well the two
>>printed versions match the definitive printed versions of the Computer
>>Modern founts in Knuth's books.  Do that comparions.  What do you think?
>
>Somehow you are stacking the deck.  If you are saying is the Metafont
>version on your ink jet closer to the the Metafont version in Knuth's book
>than the Type 1 version on your ink jet is, then we know what answer to
>expect.

(does this mean you *know* that ATM's a load of rubbish at rendering?)

Argh!  I'm not stacking the deck!  What matters is what the fount designed
wanted: the fount designer wanted his fount to have a particular
appearance, right?  The definitive look of the Computer Modern founts, as
desired by the fount designer, appears in Knuth's books.  It doesn't matter
what rendering engine was used: that is what the designer wanted.  Now
then...  Knuth fiddled around with the designs so that he got what he
wanted using Metafont at various resolutions.  The definitive appearance:
what he *really* wanted - *that* appears in his books.

As my test prints show, both ATM rendered CM/PS founts and Metafont
rendered original founts look the same at high resolution.  There is no
problem there.  The problem is that at low resolution, the ATM rendered
CM/PS version (especially 12pt at 12pt and 300dpi) looks very different to
what the designer intended.

Your aesthetic judgement has nothing whatever to do with it: all that
matters is how close the rendering engine approximates what the designer
wanted.  If the CM/PS founts rendered by ATM don't look the way Knuth
wanted them to do, then there's a problem with the quality of the CM/PS
founts and/or ATM.

Deck stacking does not come into it: the designer wanted *that* appearance,
clearly printed in Knuth's books.  The fact that he used that particular
rendering engine doesn't matter: this is the way he wanted the fount to
appear.  Forget *your* aesthic judgement, forget what *you* want the fount
to look like, forget what *you* think is the right way for a fount to look,
to be designed, and to be rendered, forget about Metafont and ATM.  What
matters is the conformance to *that* appearance on the page.  Nothing but
that: the appearance of the glyphs on the page.  Ink on paper: ink on paper
as Knuth wanted it.

I repeat: it is the appearance on the page that matters.  Knuth designed
the founts to have a particular appearance, and what he wanted appears in
his books.  If a particular rendition of the CM founts departs from that
appearance, *no matter what you might think about the appearance*, then
that particular rendition is *wrong*.

>>> So the gap is 17/1000 of an em or 0.204pt at a font size of 12pt.
>>> You don't have to make the font size very small before that is a pixel
>>> or less.  It is not a good idea to design a font with such close
>>> approaches, since it dramatically limits what adjustements are allowed
>>> in grid-fitting.
>>
>>Erm...  Surely that means it's only a bad idea if you don't pay enough
>>attention to detail when writing the code to produce the fount?
>
>Well, I tend to disagree.  It is better to write the font code so that there
>remains enough freedom at small sizes to shift things by one pixel.
>The extremely close approach of the F to the r can only be safe if
>their are strong constraints on the alignment zones which reduce the
>ability of the rasterizer to make adjustments needed for good rendering
>at low resolution.

Yes, but...  As I've tried to demonstrate, ATM makes a rotten job (departs
from the designer's intention quite radically) when rendering CMR12 at 12pt
and 300dpi, but Metafont does the job very well (conforms closely to the
designer's intention).  So I'd say that your argument holds only for a poor
rasteriser.

>>> Same in "OzTeX", and also the acronym "ATM".
>>> But with the ragged printing of the ink jet its hard to say much more.
>>
>>You've spotted yet more problems with a PS Type 1 fount rendered by ATM,
>>haven't you?  That was Monotype Joanna.
>
>I didn't know what font it was.  In that case your code for generating
>these special acronyms is too font specific (and not tuned for Joanna).

My code?  It's not *my* code.  I used the standard \LaTeX and \TeX
commands.  I didn't use any code to generate `ATM'.

>>>Anyway, I don't think we can settle this via words alone.  Maybe you can
>>>find someone with a scanner and put up the comparison for others to see.
>>>Of course, as I said, it would make more sense to show CMR10 at 10pt
>>>than CMR12 at 12pt.
>>
>>Well, no, it makes more sense to show cmr12 at 12pt, because that
>>demonstrates the problems better.
>
>You would say that :-)  I think it makes more sense to use any of a
>number of other combinations because 12pt on a 300 dpi printer
>is a special borderline case for the CM fonts.

Erm...  Precisely: that's where you can see the problems most clearly.

>>>By the way, I couldn't find information on the Deskwriter 520 on the
>
>>>HP web site, even under "discontinued products"  - so I don't know
>>>much about its properties.
>>
>>300dpi monochrome inkjet, able to print 600dpi along one axis; but OzTeX
>>doesn't allow you to use a non-square resolution.
>
>Why does OzTeX care?  With ATM you can print to any resolution isotropic,
>or not.  Which seems like another advantage, since quite a few printers
>these days have twice the resolution along the horizontal axes than the
>vertical...

I have no idea why OzTeX cares.  Best ask the author, not me.  As you know
as well as I do, Metafont can print to any resolution as well.

Rowland.