cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 

Community Tip - When posting, your subject should be specific and summarize your question. Here are some additional tips on asking a great question. X

True type vs Type 1 fonts

PhilipOakley
5-Regular Member

True type vs Type 1 fonts

When printing to pdfFactory from a mathCAD file (V14 M030). A message appears "pdfFacory supports TrueType fonts. This document uses Type 1 fonts, which will not be correctly imaged to PDF. Do you want to continue anyway?"

The document looks ok on the viewer but I have not yet tried to print out a paper copy.

Should this be the case - that is : What's the difference between type 1 and true type fonts, and why has mathcad chosen this one.

Philip

PS If you want to paste text (e.g. from Notepad) into this edit box during preparation you need to first click on the HTML icon (top right) to switch to plain text mode, then re-click it again to get formatting back.

13 REPLIES 13

Philip Oakley wrote:

When printing to pdfFactory from a mathCAD file (V14 M030). A message appears "pdfFacory supports TrueType fonts. This document uses Type 1 fonts, which will not be correctly imaged to PDF. Do you want to continue anyway?"

The document looks ok on the viewer but I have not yet tried to print out a paper copy.

Should this be the case - that is : What's the difference between type 1 and true type fonts, and why has mathcad chosen this one.

It's a postscript font. Try CutePDF instead. That's what I use, and it doesn't seem to have that problem (in fact, I have never had any problem with it, in any piece of software!)

PS If you want to paste text (e.g. from Notepad) into this edit box during preparation you need to first click on the HTML icon (top right) to switch to plain text mode, then re-click it again to get formatting back.

Or paste the text into Word first, then into the edit box. That's not any easier, just an alternative

995 was great and simple

CutePDF , I have zapped it

PDFcreator works fine but smashes my plot borders and sometimes the line in program

DeskPDF [ScanSoft $29.95] wants to install Postscript font(s) , it finds in the Windows cabinets

PDF Converter 6 [Nuance $ 69 CD] , Creates, Converts, Edits [a great tool for reasonable $]

_______________________________

In their principles, PDF are converters from windows coding [Mathcad or else] into most popular printers. If your printer has 1200 PPI, the PDF thrives to render that resolution ... which resolution you can't see on the crappie screen 96 PPI. I can't compare PDF Converter 6 with Adobe because Adobe don't have a trial version. DeskPDF has a printing quality option in % compression [don't really understand their compression principle]. "PDF's" are virtual printer doing what some software have "Print to file".

MikeArmstrong
5-Regular Member
(To:PhilipOakley)

Cute PDF or Acrobat is the way to go.

Mike

MIke Armstrong wrote:

Cute PDF or Acrobat is the way to go.

Mike

Wait Philip tries. ADOBE prices around $ CAN 700 !

You are proposing the two extremes !

MikeArmstrong
5-Regular Member
(To:jeanGiraud)

jean Giraud wrote:

You are proposing the two extremes !

Exactly.

2 options which are 2 extremes.

Mike

PhilipOakley
5-Regular Member
(To:PhilipOakley)

I've done a bit more digging.

The problem appears to be the new MathCadUniMath font (MathCadUniMath.otf) used for all the special symbols, e.g. phi and psi (or the ohm/Omega symbol) from the Greek pallette.

These are part of the UniMath font. http://www.appliedsymbols.com/um/

This font is in "Open Type" format. It looks OK on my screen with pdfFactory.

http://www.truetype-typography.com/opentype.htm

I found this explantaion of the TrueType vs Type 1 font wars on the web.

http://www.truetype-typography.com/articles/ttvst1.htm

http://www.truetype-typography.com/ttandt1.htm

Part of 'my' problem is that this is for a corporate environment where we have an old version of pdfFactory generally available, but no ability to load say Cutepdf (which I use at home).

All that happens with pdfFactory is that the screen shot looks OK but the created pdf has blank space where the greek characters used to be!

I am pursuing the option for an update to the pdfFactory version

- Has anyone used the latest version of pdfFactory sucessfully with V14 and greek characters?

Interesting,

In short and from recollection [could be confirmed], one type uses Bézier quadratic, the other type Bézier cubic. That resumes and explicits the nebulous "different mathematical representation". Free with Linux, there is a "print to file" apparently a 1/1 quality to the expensive Adobe. Hard to imagine PTC didn't think of Linux, for sure it's not much American ! meaning small $ exploitation. Readers can ignore my last comment.

I am pursuing the option for an update to the pdfFactory version

- Has anyone used the latest version of pdfFactory sucessfully with V14 and greek characters?

Their website specfically says it supports font embedding, and that all fonts will be displayed:

http://www.fineprint.com/products/pdffactory/index.html

So it should work.

PhilipOakley
5-Regular Member
(To:RichardJ)

The new version of pdfFactory (trial version, V4.0) does appear to handle the fonts OK, however I'm not sure if the font embeding has actually happened. The properties section of the pdfFactory indicated that the uniMath was not included as an embedded font (it was on the left hand side of the 'include font' option table!), but could be as a user option.

I need to regenerate a sheet and do the saving to pdf and then post the pdf so others can check.

The other issue I noticed was that if I copy the text (e.g. a variable name in greek) from MathCAD to MS Word (using paste special : unicode) then when it is pasted into the word document the font has been changed from unuimath to other fonts, with changed glyphs.

In my system the main text is Times new roman, with one of the phi characteres being shown as Lucinda unicode. I am still able to change the font (in Word) for the text back to the MathCADUniMath font and then get the same glyphs displayed as showed in the actual mathcad sheet. But again that is an extra user action!

This just shows how tricky this stuff can be and how much care is needed.

The new version of pdfFactory (trial version, V4.0) does appear to handle the fonts OK, however I'm not sure if the font embeding has actually happened. The properties section of the pdfFactory indicated that the uniMath was not included as an embedded font (it was on the left hand side of the 'include font' option table!), but could be as a user option.

I need to regenerate a sheet and do the saving to pdf and then post the pdf so others can check.

We all have the font installed though. Just look at the pdf on a computer without Mathcad (or, more accurately, the Mathcad Unimath font) installed. If you can see the characters the font must be embedded in the pdf.

The other issue I noticed was that if I copy the text (e.g. a variable name in greek) from MathCAD to MS Word (using paste special : unicode) then when it is pasted into the word document the font has been changed from unuimath to other fonts, with changed glyphs.

In my system the main text is Times new roman, with one of the phi characteres being shown as Lucinda unicode.

I see the same thing, except that all the characters in Word are Times New Roman. That means one of the phis shows up as capital phi, and changing the font to Mathcad Unimath does not fix it.

>I see the same thing, except that all the characters in Word are Times New Roman. That means one of the phis shows up as capital phi, and changing the font to Mathcad Unimath does not fix it.<

________________________

This is a very complicated business, not new, not worth a discours,

the logic diagram behind would kill a dead stone [something like

the logic of this forum].

Pierre Bézier
Born: Paris, 1910 Died: 1999
Pierre Bézier worked as an engineer for the Renault car company. In the late 1960s he conceived a new curve formulation to represent 3D car body forms on computers.
Bézier curves, in their two-dimensional form, are now the basis of almost all common graphics programs (such as Adobe Illustrator and Corel Draw), following their adoption as the standard curve of the PostScript language. Most outline fonts, including TrueType and PostScript Type 1 (but not Ikarus) are stored as Bézier curves.
The standard Bézier curve, used throughout PostScript and in many drawing programs, is a parametric cubic (or third order) curve defined by its two end-points and two control-points, which in general are not on the curve. As well as other characteristics, the control-points define the tangents of the curve as it leaves the two end-points. A Bézier curve always remains inside a polygon drawn around all of its points (the “convex hull”).
The curves of TrueType are a quadratic (second order) version of Bézier curves, having two end-points but only one off-curve control-point. The single off-curve point then defines both tangent vectors, while the curve remains within the triangular convex hull.
PS = cubic ... TT = quadratic
Tout le monde (ou presque) connaît Pierre Bézier et ses fameuses courbes. Mais ce que l'on sait moins, c'est que, parallèlement aux travaux P. Bézier (qui travaillait chez Renault), Paul de Casteljau qui travaillait chez Citroën [ 2 ] menait le même type de travaux [ 1bis ].
Paul de Faget de Casteljau est né en 1930 à Besançon. Il est mathématicien et physicien et a fait ses études à l'Ecole Normale Supérieure de Paris. Ses découvertes majeures est celle des formes à pôle dès 1959 et surtout l'algorithme qui porte son nom et qui permet de représenter des courbes de Bézier à partir de polynômes de Bernstein. Ce dernier point est capital. Sans l'algorithme de de Casteljau, la représentation informatique des courbes de Bézier et leur utilisation au quotidien n'aurait pas été ce qu'elle est aujourd'hui, voire aurait été tout à fait impossible. Dans son livre Fontes et codages, Y. Haralambous est formel sur ce point :
“Il est dommage que le nom de Paul de Casteljau soit moins connu que celui de Bézier, puisque les courbes du deuxième n'auraient que très peu d'intérêt sans l'algorithme du premier.“ D'ailleurs et fort heureusement Pierre Bézier lui-même reconnaît l'apport de de Casteljau pour ses propres recherches. “Plus sérieusement, la propriété intellectuelle sur ce travail devrait pourtant être partagée avec Paul de Casteljau, dont je ne manque jamais de citer la contribution.“ [ 1 ]
Le principe de base de l'algorithme de de Casteljau est qu'une restriction d'une courbe de Bézier est aussi une courbe de Bézier. L'algorithme de de Casteljau est dit récursif [ 3, 4 ]
_______________________________
Conclusively:
1. de Casteljay runs either Bézier cubic or quadratic for fonts
2. professional PDF converters will convert any scalable forms
3. and take bitmaps as is

"font smoothing" a Microsoft expression and option

that's what it means.

Top Tags