My department (editorial) has just upgraded to CS5.5.
We're seeing differences in the way type is scaling between documents created in CS4 when opened in CS5.5.
The font involved is Myriad Pro Black, part of our Adobe Myriad Pro family.
If I save the CS5.5 doc back as an IDML and re-open it in CS4 everything is back to normal.
I'm finding it difficult to compare the two to see whether it's point size, kerning or tracking because if I cut and paste a text box from 5.5 to 4 it pastes as an object, I'm guessing it pastes as PDF as it's seen as coming from another aplication.
Is anyone else seeing differences in character widths or line lengths?
This is potentially a major problem for us as we have a large amount of legacy material in previous issues an promotional documents that will need to be re-proofed every time an alt is made.
There are changes in the text engine with every version. Opening a legacy file will cause text to reflow as soon as you touch the frame containing it. You have two options, continue to work on old files in the original version and create new files inthe new version, or live with it.
You can force recomposition in two ways and get it over with immediately, by exporting .idml or .inx from the original and opeing that (a good idea, anyway, when moving legacy files to CS5), or by using the keyboard shortcut Cmd + Opt + / or Ctrl + Alt + / depending on the OS.
What I'm looking at here in a sample file is a single line of 11 characters at 36pt which have increased in length by 2.2mm.
Text in a regular and semi-bold weights of the same face are unaffected, body copy in tight text boxes has not re-flowed and lines are of identical length in both versions.
The difference manifests itself with kerning set to optical on specific fonts of the same family. When set to metrics the variation disappears.
So, whatever the cause, and yes, I understand that text composition calculations change accross versions and we'll have to live with it.
What complicates the understanding of this is that, in the UK at least, I've found it impossible to buy fonts direct from Adobe because we don't have a US mainland registered credit card. Therefore we've had to buy them from Monotype in the UK as they're the registered Adobe font re-seller. We've found that font files supplied to Monotype by Adobe to re-sell have certain features disabled, even though the version numbers match, such as Opentype fractions (this has been confirmed by Monotype engineers) in InDesign and as a result our font set (Myriad is our main face) is a mixture of the Adobe Myriad Pro supplied with InDesign (to be able to use native fractions) and Monotype flavour Adobe Myriad Pro for the extended set. This raises doubt over whether it's InDesign or Indesign's interaction with mixed version fonts when anomalies occur for us.
Thanks again for your prompt reply and assistance.
Can you post some screen shots of the difference you are seeing?
You mentioned Optical Kerning, and that might be the clue. Optical Kerning is using an algorithm to calulate spacing based on the glyph shapes, and that algorithm may be changing between versions. Optical kerning is really intended more for working with glyph pairs from different faces or styles, not a s a substitute for font metrics in running text.
CS4 on the left, CS5.5 on the right.
I've selected the top line but as you can see the two below have also extended. Copy wrapped in the round shape seems unaffected as is that in boxes bottom left.
All type is Myriad Pro. Text in the middle three lines shows as being mixed kerning, when set all to metric length equalises, when all to optical lengths vary.
All other text on the page is set to metric, however when this text is set to optical the results are the same in both applications.
So, in this case it looks like a change in the Optical calculation, but only with certain fonts?
Yes Peter, I've just noticed it in the two lines bottom right, they also have -10 applied and optical, so only where optical and tracking are applied.
I think this may be coincidental though. I went back to the native CS4 doc, set the tracking to zero on the selected text above and opened that in 5.5. The difference is still there minus the tracking adjustment.