@GeekestGeek How can one mark a question as (half-)obsolete?

@GeekestGeek I think that the way to go is to upvote your comment so it sticks out...

@UlrikeFischer where does miktex leave a log from format creation? tex.stackexchange.com/a/659544/1090

@GeekestGeek cm fonts have not changed since the 1980s. The question is not "outdated" it just is/was unclear. You and the OP have shown different images, presumably from different source, but as OP did not show any source, hard to say anything.

@DavidCarlisle in my system in ...\AppData\Local\MiKTeX\2.9\miktex\log\makefmt

@UlrikeFischer could you leave a comment there to say how to rebuild formats, I assume the miktex console thing has an option equivalent to fmtutil --all ?

@DavidCarlisle there is no button for it, but one can select formats and click on build. But normally miktex is quite eager to recreate formats all the time, so one should first check what didn't work here.

@DavidCarlisle I added a comment.

@UlrikeFischer thanks

@mickep I don't know, frankly.

@Rmano Maybe. I don't know the right way to deal with this question.

@DavidCarlisle Maybe, if this question is/was unclear, one should flag it?

The adjective for metal is metallic. But not so for iron.. which is ironic. (@AlanMunn @barbarabeeton)

@GeekestGeek you can always vote to close it as unclear or leave a comment asking the OP to provide an example code. Flagging would not achieve much as there isn't really anything else the mods could do

@PauloCereda breakfastic

@DavidCarlisle oh no

@JosephWright are there more -dev updates to come? I want to upload the pdfmanagement with the new xmp code, but as it perhaps can affect pdf-based tests I didn't want to interfere with the dev release.

@DavidCarlisle I cannot vote to close myself (probably due to my low rank). Still, I posted a request for clarification and flagged the post as unclear so that the moderators can consider voting to close themselves.

@GeekestGeek sorry but what is your problem that the question is a bit unclear? The question is highly active because of the useful answers which offer a lots of variants.

@UlrikeFischer No, just base and tools had changed

@JosephWright ok. Then I will make an update ok?

@UlrikeFischer Unclear in the sense of not exactly reproducible (on my computer, and, perhaps, for the others, too) as of now. You may, of course, argue on whether the circles mostly touching the digits vs. not touching the digits is that much of a difference. For my purposes, it does make a difference. (I would not use the original encircled digits i.stack.imgur.com/V5KtM.png but would use the (admittedly, nonideal) encircled digits i.imgur.com/nDZZ14q.png .)

@DavidCarlisle tex.stackexchange.com/q/659564/52406. Time to do something about eqnarray, isn't it. (Yes, I am nagging, sorry) :P

@mickep The use name is a bit frightening: imagine opening the box and a lion jumps out :)

@samcarter Dodos are not afraid of lions. :)

@mickep ohh, brave dodos!

@mickep ooh

@PauloCereda Very aluminating.

@AlanMunn ooh

@mickep Frank and Rainer did something in 1992, ported the amstex alignments to amslatex (what is now 2e + amsmath) it's not my fault @UlrikeFischer persuaded everyone not to read the documentation.

@mickep Stability

@JosephWright True stability results when presumed order and presumed disorder are balanced. A truly stable system expects the unexpected, is prepared to be disrupted, waits to be transformed. --- Tom Robbins :)

We ducks are very good at chaos. :)

@GeekestGeek Asking to clarify a question from 2010 doesn't seem like a terribly useful thing, honestly. Using TL 2013 (the earliest I have on my machine) the problem is reproducible.

@GeekestGeek And the output is still crappy with TL2022 for me.

@JosephWright i know, just teasing @DavidCarlisle...

@mickep :)

@mickep ooh

@PauloCereda Ohh, indeed

@mickep but that only works if I do not ignore you.

@DavidCarlisle Ooh no.

@DavidCarlisle Conclusion: Uniicode is a confusing mess.

@mickep corolary: everyone should use English

@PauloCereda ÖÖh nö!

@mickep I once failed to explain to some americans that 1+{}^-2 was a perfectly standard notation, especially when teaching "negative numbers" as a topic and you need to distinguish from 1 - 2. We should apply @PauloCereda's corolary

@DavidCarlisle Yes, it is not so encouraging.

I would like to have line spacing 1.5. Should I write \linespread{1.25} in the tex document?

@JaakkoSeppälä probably better to use setspace package and \onehalfspacing

@DavidCarlisle ups. We should add a test for eplain ;-(

@UlrikeFischer I need to check my email?

@DavidCarlisle graphics-def, we got bitten by \newdimen which is outer in eplain

OUTER ARGGGGGGGGGGGGGGGGGGGGGGGG

@DavidCarlisle Screaming out your love...

@DavidCarlisle well said ;-)

@UlrikeFischer I could make it not outer in miniltx rather than mess up the main code

but I'm offline in about 2 mins for this evening @UlrikeFischer

@DavidCarlisle that would be good. But we should also have a simple eplain test.

@DavidCarlisle ok.

@UlrikeFischer or pick any of these comments chat.stackexchange.com/…

@UlrikeFischer oh you mean test as lvt file? I thought you meant a run-time test \ifx\eplainthig... avoid error \fi

bye for now

@DavidCarlisle yes I meant a lvt test, that it still works. I tend to forget that it exist.

I'm trying to figure out the problem here: tex.stackexchange.com/q/659491 Does anyone know offhand where this error message (the ext4) comes from? "pdfTeX warning (ext4): destination with the same identifier (name{page.iv}) has been already used, duplicate ignored" (@DavidCarlisle. @JosephWright)

@AlanMunn It seems that your TL2013 version is more crappy than your TL2022 version (which itself is more crappy than what I posted in a comment). Anyhow, the post has to show source code so that we are all on the same track.

@GeekestGeek My point is that the question is almost 12 years old, so asking for code seems not very productive. My code was the same as yours.

@barbarabeeton it means that there are two pages with number iv and hyperref created duplicated destinations.

@UlrikeFischer -- I know what it means. What I want to know is from what .cls or .sty file the message is generated. (I'm just not up for \tracingall.) It goes away if hyperref is removed, but I've never seen this problem with amsbook + hyperref before, and nothing else I've looked at explains what the OP is describing, namely TOC and a dedication page reversed. Before I ask for more information (don't have a tidy MWE) I'd like to figure the ext4 message out.

@barbarabeeton it is a primitive message from pdftex, you get it if you use \pdfdest twice with the same name. I didn't really look at the example, but the OP does \setcounter{page}{4} and that probably creates page iv twice.

@AlanMunn In fact, your answer is productive. Simply feel free to edit the post with our source code and your output and state the TeX Live version you used (TL 2013). This would make the question clear and the output reproducible in principle. Hypothetically, the original poster might have thought of something else, but if so, he has a chance to re-edit.

@UlrikeFischer -- Yes, commenting out the \setcounter gets rid of the error message. (Actually, two of them.) I've left comments, but can't figure out any way to reproduce the reported problem. Thanks for the helpful information.

@GeekestGeek What does it matter if this particular question is exactly reproducible or not? The main value of the post are the many good answers which show alternative methods to get centred numbers. Does it really matter how far off-centre the numbers in the question are?