That's because there is an old character set hack to swap to very partial implementations of Russian and Czech. It has the basic character set but no accents. It doesn't align to any codepage out there and the hack has been explicitly disabled on newer versions of AT5 (since there is full unicode support) so the hack is better to be implemented on the back end of the IMC without the user's knowledge of the exact implementation and only when Cyrillic or Czech are used on a pre AT5 Version 12 dataset.

I'm planning to do a string replacement on the back end through the IMC but I don't know if it will make it into the next release.

There is nothing that prevents the user from doing this now if you know the formatting of the hack.

As a quick breakdown, these characters are supported through a hack in earlier versions of AT5 in addition to the Windows-1252 codepage but are not yet supported through an auto convert through IMC:

Russian (U+401, U+410-U+44F, U+451)
АБВГДЕЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯЁ
абвгдежзийклмнопрстуфхцчшщъыьэюяё

Czech (U+10C-U+10F, U+114, U+115, U+147, U+148, U+158, U+159, U+160, U+161, U+164, U+165, U+16E, U+16F, U+17D, U+17E)
ČĎĔŇŘŠŤŮŽ
čďĕňřšťůž

I'll need to test the build to make sure my changes are working but this conversion should work on my internal build.

There appears to be an issue with the Elite 5 with the letters IE/IO and on the Elite 7 HDI with the letter IO. All other units appear to render correctly.