I believe this works as intended - but only for internal layers when designing multilayer boards.
I beg to differ. If hiding a pad ring is
supposed to leave the mask layers as-is, why does the 3D Preview render without the mask aperture? So either the Gerber export is wrong, or the 3D Preview is - which is it? Anyway, if it's only intended for internal layers, why make the 'Hide' option even appear for top and bottom layers?
What you are trying to do is very unusual in that, engineers would not normally want to weaken the integrity of the soldered top pad by doing away with the plate through hole & having no pad on the bottom side.
Sorry, what I failed to mention is that the components I want to do this for are to be mounted on the bottom side. Therefore, by hiding the pad ring on the bottom, they are soldered only on their opposite side, as per normal for if it were a single-sided board. Doing so can't be that uncommon, as I have seen several double-sided boards where some through-hole components have pads on a single side only.
I want to do this because it makes the components in question much, much easier to de-solder (without risk of damage to the board) if they need to be replaced. I presume this is why other boards I have seen (in commercial products, no less) like this are done that way too.
In your case, you have found a way to do it, so just ignore the design rule errors that are generated. These design rule errors are not pushed through to the Gerbers, so the final design files you send to the board shop represent exactly what you want.
Indeed. But, having a large list of DRC errors makes the signal-to-noise ratio very low, such that it would be too easy to miss a genuine error. I'd rather not take the chance that I may miss some issue due to it hiding in amongst so many false-positives.
It would be nice, actually, if DipTrace had some feature to mark DRC errors as to be ignored. I believe Eagle has something like this.
If you feel that the software should have some changes made, please register on the DipTrace Forum & ask the DipTrace engineers to look at your request.
I e-mailed their support address about it, so hopefully they fix the issue or clarify what the intended behaviour is supposed to be.
In the meantime, I managed to work around by taking a quick crash-course in Gerber code, and figuring out where I need to hand-edit the code in the mask's file to change the size in the definition of those apertures.