Is there any specific difference from normal aliasing and specifically texture aliasing?

Also, is there any specific data structure which can be used to avoid aliasing?


1 Answer 1


What do you mean by "normal" aliasing? Aliasing is the result of under-sampling: i.e being given a signal with a high frequency (temporal and/or spatial) that is being sampled too infrequently (it needs to be sampled at a rate exceeding 2x its maximum frequency OR band-limited/pre-filtered until its max frequency is < 1/2 the sampling rate).

With texture data, the common pre-filtering approach is to use MIP maps (see Lance Williams' "Pyramidal Parametrics".

FWIW I like to explain aliasing by the following:
Imagine you work for a an office-based company (i.e pre covid WFH etc), and are at your desk solidly, 9~5, apart from a lunch break and a 15 min coffee break at 10:30. The CEO take a regular stroll around the office at 10:35 and thus always sees you in the kitchen, rather than at your desk. Because she is under-sampling your 'signal', she thus assumes you are never working.

  • 1
    $\begingroup$ Thank you for this answer. I was actually a bit confused about aliasing itself. That is why I framed the question poorly. $\endgroup$
    – Academic
    Jan 12, 2023 at 11:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.