Question regarding the term "latency" as it is commonly used in this forum topic.

Pardon me for opening a new thread on the already overworked topic of latency. But something has been puzzling me for some time, and I'm not sure, but I thought it might point to a hitch in communication that is the kind of minor issue that can easily get out of hand quickly.

What do you mean when you speak of "latency"?

latency n. 1. The time that elapses between the moment a note is played on a MIDI keyboard and when that note is heard or when the note appears in the score. This latency has become such a distraction I was unable to finish my 12-tone inversion/retrograde/inverted-retrograde assignment.

I haven't had any experience with this kind of latency because, for me, even though I tend to use MIDI keyboard input in other notation programs, I've found in MuseScore4, once I set up the shortcuts I like to use, I can usually achieve a higher speed with greater accuracy using a regular computer keyboard.

However, there is another kind of latency that has been driving me up a wall, and in fact was quickly becoming a deal-breaker:

latency n. 2. The delay between the moment the space bar is pressed to begin score playback (silently) and when sound starts coming out of the speakers. This latency is driving me up the wall, but that's not what caused all the dents.

I'm not even sure this kind of latency has anything to do with MuseScore4, since it begins playing back the score immediately; it's just that, for a second or two, the speakers remain silent, and eventually when the soundcard and/or rendering engine finally resets, the score begins playing wherever the play head is at that moment, and thereafter plays just fine ... until I stop it and start it again. I've tried allowing MuseScore4 to take complete control of the sound card so that no other programs could use it, but it still seems to lose track of it as soon as playback stops, and then has to scramble to start routing sound through the card again for a few seconds after playback has begun, which is obviously sub-optimal.

For my purposes, it would even be preferable if the playhead didn't move for a few seconds while the sound card gathers its wits enough to begin playback at the point I had selected.

As it is, with larger scores, that lag can make it impossible to hear specific measures because in order to compensate, I would have to start playback several measures before the one I needed to hear, and sometimes I would judge it incorrectly and the playhead would just sail right past the measure of interest, causing much profanity, objects launched at screens, etc.

I eventually ended up buying a whole new sound card, which helped reduce this lag but not completely eliminate it. Computer specs if anyone is interested.

At any rate, I'm not sure if this is correctly termed "latency," although I believe it is, and even if not, it seems like ordinary non-developers might still use the term to refer to this kind of thing, whereupon the developers might assume they're talking about the other kind of latency, or vice-versa.

--

Note:
Computer specs:

• Core i9-9820X processor
• 32G RAM
• RTX 2080ti

(It's a computer that is capable of rendering complex raytraced 3D scenes and editing video at 1080p just fine, so it has been a mystery why it has such a difficult time playing back an orchestral score.)