In the DJ world we have something called Digital Vinyl System ("DVS"). DVS allows you to use time-coded vinyl to control playback of digital files on your turntable. DVS has a latency measured in milliseconds (while true analog vinyl has no appreciable delay at all.) Where the delay really becomes an issue is with scratching because of all of the fine motor skills involved. Anyway, the reason this is applicable here is because when using DVS you get a good idea of how much a few milliseconds in delay can affect your ability to accurately reproduce movements that have no delay in the analog realm.

Like others have said, 1.5 ms of delay is not very much delay. It's so small that many people would not recognize it at all. However, as some others have responded variability in the delay can be a real problem. I'm a fairly advanced scratcher and the latency on my system is about 6 ms, which is perceptible but not so much that I can't adjust. It's not perfect, but it's pretty darn good. That being said, its VERY important that the delay is not variable because the reason you can deal with 6 ms of delay is because your mind and body adjust to accommodate that 6 ms. So, while my mind and body only take a few seconds to adjust from true analog vinyl to 6ms of delay in DVS, I suspect that wouldn't work if the DVS delay was a moving target. I can tell an appreciable difference between 4.5 ms delay and 6 ms of delay when doing a back-to-back A/B comparison, meaning that variability would likely be distracting. The main reason I believe that 6 ms doesn't ultimately matter too much is that it's a CONSISTENT 6 ms. Once your body adjusts then the mind just knows that everything is going to be 6 ms off and you forget about it. I imagine the 1.5 variability could really throw things off (especially for more advanced players) because your mind and body can't adjust properly.

There you go! Haha.

I was thinking about this last night. I still hold by the opinion that few milliseconds in delay is perceivable but it may be that this delay is much more pronounced in the realm of audio. I did some digging and found a interesting thread online where musicians talk about the issue.

Somebody found a study
(Nick Herbert, Elemental Mind, Dutton, 1993, p. 50.):

Quote:
How finely can we divide our little 3-second lives? The shortest perceivable time division – sensory psychologists call it the fusion threshold – is between 2 and 30 milliseconds (ms) depending on sensory modality. Two sounds seem to fuse into one acoustic sensation if they are separated by less than 2 to 5 milliseconds. Two successive touches merge if they occur within about 10 milliseconds of one another, while flashes of light blur together if they are separated by less than about 20 to 30 milliseconds.
https://www.gearslutz.com/board/so-much-gear-so-little-time/202452-perceivable-latencies.html

In the thread I linked to musicians and audio technicians talk about their own experience with latency using "soft synths" and the like. It was the "jitter" or variably that some of them complained of.

However, importantly, musicians are often using a reference when the delay is perceived. How a sound falls in a track in relation to other sounds may increase the perception. Or, it may be that the audible delay is just more noticeable for some reason.

When it comes to pinball, I'm wondering why it's just not perceivable (it's never been to me anyway). It could be that there is no clear reference, or that the perception of delay with the hand and eye is just different than the hand and ear.

The other thing I was thinking about is that maybe the tolerances in pinball just aren't small enough to be noticeable. It's not like the ramps and shots are super-super tight. There's plenty of room for some variation. It may be that 4-5 ms delay just has no appreciable effect on the ability to make a shot, therefore, you don't notice it.

Pretty cool thread regardless.

That's a really good point. I'm not sure I fully follow you though. Are you saying that pinball is all reaction time and that's why the lag doesn't matter?

I was thinking about pinball and music in terms of anticipatory timing (eg of making a shot from a flipper rolling down the inlane vs hitting a note at the correct time). My hypothesis (if you want to call it that) would be that since shot tolerances are way more generous than timing tolerances in music, the lag is such an insignificant factor that it doesn't matter. On the other hand, if you're a keyboard player and you hit the key exactly so that the note is in the pocket and the sound falls 4 ms later then the lag it's pretty noticeable.

I did hear some people talking about reaction time earlier in the thread but in my mind I was thinking about what you call "anticipatory timing".

Maybe I'm getting things twisted though.

I think you're making the same mistake I made, which is assuming pinball is directly comparable to something else where milliseconds are at play. The tolerance for lag is going to vary depending on the matter at hand. If you think about it, flying a drone is not the same fast-paced exercise as pinball. You have a lot more tolerance with the drone. Likewise the tolerances in pinball are not as tight as the tolerances in playing an instrument (especially at a fast pace like 64th notes). This is where I think I made my mistake. Also, whether or not the lag is noticeable is likely to depend on if there is a reference point in "real time". In other words, if everything is off by 300 ms you may be able to better compensate.

As an experiment I took a track I was working on and moved the bassline 300 ms later than the drums. If you can imagine the length of an average drum hit, 300 ms is about the amount of time it would take to play that drum hit from start to end). This shift is very apparent to the ear and I'm fairly certain that EVERYONE in this thread would notice that delay in button press to flipper activation. It would be distracting, even without a point of reference.

The next thing I did was shift the bassline 5 ms and I'm fairly certain that nobody here would be able to detect that shift. If a bassplayer was PLAYING the bass, and the sound fell 5 ms later, he may notice, but that's likely only because he's got a real time reference. Even then I'm not sure he'd notice. However, if the bass line was played at 64th notes instead of 1/4 or 1/8 notes, I THINK the 5 ms delay would become much more perceivable because the tolerances tighten up. This has been my own experience with small ms delay in audio. When things get fast and intricate, I tend to notice the lag a lot more. Also, if I take the drum tack, duplicate it and play one of the drum tracks 5 ms later, the 5 ms delay becomes VERY perceivable (this is called "phasing"). This noticeable difference is largely because you have a "real time" point of reference.

I'm just waxing philosophic here just like (mostly) everyone else, but I do think many of you would be surprised how 5 ms of difference can be perceivable depending on the situation. The bottom line appears to be that it's not going to be very perceivable in pinball!

Not to beat this horse anymore but I'm still trying to wrap my mind around all of this so I'm just thinking out loud here.

My original audio example came from using audio hardware to trigger sound in software. It's not just listening for delay, it's the lag between hitting a button/triggering the device and hearing the sound. The easiest way to think about it is keyboard player pressing a key. So, you press the key and X milliseconds later the sound comes out of the speaker. The way I think about this, there is the same perception, prediction, and muscle response as with pinball. If you think of a pinball rolling down the inlane to your flipper, you're waiting and anticipating the right time to hit the button so that you can make your shot. With a keyboard, it's waiting for the right time to hit your note. I don't think it has much to do with reacting to the ball flying at you (which is where I THINK the 200-300 ms might be applicable), you're just waiting for the right time to hit the flipper button to make the shot. If you hit the button at the right time, but the flipper doesn't trigger until 1/10 of second later, will that screw you up?

So in my mind the question is not your reaction time, it's how long does it take after you hit the button for the flipper to trigger, and at what point does that lag become noticeable? I think 80 ms is still quite a long time. It's not a long time if you're measuring how long it takes you to react to a something flying at you, but it's a long lime to get the response once the button is pushed. 80 ms of delay would basically feel like, if you quickly press the button, right as soon as you release the button the flipper goes off.

I suspect they put some effort into engineering the flipper lag issue and ended up around 5 ms because that's where it becomes nearly impossible for a person to perceive the lag. In that article I listed the author found that "Two sounds seem to fuse into one acoustic sensation if they are separated by less than 2 to 5 milliseconds." I'm not sure the same holds true for pressing a button and getting a response but with flipper lag being around 5 ms I wonder if that threshold is similarly applicable. Also suspect that the tolerances required in making shots in pinball aren't tight enough for 4 or 5 ms of variation to matter, which is why its not noticeable.

That's my line of thinking anyway (I am not a scientist).