- Messages
- 6,508
I like it as low as possible, but 13ms on an electronic drumkit is pretty damn playable.
Most people start to notice latency around the 15-20ms zone.
The Latency Thread?
- Thread starter pipelineaudio
- Start date
I like it as low as possible, but 13ms on an electronic drumkit is pretty damn playable.
Most people start to notice latency around the 15-20ms zone.
- Messages
- 6,508
People who buy e-kits regularly have to put up with 10+ ms, and that is entirely in the box, keeping the computer out of it. The TD50 quotes 7ms, but I'm not sure how accurate this is, or even if it is consistent across all kits or DSP loads.
You typically get 3ms just because of the scan time required to track the voltage, convert it to digital, then convert that to something usable that the synth engine can translate into a midi note.
If you're using a DAW and the soundcard is giving you 2ms round-trip (doubtful, but depends on the gear) then you've still got to add the scan-time and the MCU conversion time to that figure.
Even the polarity of the piezo's can affect the final latency.
No-one is achieving 2ms round-trip latency with electronic drumkits, AFAIK.
Here's an interesting comment:
Ages ago I set up an experiment where I had a drum module that had different latencies on different pads (it was surprisingly easy to do). And in my experience, it was only at the higher limits that drummers could tell which was slower. At fast latencies, drummers couldn’t generally tell difference between 2.5ms and 15ms.
Latency - Mike Dolbear
OK, I’m surprised I hadn’t covered this, but it appears I haven’t, and it is VERY important. If you use ANY electronic or mechanical gear, you experience latency. Latency is the time it takes for something that you are using (a car, a toaster, a clever bit of electronic drum gear) to react to you …mikedolbear.com
Also:
Last edited:
James Freeman
Rock Star
- Messages
- 3,699
Then there is a factor of whether you can hear your instrument acoustically along with the processed sound.
I have noticed that I can tolerate longer latency times if I can't hear the guitar acoustically.
I have noticed that I can tolerate longer latency times if I can't hear the guitar acoustically.
metropolis_4
Rock Star
- Messages
- 2,923
When I was using a laptop rig and could monitor latency I found I started to notice it when it got above ~7ms and it started to bother me when it was above ~10ms or so.
I also remember walking the house with a wireless 100 feet from the stage during sound check and thinking it felt a little weird but I could still easily play with the band.
The general rule of thumb seems to be the more you focus on it the more you notice it, and the more you look at and talk about raw numbers the more you start to worry about hypothetical problems you likely will never actually experience
I also remember walking the house with a wireless 100 feet from the stage during sound check and thinking it felt a little weird but I could still easily play with the band.
The general rule of thumb seems to be the more you focus on it the more you notice it, and the more you look at and talk about raw numbers the more you start to worry about hypothetical problems you likely will never actually experience
Jim Roseberry
Groupie
- Messages
- 76
Point taken, but my point is still valid.
Lower latency is always going to feel more responsive/immediate.
I think there's a propensity to want to "buck" this idea... because it can be expensive/difficult to achieve effective (glitch-free) low RTL (especially under decent loads).
Part of why playing an acoustic drum kit is more "alive" is precisely because the drum heads and cymbals are immediate/responsive.
BTW, I no longer have a TD-50 based Vdrum kit. Part of what I didn't like was the lag.
Playing the TD-50 wearing headphones, pull either the left or right off your ear... and play.
The lag is pretty obvious.
Years ago, when I created Purrrfect Drums and then Purrrfect Drums 2... I did a lot of triggering drum samples via DrumKat, TrapKat, Vdrums, etc.
There's zero doubt in my mind that when the DAW was set to ~3ms RTL, it felt a whole lot more immediate than when it was set to ~10ms RTL.
To your point, there was also latency from the trigger process itself.
Some have perfect-pitch.
Some are more sensitive to latency.
Some are distracted by fan noise. (drives me crazy - I've sent back an expensive Neve Summing Mixer because of fan noise).
Neve refers to the fan as quiet.
Some can tolerate the noise, some can't.
Lysander
Shredder
- Messages
- 2,235
I humbly disagree. Lower latency is always desirable, but there's definitely a threshold where differences becomes imperceptible.
That threshold tends to be higher than most people tend to realize too, and it has everything to do with human performance limits.
Jim Roseberry
Groupie
- Messages
- 76
And before doing that, you need to do a loopback test... to make sure the audio interface driver itself isn't causing an offset.
Many ASIO drivers don't report their latency accurately.
This causes a time offset.
Take a short spike (high transient) audio signal, route it from analog output to analog input... and record the result.
Zoom-in fully... and you'll see if the two spikes are perfectly aligned in time.
If there's any difference, you need to add a manual offset (in samples) in the DAW.
Jim Roseberry
Groupie
- Messages
- 76
Not all humans have the same exact performance limits.
Watch Stan Lee's Super Humans
Some can climb much faster than a typical human... due to muscle composition.
Same with world-class sprinters.
Some have accuracy with aim that most people can't achieve.
For someone with perfect pitch, an A4 a few cents out may drive them crazy.
Doesn't bother most folks.
While it's fun to poke at Steve Vai's 2ms comment, he may be able sense the additional lag.
laxu
Rock Star
- Messages
- 5,403
Which we can't really prove without putting Vai to a blind test. It's possible he can spot it, he is from another planet after all!
But for us mere earthlings it's not a problem.
Lysander
Shredder
- Messages
- 2,235
Which we can't really prove without putting Vai to a blind test. It's possible he can spot it, he is from another planet after all!
I can confidently state he cannot. And this is not guessing either: no one can. No more than no one being able to run 100mph, or seeing infrared light, these are inherent limits to how all of us are built.
Thing is, i'm also 100% certain he believes he can spot it... and that makes all the difference in the world. Psychoacoustics are weird; if one is convinced to hear something, you will hear it. So if Vai is convinced a few ms can throw his playing off, and builds his rigs accordingly, well... that benefits him, and all of us.
Sascha Franck
Rock Star
- Messages
- 6,079
Perceiving latency isn't exactly related to musical time, groove or awareness of either. Sure, there's overlapping areas, but they're defenitely not the same.
- Messages
- 6,508
You're a science guy yourself. I follow your posts on computing and gear because you're a knowledgeable guy. But let us present the facts.
For example, you made no mention of scan times nor microcontroller processing times, both of which typically add up to 3 or 4ms themselves. This means that the moment you hit the pad, to the moment you hear a sound, there is a baseline 3ms latency response as standard. Even if you're getting a round-trip of 2ms - more or less only possible via Thunderbolt at 88.2khz sample-rate, with specific devices - your total latency will be 5ms.
Sascha Franck
Rock Star
- Messages
- 6,079
To illustrate (and yes, for yourself, if you like) how sensitive we are when it comes to latency: Double a signal in your DAW of choice, pan both hard L/R, add some samples of delay (using track parameters, a 100% wet delay /w 0 feedback or whatever) just to one of the signals. That's how incredibly great at detecting the smallest latencies the combination of our ears and brains is.
It's likely also a matter of evolution. When the leaves of the jungle behind you were rustling back in the days, you simply had to know whether it was slightly left or right because that might've easily become the deciding factor for that approaching sable-toothed fellow to have a nice meal or to stay hungry.
For more scientific explanations, one could always look up the Haas effect.
Note: No, this isn't all too directly related to amp modeler and DAW latencies - it just goes to show that at least in general, our senses are absolute suited to deal with extremely small latencies.
It's likely also a matter of evolution. When the leaves of the jungle behind you were rustling back in the days, you simply had to know whether it was slightly left or right because that might've easily become the deciding factor for that approaching sable-toothed fellow to have a nice meal or to stay hungry.
For more scientific explanations, one could always look up the Haas effect.
Note: No, this isn't all too directly related to amp modeler and DAW latencies - it just goes to show that at least in general, our senses are absolute suited to deal with extremely small latencies.
Lysander
Shredder
- Messages
- 2,235
That is, by definition, not latency. You're experiencing phase misalignment - and yes, your ears are incredibly sensitive to it.
It's similar to how our brain picks up minuscule differences in relative pitch, and the reason why you can tune instruments by ear very precisely.
- Messages
- 2,052
Comb filtering is latency based. Generally detectable by people at around 6ms of differentiation.
James Freeman
Rock Star
- Messages
- 3,699
You're right, I should have added more info in my post.
Point was to give another perspective of how short of a time 10ms is in relation to music and how even trained musicians are never that accurate.
I believe that hearing the instrument acoustically together with the processed sound is responsible for a lot of the hullabaloo about latency.
pipelineaudio
Shredder
- Messages
- 1,931
I said this before (and received a ban for it elsewhere) but, when I was testing the iPhone iPad stuff with a few hardware companies, pretty much the lowest Round Trip Latency we could get was around 20msec....And this was with chains where people who complained about the "much too high" round trip latency of like 4-8 millisecond on a DAW, yet somehow were just fine playing onstage at 20 ms.
Pretty comprehensive studies have been done for musicians tracking (when there wasn't an audible parallel path within 6dB of the amplified level at the same time) came out with some Paredo principle type numbers of like 75-80% being ok until 11-13 milliseconds. Among the leftovers were people who were really finicky even at 5ms and people who could discern it but didn't bother them until that 11ms or so
Pretty comprehensive studies have been done for musicians tracking (when there wasn't an audible parallel path within 6dB of the amplified level at the same time) came out with some Paredo principle type numbers of like 75-80% being ok until 11-13 milliseconds. Among the leftovers were people who were really finicky even at 5ms and people who could discern it but didn't bother them until that 11ms or so
Lysander
Shredder
- Messages
- 2,235
Yeah, but "differentiation" is the keyword here: you need two sources at play. Comb filtering emerges from phase interactions on two identical signals propagating, and merging back, at different speeds.
Not what people refer to when discussing latency, and particularly digital latency, which is, by its very nature, frequency- and phase-invariant.
Last edited:
