Distance to your amp plus wireless latency = unacceptable delay?

[Foz]

...From experience, timing (or at least the feel of timing) becomes an issue starting around 25msec...

Interesting that you independently arrive at that figure - because 25ms is an oft quoted approximation for the Haas effect breakpoint* for percussive musical signals. And yes the Haas effect will vary quite a bit for different listeners - and for the same listener for different types of signal [for example speech versus complex musical signals].

*the delay time at which a second instance of a signal coming from a different source vector is perceived as an echo versus a part of the original signal = the second signal is ignored for purposes of localization of the source if the delay is below the breakpoint

 

[Foz]

I very often have been set up at least 20-25 feet from my drummer, often with no mics on the drum kit. Not a problem at all IME, at least after doing it dozens of times. That particular drummer has a strong preference for being able to see my hands though, FWIW.

I have never had a problem as long as my rig is next to the drummer so that wherever I go the path length difference between my ear and the drummer versus my ear and my rig - is small. I have had weird problems when me and my rig where way over to the side of a large stage well away from the drum kit. Also - drummers get to look at my ass and that should be good enough.

 

[Honch]

There have been studies done on the effects of sidetone delay in communications. When it is too long it can cause real problems for the speaker. We limit the sidetone delay in our communication Invalid Link Removed to no more than 10 ms. They typically run less than five.

Make sure you use the word speaker correctly here. That article means the speaker (person) who holds a speech. I thought it was about the driver inside our 8x10 cabinets, which is - too - a speaker. Loudspeaker that is. That's why I clicked the link and thought "what have they've been smoking..."

 

[seamonkey]

Radio is the speed of light.

Analog -> Digital (A/D) and back again D/A takes micro seconds, less than 1ms on each end.

Assuming digital wireless, then there is going to be buffering, but because it has to be realtime it's a fixed delay. For line 6 it's less than 4ms for A/D, buffering, Transmit/Receive, D/A

That is way less than it takes the string itself to ramp up to a full note.

If you're 20ft from your amp with a long cord. then with wireless it goes from 20ms to 24ms. 25ms is the point where you can hear a slapback echo. It all depends on the player, and their ability to learn without thinking about dealing with latency. I think if you over think it, it's will be harder, than just sitting down and practicing for an hour. Using a PC and DAW you can add variable delay and hear and practice with it. Somehow many musicians across the world just deal with it.

 

[seamonkey]

Spectral analysis waveforms, or it didn't happen!

Search around, or pick up a free sample library, if you have a DAW they probably included some bass samples. Depends on how you play but here's a guitar string that someone was complaining about "lag" while playing on a keyboard. Keyboard fingering latency, plus natural sample delay and a a "keyboard" player may think it's lagging, when it's just sounding like a plucked string. Keyboardist just get over this with some practice. Like a pipe organ sitting way far away from the pipes in a church.

 

[agedhorse]

Latency for a digital product also includes any processing, in addition to A/D/A conversions. Often delay is added in order to provide time for side chain type processing. This is fundamental to how "look-ahead" compression and limiting are achieved.

 

[DanAdams]

Long answer... it's complicated.
Short answer... don't worry about it.

 

[Old Garage-Bander]

Search around, or pick up a free sample library, if you have a DAW they probably included some bass samples. Depends on how you play but here's a guitar string that someone was complaining about "lag" while playing on a keyboard. Keyboard fingering latency, plus natural sample delay and a a "keyboard" player may think it's lagging, when it's just sounding like a plucked string. Keyboardist just get over this with some practice. Like a pipe organ sitting way far away from the pipes in a church.

Woohoo! Pictures! It DID happen!!

 

[Downunderwonder]

The delay is a result of sound moving through air --- not electrons.

another post from someone who doesn't know what latency in electronics is, didn't read the link in the OP, yet still feels like he has something to contribute.

 

[Downunderwonder]

It's the same with some modelling on computers. When vendros says this or that milliseconds latency are enough because in real life the sound travle 6 ms at 5 feet away from you amp. What people forget is that natural latency are added on top of the digital conversion latency of computer amp modelling. So you have modellings 8-9 ms latency added ON TOP of the natural 5-6 ms latency there is. After 12-15 ms I start to get uncomfortable. On large stages you just have to use in ear monitors or ample bass on stage fills, or wedge monitors.

you are the opposite end of the scale from Passinwind.

I didn't get the impression from the Whirlwind report that there is so much variation between individuals.

 

[InhumanResource]

not so. Please, if you can't hear delay from your own monitor in the sub 30ft range just shake your head and take the ''carrots'' option but don't try to tell me it is insignificant within the practical distances of gigging.

I am deeply sorry my answer did not fit within your model of acceptable responses. Next time I will be sure to select the joke option on your survey and move along to something more meaningful.

 

[Honch]

The Sidetone delay "report" is more interesting. I remember in the 80's when drummers had to adjust and adapt to Simmons electronic drums. The sound didn't come when the sticks hit the pads. The rubber/plastic pads own acoustic click or whip sounds heard first (which was another annoying thing) and THEN came the sampled or whatever drum sounds. Even when not triggering by MIDI (which was/is by todays standars still sluggish). Even when triggered by true cv/voltage controls of each pad. The sound didn't arrive. They had to play with cans (headphones) or in ear monitors which were still in their infancy back them. I've played with several, and especially those who though that the actual drums could be electronic but the cymbals was real. Immense timing problems, that wasn't resolved until they had isolated headphones that cut the acoustic (quite loud) sound out. There's a slight phenomenon that happens, but I think in the bass world it's not so much of a problem because low notes takes time to bloom and happens. You have to stand several feet away before the lowest notes of the fundamental sine wave hits you. But...then it's too late becasue the drummer has already hit one.

It's a dilemma, and something you have to deal with and hone in for a long time. That's why huge band sounds tighter after a while touring.

 

[NoBlackTShirts]

The smallest interval that I've shifted a note to get the "feel" right when comping tracks was 3 milliseconds.

That doesn't happen often but it did happen.

 

[Slevage]

another post from someone who doesn't know what latency in electronics is, didn't read the link in the OP, yet still feels like he has something to contribute.

Sorry I bothered you

 

[Downunderwonder]

The smallest interval that I've shifted a note to get the "feel" right when comping tracks was 3 milliseconds.

That doesn't happen often but it did happen.

I don't think I would notice any difference between the tracks even if was pointed out, although I have had band directors insist on parking bass amp on top of the drummer as if those 3ft were vital.

 

[gumtown]

I have had band directors insist on parking bass amp on top of the drummer as if those 3ft were vital.

Not sure if 3ft is vital, but that does raise the issue that there are 3 elements of perspective involved.
The placement of the drums, the placement of the bass speaker,
and the relative placement of your ears in relation to the positioning of the two sound sources (in wide spaces).

Your ears are getting the timing from the drums, so the distance of your ears from the drums will add a few ms latency.
Your perception of locked timing will depend on the time you hear the drums and the time you hear your bass from the speaker.
If you, your speaker are next to the drums, everything is in time.
If you are next to the drums but your bass speaker is 50 feet away to the side, you will be actually playing ahead of the beat (perception from audience point of view).

So I suggest for the best timing, stand next to the drummer, or use analog wireless & IEM.

 

[Sid Fang]

another post from someone who doesn't know what latency in electronics is, didn't read the link in the OP, yet still feels like he has something to contribute.

I wouldn't be so harsh. I didn't read it either, until now, as I'm a EE/SW engineer myself and understand these things reasonably well. Having read it, I should have commented on earlier. It needs to be understood that it's marketing literature for an audio cable company, whose sales are directly impacted by the use of digital technology. Like all good technical marketing material, it is factually correct. It just chooses its data points carefully. All the exposition on sound propagation and psychoacoustic effects of delay seem OK and legitimately helpful. But everything about the delays involved in digital audio transmission and manipulation is made-up. They don't mention A/D and D/A conversion delays, which are the things you can't easily finesse. They call out delays like "Digital Snake 3-7ms, Digital Console 1-3ms, Digital Signal Processor 5-10ms, Total 9-20ms" without saying where those numbers came from, or what they include. They are grossly inflated relative to the actual performance of products on the market. Roland specs 1.2ms with A/D and D/A included for their digital snake. If you're running a digital snake into a digital mixer, you may get to skip the snakes D/A and the mixer's A/D stage, which is a lot of the latency. DSP chips are freaking *fast* these days, relative to audio data rates. Yes, there's latency involved in the use of digital wireless, but it's *way* down there in the noise - or I wouldn't be using it.

 

[Downunderwonder]

I wouldn't be so harsh. I didn't read it either, until now, as I'm a EE/SW engineer myself and understand these things reasonably well. Having read it, I should have commented on earlier. It needs to be understood that it's marketing literature for an audio cable company, whose sales are directly impacted by the use of digital technology. Like all good technical marketing material, it is factually correct. It just chooses its data points carefully. All the exposition on sound propagation and psychoacoustic effects of delay seem OK and legitimately helpful. But everything about the delays involved in digital audio transmission and manipulation is made-up. They don't mention A/D and D/A conversion delays, which are the things you can't easily finesse. They call out delays like "Digital Snake 3-7ms, Digital Console 1-3ms, Digital Signal Processor 5-10ms, Total 9-20ms" without saying where those numbers came from, or what they include. They are grossly inflated relative to the actual performance of products on the market. Roland specs 1.2ms with A/D and D/A included for their digital snake. If you're running a digital snake into a digital mixer, you may get to skip the snakes D/A and the mixer's A/D stage, which is a lot of the latency. DSP chips are freaking *fast* these days, relative to audio data rates. Yes, there's latency involved in the use of digital wireless, but it's *way* down there in the noise - or I wouldn't be using it.

thanks for the insight. Interestingly it was a wireless manufacturer's claim that an additional 15ms from his unit was inconsequential that got me looking into it. From the responses here it is clear to me that the extra 15ms is very important to some like me and quite liveable to unnoticeable for others.

 

[Sid Fang]

thanks for the insight. Interestingly it was a wireless manufacturer's claim that an additional 15ms from his unit was inconsequential that got me looking into it. From the responses here it is clear to me that the extra 15ms is very important to some like me and quite liveable to unnoticeable for others.

15ms isn't very perceptible in itself - I usually run my DAW with a 16ms latency for live monitoring - but it's a lot for a wireless unit to add. Line6 calls out 4ms for theirs.

 

[agedhorse]

Latency in digital devices (not including A/D/A) depends on the amount of processing required and the speed of DSP. Every signal path is then delayed back to the slowest total series of computational operation by adding bits of latency as required to the faster operations. This way, all signal elements arrive at the same time regardless of processing paths being used.