1. Please take 30 seconds to register your free account to remove most ads, post topics, make friends, earn reward points at our store, and more!  

How to generate latency in live situation?

Discussion in 'Recording Gear and Equipment [BG]' started by Clemouze, Apr 8, 2019.


  1. Clemouze

    Clemouze

    Sep 1, 2016
    Paris - France
    Salut all

    I am using a MOOER Radar pedal out of the line out of my amp head.
    Super cool bass cab sim tone (taken from my own cab )

    I like using it to play with headphones, to record or in live situation to go to FOH.
    BUT
    In the two last cases I also like to take the DI pre-EQ from my amp.
    BUT
    The MOOER has a 0,7 ms latency
    Almost nothing !
    But when taking the DI at the same time happens a slight out of phasing.
    When it is for recording, it's simple to correct, delaying the DI track from 0,7 ms.
    But what could I do for live setting ?
    Is there a way to generate 0,7ms ?
    A pedal ? A rack thing ? A delay thing ?
     
  2. silky smoove

    silky smoove Supporting Member

    May 19, 2004
    Seattle, WA
    Latency of 0.7ms in a live setting is meaningless. It's far too low to be perceptible by humans except under the most scrutinizing of situations, which live is most definitely not. Don't sweat it. No need to purchase any additional gear.
     
  3. monsterthompson

    monsterthompson The Eighth Note Wonder Of The World Supporting Member

    Nov 25, 2008
    Hollywood
    Could you put a second of the same pedal in the signal path with a “blank” IR so it has the same delay time but doesn’t alter the tone?

    Just a thought. As mentioned above, it may not be a significant real world issue. I’m unfamiliar with the device and the application so I’m not sure how big an issue this is or isn’t.
     
  4. micguy

    micguy

    May 17, 2011
    There are a lot of ways you can make a delay - at least electronically (I've desiged a few). The odd thing is that digital delays start at a couple of milliseconds - the A/D and D/A conversion tends to take a millisecond or two.

    .7 milliseconds sounds...odd. That's too much for it to be a typical analog related delay, and not enough to be a digital one - are you sure about that number?
     
  5. silky smoove

    silky smoove Supporting Member

    May 19, 2004
    Seattle, WA
    Given that 0.7ms would be an imperceptible amount of latency to humans, I'm feeling pretty confident in my guess that it's a complete non-issue :smug:

    Sound travels roughly three meters in one millisecond (not accounting for a number of factors that will change this). So we're talking about 2.1 meters at the speed of sound here. Play a sound while another identical sound plays at the same time but placed 2.1 meters away and you'll have a crude model of this latency. You won't hear any "flamming" of the two sounds. They will appear to your brain as the same sound.

    There are some other issues related to latency beyond the "flamming" issue, such as changes to frequency response through comb filtering, or perceived widening of a stereo field through the Haas effect, but I don't think that's what's concerning the OP.
     
  6. silky smoove

    silky smoove Supporting Member

    May 19, 2004
    Seattle, WA
    Modern Thunderbolt 2 interfaces are claiming sub-1ms latency in their specs. How accurate that is I can't say, but they're certainly claiming it. This isn't that type of product, but my point is that digital appears, on paper, to be in that range in the modern world.
     
  7. micguy

    micguy

    May 17, 2011
    If it's in one signal path, yes that's very short (oddly short, in that I don't know how they'd do that). But, if you are combining the path with the short delay with a path without that delay, it'll give you a comb filter than will be VERY audible - .7 milliseconds is half a cycle delay at about 700 Hz At that frequency, you'll have a complete null if the two paths have equal amplitude. You'll also have nulls at odd (3, 5, 7, etc.) multiples of that frequency.

    But, I'll bet it's latency is longer than that (see my posts about A/D/s and D/A's), which makes the situation worse - the comb filter starts lower in frequency. The real solution is probably to avoid combining the signals if at all possible.
     
    Last edited: Apr 8, 2019
    PlatoFunFactory and Wasnex like this.
  8. micguy

    micguy

    May 17, 2011
    The network interface might have sub 1 millisecond latency, but that's how long it takes to move data between digital devices, not how long it takes for audio to get in, through, and out of the system. That Mooer product has an analog input and output and samples at 44.1 kHz. That means there's an A/D and a D/A in the audio signal path, and with that low sample frequency, I'd be willing to be the signal latency of those two conversions added together is well above .7 milliseconds. If the processing has any latency (which is likely), that'd be tacked on as well.
     
  9. Clemouze

    Clemouze

    Sep 1, 2016
    Paris - France
    Yes I know it is not audible .
    The issue is when mixing this 0,7 ms latenced signal with the DI signal . It creates a sliiiiiight out of phasing !
     
  10. Clemouze

    Clemouze

    Sep 1, 2016
    Paris - France
    Ok I can't find where those 0.7 ms where from.. semms very very little indeed

    Well it is not very important since the issue will be the same whatever latency it is : out of phasing.
     
  11. Clemouze

    Clemouze

    Sep 1, 2016
    Paris - France
    Appears to be less than 2 ms : ( From MOOER website ) Screenshot_20190408-230159_Chrome.jpg
     
    40Hz likes this.
  12. Clemouze

    Clemouze

    Sep 1, 2016
    Paris - France
    Currently my solution is to... not use my DI signal blended with the cab sim'ed signal.
    Otherwise the out of phasing is causing a slight low end drop, you know it

    Hmrf
     
  13. silky smoove

    silky smoove Supporting Member

    May 19, 2004
    Seattle, WA
    Agreed. I've never really believed those numbers, but they are being published by multiple manufacturers, so either they're all getting away with fudged numbers or there's some semblance of truth to it, even if it's not quite as low as advertised.

    I mentioned this in one of my posts as well. I disagree that it will be VERY audible in a live context. Maybe in a "live" A/B comparison of the signal with and without the comb filtering and nothing else going on in the mix. With a full band in a live setting? No way. Nothing that the soundguy wouldn't already address and make a non-issue.

    Absolutely agree

    Yes, a slightly out of phase signal (assuming the 0.7ms is accurate) with another that will be imperceptible from the standpoint of distinguishing one from another, and one that, in the mix, will have negligible or at least easily handled EQ impacts. Nothing to worry about at all from a practical perspective.
     
    Clemouze likes this.
  14. sawzalot

    sawzalot Supporting Member

    Oct 18, 2007
    I get it, when you mix a direct signal (pre-eq from your head) with a signal with a .7ms delay, you hear the phasing in the upper frequencies. If you weren't mixing in the pre-eq signal it wouldn't be an issue. The MOOR is a digital pedal so the .7ms is almost certainly the time to do the A/D and D/A conversion.

    Basically you have two choices: 1) don't mix in the signal from your amp head. You'll never hear the 0.7ms delay but you won't get the dry sound mixed in 2) somehow incur that delay on your clean signal: use another MOOR Radar on your pre-eq signal with a setup that doesn't do anything to your signal except incur the A/D and D/A latency, use a digital mixer and delay the channel with the direct signal from your head by about 0.7 ms, find a digital delay that will do the same thing, etc.
     
    Clemouze and D.A.R.K. like this.
  15. geeza

    geeza

    Mar 15, 2009
    Earth, but just barely.
    I'm not wearing pants
    Stand .7ms further away from the amp
     
  16. thumbknuckle

    thumbknuckle In Memoriam

    May 23, 2012
    Westfield, MA
    That's 16 sample frames each way at 44.1K, a super common figure. The dbx Driverack units and most digital mixing consoles are about the same.

    Mult a signal out and record it straight in to your interface and through the unit in question and compare. There is nothing mysterious going on.
     
  17. micguy

    micguy

    May 17, 2011
    Latency quoted from the dbx Driverack 360 specs:

    Analog input to output: 2.57 ms (48 kHz) - yes, that's more than .7 msec

    Latency would be more for a comparable unit run at 44.1 kHz (likely 3 msec-ish). Consoles that run less latency are typically running 96kHz sampling, which reduces latency in the A/D's and D/A's. Latency of .7 msec for a unit running 44.1 kHz is odd - I doubt that's signal input to output. A latency of .7 msec (16 samples) for processing in the digital domain I can see, but not input to output, which is what the user here is fighting.
     
    Last edited: Apr 8, 2019
  18. thumbknuckle

    thumbknuckle In Memoriam

    May 23, 2012
    Westfield, MA
    I've never used that unit. The driverack PX is 0.8.

    Behringer XR18 is 0.85
    A&H Qu series is 0.7.
     
  19. MattZilla

    MattZilla

    Jun 26, 2013
    CNY
    by...


    ...doing something other than turning down one of the two bass guitar signals?

    If they're a legit Sound Engineer with full equipment then they're already applying some amount of delay uniformly on the whole band, yeah?
     
  20. micguy

    micguy

    May 17, 2011
    Can't find spec on the driverack PX, other than it's using a rather interesting conversion system. The A&H is .7 from input to AES (digital) output - at 48 kHz - that's only one conversion. From analog input to analog output (adding in the d/A converter) its 1.2 msec. which is pretty good at that sampling rate - assuming they haven't short changed the filtering.
     
  21. Primary

    Primary TB Assistant

    Here are some related products that TB members are talking about. Clicking on a product will take you to TB’s partner, Primary, where you can find links to TB discussions about these products.

     
    Apr 23, 2021

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.