Latency between patch switching

[isdirg1]

Has anyone had an issue with latency? I ask because I have recently noticed that on a few (not one!) patches the sound seems to "fade in" to full volume over a period of 2-3secs. This is driving me nuts.

Has anyone else noticed this and if so how have you fixed it?

[Joepeggio]

I notice that too. Not the XX ms patch change but a very slow (2-3 sec like OP states) for the volume to ramp up. It seems maybe associated with compressors or lots of modulattion in the patch??? Because some are loud right away.

[sammycraig]

Certain patches using more DSP take a second or two to load up whereas patches using less DSP load quicker. An unfortunate occupational hazard of DSP hungry patches. 

[phil_m]

Are you connected to the editor while seeing these really long transition times. Even with complicated patches, the longest gap should be in the 150ms range, definitely not 2-3 seconds. If you're changing patches while using the editor, however, I'd say all bets are off as far as patch change latency.

[RobBon]

If you have placed the LA studio comp after the amp, move it to the front of the amp, (it will sound different) then switch patches, you will find that the volume will be instant.

I use the Deluxe comp after the amp, with a little tweaking, achieving a similar result to the LA comp. It has no volume issues when switching in the patch. Hopefully this works for you.  :D

 

[klangmaler]

There are several threads dealing with this, for example:

http://line6.com/support/topic/17047-noticeable-delay-when-switching-patches/?hl=latency

[Joepeggio]

Maybe the wording is off. It's not patch change latency. The patch changes very quickly.

 

I'll have to experiment more to explain better. After the patch changes and sound is coming through, the volume takes a few seconds to rise up. When I have the time to test this, I will try to replicate it and share the patches.

[FullMotion]

I have noticed the volume ramp up when changing patches. I haven't managed to figure out what causes most of my patches to do it.

[klangmaler]

...maybe you have one ore more compressor blocks active?!

[silverhead]

Look for a compressor block that has an Attack parameter. This defines the length of time from gate-opening to full volume; essentially the 'fade-in' time for the sound. A high value for this parameter explains the symptom.

 

For instance, with the Deluxe Comp FX, an Attack time of 200ms and a Mix of 100% means no dry signal gets through, and the compressed signal will fade-in over a period of 2 seconds.

[duncann]

Look for a compressor block that has an Attack parameter. This defines the length of time from gate-opening to full volume; essentially the 'fade-in' time for the sound. A high value for this parameter explains the symptom.

 

For instance, with the Deluxe Comp FX, an Attack time of 200ms and a Mix of 100% means no dry signal gets through, and the compressed signal will fade-in over a period of 2 seconds.

 

This doesn't sound right to me. There appears to be varied and nuanced definitions of what the attack on a compressor does, but I've never come across one that can basically fade-in the signal over time. I would actually love this, because then the need for a auto-volume would not be needed. :)  I could very well be wrong, but my understanding is something along the lines of the amount of time that passes before the compressor starts compressing when the threshold is passed.

 

I have noticed a fade-in and -out when switching presets, but nothing on the scale of seconds. I always thought it was there to suppress sudden volume spikes when switching, and generally making things sound smoother.

 

Is there any external equipment, either real or virtual, that could be causing this?

[Joepeggio]

after further testing it is associated with compressors.  Mostly with the LA Studio Comp, a little with the Red Squeeze, and no issue with the Deluxe Comp (even with attack all the way up).

 

May be how those models act in the real world, and since the Deluxe comp is a Line 6 creation, no issues.

[zooey]

Look for a compressor block that has an Attack parameter. This defines the length of time from gate-opening to full volume; essentially the 'fade-in' time for the sound. A high value for this parameter explains the symptom.

 

For instance, with the Deluxe Comp FX, an Attack time of 200ms and a Mix of 100% means no dry signal gets through, and the compressed signal will fade-in over a period of 2 seconds.

Sorry, but this is definitely not right.

 

First off compressor attack time controls how long it takes for volume to be REDUCED when signal above its threshold comes in. Longer attack time means more of a peak gets through before the compressor clamps down. Sometimes that helps preserve attack transients in a good way, sometimes it's just ugly.

 

Second, 200ms is 200 milliseconds, 2/10ths of a second, not 2 seconds.

[njglover]

It may instead depend on the RELEASE setting of the compressor in question. In this case, the release controls how long after the compressor engages to wait before releasing the compression. In effect, how long to wait before going BACK to loud. So that could explain going from quiet to loud.

[jeremyn]

It sounds like a compressor is being 'kicked' by a huge input spike in the amp model when it is switched. The attack time will affect how small a 'glitch' will cause the compressor to compress, so if a smaller number here has a more noticeable effect, then the 'glitch' is very tiny. And obviously the release time will affect how long it takes before the compressor fades back to full gain.

 

This is something worth trying to replicate, as it may be an issue with some amp models or effect blocks that glitch after switching.

 

If the Helix mutes the output for a short time after switching, then you don't necessarily hear the glitches. But, a downstream block like a delay, reverb or a compressor that moves the glitch forward in time will end up with an audible affect beyond the output mute time. Try a delay set to 1 second instead of a compressor in the same location and see if you can hear a click or pop. That said, it is possible that it is something that the Line 6 engineers have hidden with a simple muting technique that has been applied to not only the output, but also to some subset of input blocks like delays and reverbs. It is also possible that some compressor models have a small bug in the signal or side chain where they start with a DC offset that hasn't been set to the quiescent DC level when initialising the model.

[silverhead]

Right - my interpretation of 200ms was off by an order of magnitude - don't know what I was thinking there.

 

But as for the fade-in behaviour of the attack parameter, I took that from wikipedia - unless I misinterpreted the description there, and that interpretation was done at the same time that I thought 200ms was 2 seconds.... so perhaps that should be taken with a grain of salt! :lol: :P

 

All in all, seems like a big fail on that one!

[isdirg1]

Are you connected to the editor while seeing these really long transition times. Even with complicated patches, the longest gap should be in the 150ms range, definitely not 2-3 seconds. If you're changing patches while using the editor, however, I'd say all bets are off as far as patch change latency.

No, this problem occurs when playing live. It is particularly noticeable when switching from my "Joe Walsh" tone to clean when playing "Lifes been good". I tried to solve the issue by having the clean and crunch on one preset but ran out of dsp so I have to use two presets.

[Perfect]

Try switching between two clean patches that both have chorus - gives a very weird sound as the second patch kicks in.