Hybrid Rig Setups!

[Sascha Franck]

I always think the acoustic latency comparison to digital latency is irrelevant really.
A few meters of acoustic latency is generally way less disruptive to playing feel compared to digital processing latency, and acoustic latency is what our ears are used to dealing with - it's natural.

AD/DA is like playing through a delay pedal on a short delay time that has no dry signal (fully wet), and the other is like playing in a room that has some nice natural ambience.

Or in other words: If you were to play guitar into an interface that was setup with a buffer of 512 or 1024 or something. it's basically like having a short delay pedal. That is never how it feels or sounds if standing several meters from an amp, of any kind.

I basically agree (and this is also why you can partially mask digital latency by adding some kinda ambience), but as usual, it totally depends. For instance, when playing open air, there's very little (if any at all) audible clues to "help you explain" the latency of your cab standing a few meters away. It's pretty much comparable to what digital latency is doing.

Besides, for almost aeons, people were never thinking about latency. Or does anyone remember any latency related discussions when it was about their Boss DD pedals (and whatever else) almost decades back? Most of them had no analog dry through, either.

As said, I'm pretty anal about latency, that's why I measure pretty much all of my guitar devices and interfaces carefully, but IMO, unless you're Steve Vai or Donald Fagen, there's always a certain point when you can just let go and accept things.

In general, it's however still clever to start with the lowest possible latency numbers as you may run into situations adding latency beyond your influence (especially live and even more so once there's some digital IEM solutions provided), which may as well tip you across a certain tolerance area (which, fwiw, is vastly different from person to person).

 

[Sascha Franck]

Even if I've been comfortable playing full digital or hybrid with digital in front and in the loop, therefore double AD/DA conversion, when I went back to full analog I really enjoyed the immediacy of latency absence.

Are you sure it was about latency?

 

[Frodebro]

As said, I'm pretty anal about latency, that's why I measure pretty much all of my guitar devices and interfaces carefully, but IMO, unless you're Steve Vai or Donald Fagen, there's always a certain point when you can just let go and accept things.

Let’s see how sensitive to latency Vai remains if he ever loses his Synergy endorsement gig.

 

[HotRats]

Are you sure it was about latency?

Yes, pretty sure it was latency because the almost imperceptible disconnection between picking and hearing has gone.

A small disconnection that to me feels like you have to dig more to get the note.
Sounds like nonsense, I agree, but that's what I experience.

As I've said, I have no problem playing with the usual latency I get from a digital unit in 4cm but I can't deny that it feels different.

 

[Sascha Franck]

Yes, pretty sure it was latency because the almost imperceptible disconnection between picking and hearing has gone.

It could really be something else. As in the attack simply being kinda washed by whatever digital algorithm.
Latency in the lower ms ranges (lets say <5ms) is pretty tough to decipher/pinpoint for most people, especially when not playing through headphones. And yes, I've done quite some tests with some mates in the past.
Please note: I'm not saying that no people are able to detect lower range latency differences. But very often it's something else. To find out, you'd need to compare 100% the same thing, the only difference being additional latency.

As I've said, I have no problem playing with the usual latency I get from a digital unit in 4cm but I can't deny that it feels different.

Depending on the device, 4CM often is fine to send you into "not acceptable territory".
Let's just assume that total latency values around 5ms are acceptable (most folks can easily get away with much more in case it's "masked", but let's not get into that). In 4CM (hence doubling latency, at least usually), this would already show plenty of modelers the way out (some prominent ones such as the Kemper, the Quad Cortex, the FM3 and what not being among them).

 

[Sascha Franck]

And fwiw, this is also why the GT-1000 is a kickass device for hybrid setups. Roundtrip latency of GT plus loop plus Amplifirebox in loop is still pretty much exactly the same as, say, just an FM3 on its own.

 

[HotRats]

It could really be something else. As in the attack simply being kinda washed by whatever digital algorithm.
Latency in the lower ms ranges (lets say <5ms) is pretty tough to decipher/pinpoint for most people, especially when not playing through headphones. And yes, I've done quite some tests with some mates in the past.
Please note: I'm not saying that no people are able to detect lower range latency differences. But very often it's something else. To find out, you'd need to compare 100% the same thing, the only difference being additional latency.



Depending on the device, 4CM often is fine to send you into "not acceptable territory".
Let's just assume that total latency values around 5ms are acceptable (most folks can easily get away with much more in case it's "masked", but let's not get into that). In 4CM (hence doubling latency, at least usually), this would already show plenty of modelers the way out (some prominent ones such as the Kemper, the Quad Cortex, the FM3 and what not being among them).

Yes could be somenthing else, I can just make an educated guess based on my experience.

In the end I've not said the change in feeling I get when doing 4cm with the Helix is not acceptable, I've only said that full analog feels better. Assuming the difference is due to latency, no latency feels better than latency.

In my Helix presets for 4cm default sounds (basic clean and basic dirt) have no blocks in the way other than eq, a small amount of reverb and a very light and slowly closing noise gate for the dirt. Turning gate off it makes no difference.

Don't know what the total latency is and I don't really care because, again, I can play fine with the Helix in 4cm but I can't say there's no difference.

 

[Sascha Franck]

Don't know what the total latency

Fwiw, total latency of the Floor with both paths and a loop used should kick in at around 5-6ms. A single path with no loops should be around 2ms (like the Stomp), dual serial paths add some more (somewhat around 1ms, IIRC) and a loop adds another 2ms. So in case you're using this with an am/cab around 2-3 meters away, it'll add another 2 "virtual meters". Which could be good enough to go beyond whatever might be "acceptance level". In fact, a cab 5 meters away is already qualifying for a noticeable less direct impact.

 

[northern_fox]

Here's something weird - sometimes I play better with extra latency, because of the way it separates/shifts my focus from being on the immediate auditory feeedback for reassurance that the notes are correct to being more on just what I'm actually playing with my fingers. Then I hear the sound a bit later on as a retro "verification" it was correct. I don't necessarily prefer it, but it's something I've noticed switching a lot.

 

[HotRats]

Fwiw, total latency of the Floor with both paths and a loop used should kick in at around 5-6ms. A single path with no loops should be around 2ms (like the Stomp), dual serial paths add some more (somewhat around 1ms, IIRC) and a loop adds another 2ms. So in case you're using this with an am/cab around 2-3 meters away, it'll add another 2 "virtual meters". Which could be good enough to go beyond whatever might be "acceptance level". In fact, a cab 5 meters away is already qualifying for a noticeable less direct impact.

And that's exactly my case.

 

[DrewJD82]

Going 4CM with my FM9/EVH 50-watt provides no latency I can detect, nor does it create any additional noise. Fortunately, it does the opposite via noise gate on that Red channel which would drive me nuts at gig/band volume.

I’m most sensitive to latency when playing fast, I can’t detect a thing. Conversely, playing through my Shiva with no boost feels like running through mud in comparison when it comes to playing fast and there’s no AD/DA going on there!

 

[Orvillain]

I don't notice latency in 4-cable-method with the Axe FX III or the VP4. I do notice it with the Helix Floor though. I provided measurements in the past. The on paper differences are not huge, but the practical realities mean that Axe FX III just does feel quicker to play through.

 

[Sascha Franck]

Conversely, playing through my Shiva with no boost feels like running through mud in comparison when it comes to playing fast and there’s no AD/DA going on there!

The on paper differences are not huge, but the practical realities mean that Axe FX III just does feel quicker to play through.

This is a) why I said there's plenty of other factors and b) exactly my experience as well.
I mean, "smeared" transients are a thing with some stuff, regardless whether it's analog or digital. And once they kick in, it's pretty obvious that you will feel less of an impact following your picking. It's possibly the same reason for some amps to act more forgiving while others don't.

If you really want to know your latency tolerance, there's no way to find out than to set up a proper test, read: Apart from additional latency, all other things must be equal. Switching from one amp in (digital) 4CM method to the same amp without the digital device isn't sufficient, unless you've actually measured that the digital device is causing *zero* other side effects.

It's also astonishing how wildly different oeople perceive and deal with latency (pretty tough to tell, though).

And having said that: In case you can't detect latency, it's defenitely not necessarily a sign of inaccurate timing feel or anything.
Sure, there's been folks telling me there'd be no latency when they were playing early amp sims through early versions of iRigs and early iPads - and you couldn't get the latency below 15-17ms back then (I checked). This is how I learned to take any "no latency" statements with huge grains of salt.
However, in general, latency and how it may (or may not) impact you is pretty much a "reference" thing, especially when dealing with guitars. It's likely not as easy, but simply put: If you always rely on the feel of your pick/fingers hitting the strings, latency may not impact you too much (unless you're recording things, but that's yet another story...). If you however rely on the acoustic information provided by whatever monitoring sources, you may get annoyed by latency a lot earlier.
But as said, that's just a *very* simple explanation (if at all) because in reality, things may shift between these two quite a bit, depending on the situation, actual amount of latency, guitar used, sound used and what not. And it even seems to greatly vary based on your personal shape, feel, "did-I-fart-well-enough" and what not. At least that's what I observed with myself.

 

[HotRats]

In my experience there's a difference between latency you can clearly detect as a delay between your action and your hearing (that one starts probably a bit below 10ms) and latency you can't clearly detect, meaning you can't hear the gap but you feel is there.

I know, I know, this is snakeoil bullshit but that's my experience.

 

[laxu]

In my experience there's a difference between latency you can clearly detect as a delay between your action and your hearing (that one starts probably a bit below 10ms) and latency you can't clearly detect, meaning you can't hear the gap but you feel is there.

I know, I know, this is snakeoil bullshit but that's my experience.

That's pretty much how I perceive it.

I think that "felt" latency can become a problem if you are playing e.g very fast stuff. I find my precision can be thrown off if the latency gets too high, and I'm certainly no Steve Vai.

The max latency of my pedalboard rig is about 3ms if all my Strymons in front of the amp are on at the same time. That's firmly in the "I don't care" territory. Double that to say 6ms and we are in "I can tell, but I got something in return". Above that? Then it can be a problem especially for fast leads.

Overall I just don't give a damn about things like AD/DA conversion stuff. We are long past that becoming any sort of issue with good gear as long as levels are set correctly.

 

[JiveTurkey]

This is what I will perhaps change back to one day, but with the GT-1000 it does work quite fine with digital-analog-dogital for me, at least so far.
Not gonna lie, I have tried it without the pre-loop-digital stuff and some of the analog dirt boxes do feel sort of noticeably different, some more than others. But I think it's not always better, even if they may not sound 100% as intended anymore.
Whatever, atm practicability dictates doing it as I am for the time being. To get to the same level of comfort, I needed an additional remote controlled loop switcher, an additional wah, a compressor and a dirt box - and these are only the things to get started, as I also like to use some modulation FX of the GT pre-dirt. As a result, I'd likely also need a new pedalboard or get rid of the GT and find something else for my post-FX duties (I could use the Stomp for that). In the end, it'd be pretty much like my old setup, but I would never go back to that but needed a pimped up version.

I'm defenitely having an eye on the Tone King Imperial pedal and should I ever get one and like it, it might be worth rethinking my current approach and build a new setup around that one. Which would then be strictly analog until the pedal is hit.

Honestly; the only modeler I liked putting analog drive pedals in front of was the Helix stuff. Never tried it with Kemper because I liked the M style stuff and whatever oddball "here's a tubescreamer and a bunch of eq blocks" thing you could do. The FAS stuff amp tones and boost options are spectacular so no need out of the box there and Boss was fine because I knew what I was getting into and the GT1000 form factor is perfection for that era of boxes, imo.

I agree in principle, but there are a few notable exceptions. The Strymon overdrives are digital, but are absolutely worth it in front of the amp. Also the Digitech whammy and drop pedals are all-digital and don't hurt your tone in front of an amp. The Boss OD200 is also excellent in front of the amp.

I don't disagree with this at all. At least on the whammy part? Always loved those. To a fault in some cases probably Never tried anything Strymon.

 

[Sascha Franck]

Honestly; the only modeler I liked putting analog drive pedals in front of was the Helix stuff.

The HX units indeed take pedals like champs. But the Amplifirebox' Bman and Deluxe are absolutely great, too, at least once you tweak the internal EQ - defenitely needs a good amount low cut in front and some low end compensation after the amp, fortunately the internal EQ provides both.

Fwiw, re: Whammy pedals. V2 has an incredibly great working buffer. A mate of mine even soldered it out when he didn't need the Whammy itself anymore and slapped it into a little box.

 

[la szum]

In my experience there's a difference between latency you can clearly detect as a delay between your action and your hearing (that one starts probably a bit below 10ms) and latency you can't clearly detect, meaning you can't hear the gap but you feel is there.

I know, I know, this is snakeoil bullshit but that's my experience.

I agree. When we play it is pretty much only experiential, not theoretical.... and I certainly
don't contemplate graphs and measurements when playing.

 

[TSJMajesty]

If I have enough latency, I will break strings.

Not with digital gear, but for the occasional, playing-outside-far-away-from-my-amps-using-a-wireless, (it's fucking awesome! I have these woods behind me, and the guitar reverberates through the trees!) I tend to dig in to compensate, knowing full well what the issue is.