I want as little latency as possible, should I get a analog mixer?

http://www.presonus.com/news/articles/The-Truth-About-Digital-Audio-Latency

This explaination helped me understand some of it better. I think a lot of times the issue is not that the player is imagining the perception of latency, but there is just really a lot more latency than they think. They may be just calculating the asio buffer latency with sample rate and number of samples or see a latency measurement in the daw and think thats the entire end to end latency.
I am thankful for this dilema however. I gave up on my fast track pro trying to get tolerable latency and bought the axe

Axe into usb into computer is just a treat to play. I was blamed for being impossibly picky about latency in another forum. If there is any latency. In the axe, it truly is in that non perceivable range. Imho anyway
 
cakewalkgg said:
Axe into usb into computer is just a treat to play
Click to expand...

Actually, that would be a much better way to reduce latency. Or get a card with a digital input and direct monitoring.

Inserting a crappy mixer into the signal chain... Well, the distortion and frequency response problems may be actually within limits of human hearing.
 
About those 6 microseconds. They aren't about detecting latency at all. They are about detecting frequency as in f = 1/t. It does not mean one can hear latency this small. Here's the research the white paper refers to: http://boson.physics.sc.edu/~kunchu...olution-by-bandwidth-restriction--Kunchur.pdf

kgk said:
Can instruments detect the difference between "cheap" and "expensive" op amps for audio signals? If so, what is the difference?
Click to expand...

Oh yes, absolutely. The difference is in dynamic range, frequency response, harmonic and intermodulation distortion. It can be huge. Some of those things are hard to hear, some are outright annoying.

kgk said:
And can educated humans tell the difference in double-blind tests?
Click to expand...

At some level, anybody can hear those things. However, usual humans may have bigger problems such as crappy speakers and untreated rooms.

That said, distortion and coloration add up. The sound card is already crappy with 90% certainty, monitors are even crappier, so adding even more crap can tip the scale.
 
I must be missing something obvious, and if I am I apologize, but why would you not just skip the dig and just use the Axe FX as your interface? I jam to backing tracks from guitarbackingtrack.com all the time with my Axe FX as the sound card.

I HATE latency. It has made pretty much every other device (yes even the Kemper) unusable to me.

But with the Axe FX to me the latency is not perceptible, or atleast not any more then playing with my amp.
 
I play my Axe-Fx II and AX8 through a digital X32 Producer all the time at home.
No noticeable latency.
 
lqdsnddist said:
Its simply not possible to detect timing differences of a few samples or a few milliseconds. Extensive research within the field of psychoacoustics has shown the limits of temporal auditory resolution in humans to be around 10-20 milliseconds.

You'll certainly come across people on places like gear forums who will claim they can tell a difference between say 3 and 5 milliseconds but its physiologically not possible
Click to expand...

Well, you are right if he keeps the buffer at 128. But if he sets the playback buffer to 1024 because he's not feeling the latency anymore through the analog mixer he 's gonna be surprised when he plays it back. I am just suggesting that whatever the playback buffer size, just be aware of what its actually doing and how the analog mixer way of routing is going to effect it. And I can absolutely feel a difference between 256 and 128. 256 feels the tiniest bit spongier.
 
vangrieg said:
About those 6 microseconds. They aren't about detecting latency at all. They are about detecting frequency as in f = 1/t. It does not mean one can hear latency this small. Here's the research the white paper refers to: http://boson.physics.sc.edu/~kunchu...olution-by-bandwidth-restriction--Kunchur.pdf
Click to expand...
Although this is different than perceiving throughput latency isn't the interval in and of itself a measure of latency? Two events distinguished by temporal displacement.
 
Gearzilla said:
Although this is different than perceiving throughput latency isn't the interval in and of itself a measure of latency? Two events distinguished by temporal displacement.
Click to expand...

I only skimmed through the original article quickly and can't pretend to understand the science behind it, but as far as I understand, it doesn't mean that human auditory system can discern these two events as separate. What they did was basically playing back pulses in rapid succession, and introduced low pass filters to cut off higher harmonics (or very fast changes in sound pressure, in other words). Those pulses are perceived as continuous sound at a certain frequency, NOT AS SEPARATE PULSES. The shorter the gap between pulses, the higher the frequency. So yes, human auditory system can react to such speeds. But no, it cannot discern those tiny gaps. Moreover, the point of the experiment was to determine whether humans can react to ultrasonic frequencies, and what it allegedly proved was that yes, they can. That 6 microsecond gap means what, 167 kHz or so? The experiment says that when there's a signal that contains that frequency and a signal that doesn't, humans characterize the former as brighter or something, so yes, they do perceive it. But they don't "hear" it in the traditional sense.

And more importantly, let me reiterate. If a person is playing 1.5 meters away from a cab, he gets 5 milliseconds of latency. Yet nobody ever complains about it. But you can find tons of people who claim to hear this exact latency in digital equipment. This simply means that whatever they hear or not hear has little to do with latency itself.
 
vangrieg said:
About that 1 ms delay you're talking about. Sound travels at roughly 1 foot per second. I'd venture to guess you never listen to a real guitar cab yelling at you one foot away. So probably everything in the range of 10 ms is rather realistic and shouldn't create problems. Than again, Axe FX has latency of it's own. I don't know what it is exactly, but lets say it's somewhere under 3 ms. If that Digi of yours is really great, it may have about the same latency. But even if it's twice that, it shouldn't be a problem.
Click to expand...
Actually sound travels at 343.2 metres per second (which is 1126 ft/s) in dry air at 20°C. At these conditions, in 1.0 millesecond sound waves will travel a distance of 0.3432 metres or 1.126 feet.
 
Oh, and by the way. When people talk about latency while playing guitar, it's not just about the auditory system as such. We are talking about hitting a string with a pick and expecting to hear the resulting sound "immediately", whatever that means. So we have a lot of "subsystems" involved here, and a lot more things to process for the brain than just reacting to sound pressure level changes.

Just to put things in perspective, if you play 32nds at 180 bpm, that's like 24 notes per second, or about 41 millisecond intervals. Can someone truly claim that he can reliably discern like 10% timing inaccuracy at such speeds?
 
vangrieg said:
Actually, that would be a much better way to reduce latency. Or get a card with a digital input and direct monitoring.

Inserting a crappy mixer into the signal chain... Well, the distortion and frequency response problems may be actually within limits of human hearing.
Click to expand...

I can confirm that using a cheap analog mixer will diminish your experience. I have a cheap little Beringer 8-channel unit that I've been happy with for a few years. Then I got used to AxeFX II XL to CLR. Then I routed AxeFX and a few things to the mixer then to CLR. The sound quality was noticeably worse. At first I thought I had a lot of noise from my guitar or the axe patch, then I isolated the problem to the mixer.
 
vangrieg said:
True, but what's your point?
Click to expand...
The point was that the post I was responding to was incorrect as it quoted "...Sound travels at roughly 1 foot per second....." which it does not, at least not in dry air at 20°C as the figure is 1126 feet per second which is significantly faster than the incorrect value quoted. I was simply providing a better informed value of the speed of sound for these conditions. (I forgot to include a quote referring to the original post in my previous message).
 
6stringscott said:
I can confirm that using a cheap analog mixer will diminish your experience. I have a cheap little Beringer 8-channel unit that I've been happy with for a few years. Then I got used to AxeFX II XL to CLR. Then I routed AxeFX and a few things to the mixer then to CLR. The sound quality was noticeably worse. At first I thought I had a lot of noise from my guitar or the axe patch, then I isolated the problem to the mixer.
Click to expand...

I stay away from Behringer like I stay away from my abominable ex girlfriend: with a restraining order. Had a Bugera (company is owned by Behringer). It´s as unstable as a woman on her period.

I got a used SM Pro Audio pm8 Summing mixer. The thing was quite expensive new so it should be decent, and it can run passive so it wont colorate the sound in any way.
 
Most decent interfaces will have a way to directly monitor the analog inputs to the outputs, so latency shouldn't be an issue. I'm not familiar with the Digi stuff.
 
You must log in or register to reply here.
Back
Top Bottom