From sbac@alum.berkeley.edu Sun Mar 30 00:09:42 2003
Subject:More latency tests with the Quattro

I found out that increasing the Quattro ASIO settings to 24 bit, from 16 bit, reduces the
latency. So I did that, and my Audiomulch now reports a latency of 3ms. Now, I tried testing
that by listening to the internal speakers on my keyboard and comparing it to when the sound
came out of the speakers attached to the Quattro, and even though I couldn't tell if there was
a lot of latency initially, I was suspicious. Even though I couldn't hear the latency from
playing a single note, I could definitely feel the latency when I played a fast rhythm, like a
boogie-woogie. It was almost like trying to run in water, you are trying to go fast, like on
the ground, but the viscosity of the water is holding you back. Well, that's my analogy
anyway.

Anyway, I was suspicious, so I did another technical test using a microphone and Sound Forge.
I put the internal sound on my keyboard to a saxophone, and the VST sound on Audiomulch to a
piano. I looked at the sound wave generated, and since you can see the two different
sinusoidal waves, I saw that the sax wave was going about 30 ms before the piano sound. So, I
actually think there is still a 30 ms latency. I noticed that I had placed the microphone
right next to my keyboard's internal speaker and the Quattro speaker was about 10 feet away.
I wonder if that matters. OK, the speed of sound is about 1130 feet/sec. So the time it
would take sound to travel 10 feet is 9 ms. Taking this into account, the latency is really
about 20 ms, which is still too much for comfortable playing. Now, my question to you experts
is, what can I do to get rid of this additional 20 ms latency. I know Deke is going to say to
remove my extra soundcard, but that is my last case scenario. I can't tell you how much I
love having the two soundcards, and being able to play along with my mp3s. I put on some
Booker T or Jimmy Smith and I can practice for hours. Pure bliss!

Thanks,
Aram