Benchmarks

[Lakedaemon]

I see....

Well...then this renew my interest in seeing what kind of wave is produced by a piezo transducer...
(maybee I should try to record some signals).
As I have been wondering for some time now how Roland does it's positional detection trigering,
it would help me understanding that part.

Would it be safe to plug a stereo jack of a pad in a line entry of a soundcard ?
does it need amplification (what are the specs of the line level ?)

If it is, It would be very easy for me to record waves of different piezo signals (rim and meshhead piezo...).

[Marctwo]

I put mine straight into the line input. You're best to put a pot before the input so you can adjust the level otherwise you'll most likely get a lot of clipping.

I don't know how Roland does position sensing, but it should be possible to achieve by comparing the velocity ratio between the head/rim triggers.

[elrules]

Roland positional sensing can be achieved even with mono mesh pads. It is only needed a mesh pad with a piezo in the center. The frequency is different if you hit the pad near the rim or near the center. PS does not depends on velocity.

[Marctwo]

The frequency is different if you hit the pad near the rim or near the center...

Ah, yes... of course. Cheers.

[baguette]

I don't really understand why you bother that much about latency - In my opinion it does not make any difference if there's 3, 6 or even 10ms latency.
Sound travels at about 330m/sec (3msec/meter), which means that if you use some monitor box instead of headphones you will get say 6ms extra latency.
The latency of an acoustic drumset is 3ms between the drum head and your ears, maybe 4 if you are a tall guy? (->this is all a little nonsense )

[baguette]

Somehow, it contradicts the 2-6 khz vibrating frequency (of the piezo specs...maybee I misunderstood that)

Usually these piezo elements are supposed to work the other way round, i.e. you apply some frequency and they make sound (i am sure you have at least one in your house inside a digital alarm clock).
I think the frequency in the specs is just the frequency you should apply to make them work the best.

[Marctwo]

I think the freq quoted for piezos is probablly the resonant freq. Which will have an influence on both input and output.

Latency is or isn't a big deal depending on how you work with it. I approached this problem as a programmer, not a drummer/engineer. On examining the waves, it was clear to me that delays larger than ~0.5ms were not necessary. No big deal, just an instinctive conclusion.

For me as a drummer, latency below 10-15ms is not really noticable but a lot of sound cards may struggle to achieve this, even using Asio4All. In this case, latency becomes a big issue.

Of course, if you're desperate you can always chop the front off your samples.

[Lakedaemon]

Well....it is quite hard to say at which point having too much latency gets annoying : you decide !
125 ms is an 16th note at tempo 120
62ms is an 16th note at tempo 240

I personally feel than 10ms is very good
And 3-6 ms is really excellent (especially since it can't be avoided).

Now, when you do recording/mixing/mastering work....
you want the best for your songs and so, it is natural to look for the less latency...
With less latency, your various tracks have more chance to be in sync when you overdub.

Besides, a byproduct of having low latency is that...you can increase latency and get some cpu overhead to add more effects, etc....
(Well...this reasoning only work for a soundcard ^_^)