www.megadrum.info

Hi All,

Newcomer here who is looking very seriously at the MD for my A2E conversion. I apologize if this post is in the wrong board (software maybe is where it should go).

Obviously latency is a very big factor due to the haptic nature of playing the drums. If the latency is unbearable and with no return policy I'd want to do a little homework first to see if the MD will work for me.

Is there any way to simulate or measure (without owning the MD) the latency over a USB connection or is my only option here to look at what typical values others are measuring is? I ask because I assume this would be very PC dependent but maybe it's more hardware dependent on the MD side.

Any advice or anecdotal data points are helpful. Thanks.

PS - Bonus points question - any idea what I should budget for the positional sensing add on?

1. For my measurements with MegaDrum see viewtopic.php?f=3&t=2079 . I don't know how you can measure MegaDrum (or any other drum module) latency without owing it.
2. Most (60-80%) of latency is still due to sound synth even with best sound cards.
3. It's generally thought that any latency below 10ms is very hard to detect by human ear.
4. Acoustic drums also have latency - it takes around 3ms for sounds from drums to reach ears.

Positional sensing addon is 25GBP, see viewtopic.php?f=3&t=2052

Just to add to what Dmitri has already said, The latency on MD is negligable. There are other places in the chain, notably the sound card, where savings can be made.

airflamesred wrote:The latency on MD is negligable

+1

I never had problem with latency, even with relatively old PC. Playing on MD feels good. Your chain must have <10ms cumulative delay for good response which I have with xeon l5420 and integrated sound card. I think that latency WAS big problem for modules in the past, luckily now low latency is standard in this industry.

The best measurement in my opinion is take an SM57 or small condenser mic, and put that up to the trigger. Then take whatever makes the sound of the drums and plug that into the same set of inputs as the mic. Record and locate the impulse peek in both tracks. set locate movement units to samples. then count the samples between. convert samples to seconds based on your sample rate. in this way, both signals get the same input delay, and your are checking time from trigger through audio delivered.

Edit. Somethings to consider.
First off its not the peaks. it's the peak of the trigger mic'd sound, but the start of the sample playback audio. The difference is the samples sound design lag.
Second, 2 ms+-1 are a result of trigger settings for sensing time. Since that is known, it's not really a lag, just a setting.