Making It Make Sense
Another factor is lack of context. Sure, high bit and sample rates make a difference – to a point. We definitely want to use the highest sample rate possible, within reason, and the largest bit depth.
Around the year 2000, I was involved with a demo in the Washington DC area utilizing high-end converters and a Sennheiser MKH800 microphone with a bandwidth specification up to 50 kHz.
The goal was to expose a roomful of engineers to the differences of sample rate within a controlled environment. We were able to compare a live mic feed, 16 bit/44.1 kHz sampling (the Red Book CD standard), and 24 bit/96 kHz sampling. Most people in the room were confident that they could hear the difference.
Interestingly, as good as the 24/96 sampling proved to be, it still wasn’t up to par with the live mic feed. So, what does this all mean? Was the 16/44.1 sampling terrible? Certainly not. It actually sounded very good, with only subtle differences between it and the other formats.
Perhaps the highs were slightly “grainy” or the dynamic gradations weren’t as smooth. Maybe the ambient detail was not quite as smoothly incorporated into the direct sound. But would these kinds of differences be evident in a PA situation?
In addition, there are questions of workflow, cost, and potential technical downsides. One way to look at it is that there’s probably a “lowest common denominator” somewhere in the system. Perhaps it’s a mixing console, a drive rack, or an outboard A/D converter with a native conversion rate of “only” 24 bits and 48 kHz. Does it make sense to run everything else at 24/96 or even higher?
Not to me it doesn’t. Any connections between equipment at different sample rates necessitates sample rate conversion, with the tradeoff of additional latency. Besides, what’s the point? If the signal ends up as 24/48, then any advantages of running any part of the system at rates higher than that are subtle at best, and certainly lost in the overall picture.
Back To Basics
I’ve previously expressed the view that we need to focus on getting the midrange right before worrying about the extremes. I’d like to go even further by adding that in terms of sound reinforcement systems, we need to first get the basics right, such as power, grounding, gain structure and sample rate consistency. By doing so, the rest of the job gets much easier.
Rather than focusing on how the latest gear has FPGAs running at PetaFLOP speeds, let’s put our efforts toward getting the most out of the gear available. As has probably been said thousands of times in this business, if we can’t get it right with staple consoles, microphones and loudspeakers, should we really be calling ourselves professional live sound engineers?
Karl Winkler serves as vice president of sales/service at Lectrosonics and has worked in professional audio for more than 25 years.