"Rusty Boudreaux" wrote in message ...
However, your argument using general processors is not
valid. In the realm of DSPs and ASICs this level of processing
power is certainly achievable at modest cost.

I disagree. If we're using brute-force FIR techniques, name me an
affordable computation system that can achieve 16 billion 32-bit
integer MACs per second. That's for a two-channel, 96 kHz system for
24-bit audio, and assumes 32 bits are enough accumulation precision
for such a long filter. The newly announced TI C6412 claims to do 2
billion 16-bit MACs per second, so if we generously assume things
scale linearly, you'd need at least 16 of them to achieve the required
computational throughput. They're about $40 in quantity, so parts
cost for the CPU alone is $40*16=$640, which translates into
approximately 6*$640=$3840 MSRP difference, not counting support
circuitry, the case, power supply, R&D time, etc. to support the extra
CPUs. Now if we want to do it for a typical home theatre system with
6 or 8 channels, multiply that by 3 or 4. I don't think that's a
modest cost.

One solution, as Denis points out, is smarter algorithms that require
less computation load. Optimized FIRs, or decimation are two ways to
getting there. The Lake DSP stuff was what I was thinking about
actually for reducing latency.

--Andre