[sdiy] Analysis of frequency variation in analogue synths

[Eric Brombaugh]

ASSI wrote:
> On Freitag, 4. Mai 2007 23:42, Tom Wiltshire wrote:
>> This is a 32-bit shift register noise generator producing 8-bit
>> values, which are then fed into a simple averaging filter.
> 
> If you take more than a single bit per cycle out of an LFSR, then those 
> bits are highly correlated in time.  This transfers to the words you 
> build from those bits and the correlation creeps into the averaging you 
> perform as well.

True. What will that correlation look like on the output of the 
averager/filter? Given that the LFSR is 32-bits, how narrow would the 
filter have to be (how many cycles) before the effects of the 
correlation are negligible? I'd think anything > 32 cycles would 
approximate the same results you'd get by just clocking the LFSR 32x 
faster. Also, what would be the effects of scrambling the LFSR bits when 
forming the output word?

It's probably worth throwing together a few Matlab simulations to 
investigate. Are you aware of any publicly available papers on the subject?

Eric