etxbrfa@kklm01.ericsson.se (Bjoern Fahller) (09/27/88)
OK guys, I really appriciate your funny answers about what signal processors
are for. That wasn't really what I wanted to get. It seems as the whole
world (except for someone at TEK whos name I've forgotten, has missunderstood
my question (thanks a lot)). The problem is not what kind of device to use.
The problem is the software, the maths behind. And the problem is not to
make a real time FFT (thats not very hard). The problem is the get the Envelope
of a sound's harmonics independent of its frequenzy.
No more quick answers please.
Bjoern
SNAIL: ERICSSON TELECOM Vox humana: (+46) 8 - 719 62 52
Bjoern Fahller
KK/ETX/TT/MLG
S-126 25 STOCKHOLM
SWEDENdtynan@sultra.UUCP (Der Tynan) (09/29/88)
In article <1853@erix.ericsson.se>, etxbrfa@kklm01.ericsson.se (Bjoern Fahller) writes: > The problem is the software, the maths behind. And the problem is not to > make a real time FFT (thats not very hard). [...] ^^^^^^^^^^^^^^^^^^^??? Ok, do it! come back in two years and tell us all about it!!! Anyway, I have a program written in C that performs FFT's (Singletons Algorithm, Radix-2, DIT) It goes up to 8K (A memory limit). It is fairly optimized, but basically simulates a piece of hardware, so it's not generic. If you want it, let me know. I'll mail you the source. Failing that of course, you could build your own 20MHz real-time box in a week or two, right? - Der -- Reply: dtynan@sultra.UUCP (Der Tynan @ Tynan Computers) {mips,pyramid}!sultra!dtynan Cast a cold eye on life, on death. Horseman, pass by... [WBY]