home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!kithrup!stanford.edu!agate!biosci!uwm.edu!linac!pacific.mps.ohio-state.edu!cis.ohio-state.edu!news.sei.cmu.edu!fs7.ece.cmu.edu!crabapple.srv.cs.cmu.edu!andrew.cmu.edu!cc4b+
- From: cc4b+@andrew.cmu.edu (Christopher Brian Cox)
- Newsgroups: comp.sys.mac.programmer
- Subject: Re: FAST fixed point math?
- Message-ID: <Qf2aTW600Vpc4_Vl1D@andrew.cmu.edu>
- Date: 18 Nov 92 15:57:22 GMT
- Organization: Senior, Physics, Carnegie Mellon, Pittsburgh, PA
- Lines: 19
-
- (mail to cs.oberlin.edu bounced for 4 days, so I'll post this to the bboard)
-
- response to mhall@mark1.cs.oberlin.edu message:
-
- If the samples are all -127..128 , then do the whole thing in integer math.
- Also, with this small a range you could do the multiplies with look up tables.
-
- preinit:
- Q[i] = r*a*[-127..128]
- R[i] = r*b*[-127..128]
-
- in loop:
- xTemp = Q[x] - R[y]
- y = R[x] + Q[y]
-
- Also, unroll the loops as much as possible (if you are doing a fixed
- number of iterations (ie:fft) then you can unroll it exactly.
-
- Chris
-