Re: [oc] Beyond Transmeta...

[Lars Segerlund <lars.segerlund@comsys.se>]

  There are quite a lot of bit streaming techiniques in use, look at a 
delta-sigma , or the ancient MILDAP which was a 1 bit array processor 
using SIMD, I think bit streaming operations is most usefull for SIMD, ( 
IMHO ), since it's the multiply/divide algorithms that are really hard 
to do.

  For fast addition you can get O(1) addition of N bit's if you switch 
to tertiary base ( -1, 0 , 1 ) .. and the log N multiplication that 
follows on that. However, the conversion to binary from this format adds 
log N in time complexity ( carry propagation ).

  I think you bring up the real interesting thing, which is latency ! 
Which tends to complicate all bright ideas.

  / regards, Lars Segerlund.


Rudolf Usselmann wrote:
> On Mon, 2003-02-10 at 00:54, Jim Dempsey wrote:
> ...
> 
>>Does anyone get paid to think about this? Give me a call please.
> 
> 
> Last time I got paid for something (remotely) similar was a
> gig in 1995 for Synopsys. It was a true 256 bit processor that
> serialized operations over many "time slots" to work on very
> large vectors.
> 
> 
>>Jim Dempsey
> 
> 
> Cheers,
> rudi
> ------------------------------------------------
> www.asics.ws   - Solutions for your ASIC needs -
> FREE IP Cores  -->   http://www.asics.ws/  <---
> -----  ALL SPAM forwarded to: UCE@FTC.GOV  -----
> 
> 
> 
> 
> --
> To unsubscribe from cores mailing list please visit http://www.opencores.org/mailinglists.shtml
> 

--
To unsubscribe from cores mailing list please visit http://www.opencores.org/mailinglists.shtml