|
Forum Index | Read LUG: Policy/Rules Messages Threads Digests | Post New Message | Search!
--- In logic-users@yahoogroups.com, Maurits van de Kamp <maurits@...>
wrote:
>
>
> On Jan 30, 2006, at 8:34 PM, rOLf peCHUkas wrote:
>
> > there is an ALGORITHM involved
>
> Yes, x=a+b
I believe that you are incorrect here. It is not aquestion of just addition,
at least not in
floating point calculations. Please see post no. 7 on this thread:
http://www.3daudioinc.com/3db/showthread.php?tY10
> Have you never wondered why .. nowhere in ANY daw can you choose a
> "summing algorithm"? It's because there is only one.
The link above refutes this. Also, you can choose different algorithms by
using a different
DAW. Tracktion has 64 bit in the box summing. Maybe there is a difference.
Maybe we
should have the option of 80 bit offline summing in Logic, if the Metric
Halo is really so
close to a real SSL as some have suggested.
>
> Funnily enough, Pro Tools summing has been proven to be identical to
> Logic summing in the DAWSum project.
>
This test was made in 2001 I think, if we are looking at the same board.
Logic 5 came out
in the middle of that year and offered 24/96 for the first time. I can't see
which version the
tests are done with, but if 16 bit then yes it would be exact, but if 24 bit
I think it would
not. I could be wrong here as I am not a mathematician, but the link above
delves into the
math as does say that floating point summing is NOT just addition.
C
Forum Index | Read LUG: Policy/Rules Messages Threads Digests | Post New Message | Search! © 1994-2008, All Rights Reserved. |