Re: AutoArk

[fadden@uts.amdahl.com (Andy McFadden)]

In article <1993Jan11.020733.25905@nuscc.nus.sg> ltchean@iss.nus.sg (Lim Thye Chean) writes:
>fadden@uts.amdahl.com (Andy McFadden) writes:
>: In article <1993Jan8.055606.10289@nuscc.nus.sg> ltchean@rambutan.iss.nus.sg (Lim Thye Chean) writes:
>: >                          The compression and decompression takes up a
>: >lot of time, and I don't think this is the fault of AA.
>: 
>: Wrong.
>
>Why? Please tell me more about it. I am very interested. 

I can't tell you the "why" of things, because I have never seen AutoArk.  I
do have some rough time comparisons, but they were not very scientific.  I
will say that HP uses some algorithms which are inherently faster than those
available with AA.

>What does HP actually do? Expand the file only in memory (like what I always
>thought AA did before I bought it)?

Yes.  Plus some other amusing tricks.

>How fast is HP?

Very.  I laugh at YankIt (which, BTW, I also wrote).

>So you are the principle programmer of HP? Wow.

I am the only programmer of HP.  Which is why it has taken this long.

>: Random numbers:
>:     - MODs were compressing by 20% (went from 11.9MB to 9.4MB)
>:     - TrueType fonts were compressing by 25%
>
>I think AA performs better than these (in case of MOD songs... am I right,
>DOTW?)

Try some TT fonts.  Even GSHK was only getting around 8%, I believe.  However,
my sample was pretty small (the Pointless fonts + a couple of others).

You will probably not do much better with MODs, because they are largely
digitized sound.  The 20% is an exact figure (unless I read it wrong); HP
provides the statistics for you.  HP may be able to do better, but it would
take longer than most people want to wait. :-|

-- 
fadden@uts.amdahl.com (Andy McFadden)
[ Above opinions are mine, Amdahl has nothing to do with them, etc, etc. ]