Mail Archives: opendos/2001/01/26/12:09:05
Joe wrote:
>... Is that why some people report the 16 bit version of DJPEG (not
>sure it's exact name) runs *faster* than the 32 bit version??? ;-)
They do? I haven't seen any such thing. One reason for this diffrence could
be considered being the CPU involved. But I tested on my K6-2 400 and
Clarence Verge reported about the same results on his 386 so I kind of
doubt it has any major impact.
>Well, the above is because (IIRC) the 16 bit version outputs to a
>different format. ;-)
They are the same AFAIK.
>The point I am making is that it is very hard to make such comparisons,
>because you are not comparing "apples with apples" - often there are
>functional differences, not to mention different compilers (with different
>optimizations).
That is a valid point, I compared 6a 16-bit and 6b 32-bit (and on a
RamDrive so HD wasn't affecting speed of course). I never did a test
between 16 and 32-bit 6a but it seemed faster. Besides you are probably
refering to 6a 16-bit (as seen in 16bit.apm) and 6b 32-bit (default).
>Later versions (eg. 2.1b2) were a bit faster, but still they were *not*
>any faster than their 16 bit cousin (Arachne ;-) ... ?
Hmmm... "cousin" - parent would be a better word. I didn't use WebSpyder
much so I can't say. However it's my strong belief that a 32-bit
multi-threaded Arachne with built-in support for many plugins would be
faster than the current 16-bit version. (FWIW: That would be Arachne 2.x).
Ok, it might not be faster on 386's - but seriously people do throw away
486 everywhere, and Pentiums aren't rare to see either you know - so I
really don't understand why so many out there are still using 386's.
//Bernie
- Raw text -