[Info-vax] web browsers for VMS on ALPHA

[John Wallace]

On Feb 21, 10:09 pm, hel... at astro.multiCLOTHESvax.de (Phillip Helbig---
undress to reply) wrote:
> In article <RTcuQUHQU... at eisner.encompasserve.org>,
>
> koeh... at eisner.nospam.encompasserve.org (Bob Koehler) writes:
> > In article <ijuikh$lo... at online.de>, hel... at astro.multiCLOTHESvax.de (Phillip Helbig---undress to reply) writes:
>
> > > I thought upgrading to an XP1000 would speed it up somewhat, but it is
> > > still slow.
>
> >    I don't know wht chip was in an XP1000, but I thought it was an EV4.
> >    You need at least an EV5 to get decent speed in most browsers.
>
> EV67 :-(.
>
> COMPAQ Professional Workstation XP1000
>
>                 Type...........: EV67, Pass 2.2.2
>                 Speed..........: 666 Mhz
>
> OK, I haven't had it up that long and haven't done a reboot after the
> last AUTOGEN.  We'll see.

Haven't re-read the whole thread here, but the usual performance stuff
applies - identify the bottleneck (MONITOR the system) and proceed
from there.

Additionally, and subject to correction from the audience: Where did
you get your browser? Do you know how it was built (specifically, how
it was compiled)? EV4/21064 didn't have byte instructions as such, so
byte manipulation as used in lots of PC-centric software becomes a
sequence of instructions. If your binary was compiled to be run-
anywhere ie EV4-compatible it will be unnecessarily slow when run on
hardware such as EV6 which does have the byte stuff built in.