Port-vax archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Address of second TMSCP controller



Den 2021-01-26 kl. 15:12, skrev David Brownlee:
On Tue, 26 Jan 2021 at 13:08, Nigel Johnson <nw.johnson%ieee.org@localhost> wrote:

Right now, just 12.  I have to make up a four-way cable to put in the third 4mB board! It's slow as molasses in January!
It's a uVAX II - the only way to make it fast is using a trebuchet :-p

Seriously though, modern NetBSD is targeted towards a much heavier
memory and compute footprint.

One thing to check is if you have makemandb running on initial boot -
it will chew up pretty much the entire machine for quite a while until
it rebuilds the manpage search index. You can set makemandb=NO in
/etc/rc.conf

If you're looking for a possibly interesting benchmark to run, you
could build a distribution with NOPIC defined and compare how long it
takes to boot (binaries will be bigger, but you lose the dynamic
linking overhead which I think is quite high on NetBSD/vax)
Note that compiling the NetBSD 1.2 distribution on an 11/780 took 11 days (and two additional days to gzip). Thinking of the speed of a MV2 and the size of NetBSD today it might be finished this time next year :-)

Actually, it's not only NetBSD that is slow.  Even compiling a 4.3BSD kernel on an 11/750 took a full day (around 6 hours IIRC).

-- R


Home | Main Index | Thread Index | Old Index