drochner@zel459.zel.kfa-juelich.de wrote:
> 
> jonathan@DSG.Stanford.EDU said:
> > > The /240 is >humming along at about 35-40 kKey/sec. while the /133
> > > crawls at about 4 >kKey/sec...
> >
> > The obvious question is, "How fast does the client ru on a native
> > Ultrix 5000/133"?
> 
> I've done it the other way some weeks ago, running in NetBSD's emulation
> on a '240:
> [Jan 15 13:19:16 GMT] Summary: 61 RC5 blocks 23.15:17:49.4294967240 - 
> 							[31.55 kkeys/s]
>                       Summary: 25 DES blocks 5.17:46:57.15 - [106.07 
> 								kkeys/s]
> 
> (Yes, it was running for about 4 weeks.)
> So it is a bit slower than your native result, but not _that_ much.
> 

Somehow I have the feeling that either the /133 or my 10base2 network at
home is getting a bit flaky recently... The /133 had been running for
about 20 days doing nothing by crunching RC5 blocks. Last night it
paniced with a 'corrupted amap' message while doing a 'make build'. 

Since it runs completely diskless I suspect something went haywire(?)
while swapping over the net. The machine has only 32MB of RAM, which may
be a bit tight for compiling some parts of the source tree. I'll try to
put in some more RAM later the coming week and hooking up a local disk
to see if this helps in any way...

later,
Marc