[Simh] performance of the emulated VAX on different hosts

[Andreas Davour]

On Tue, 11 Jan 2005 A.Bucher at alcatel.de wrote:

> Hi,
>
> since I did not read the whole archive I am not sure if this was mentioned
> or asked before:
>
> Is there any comparison in performance for simh emulated VAXes to real ones
> ?

I asked a similar question once, on comp.os.vms I think. Somebody gave 
me a DCL command file what made some calculations and gave you a VUPs 
rating for your simh VAX. I can see if I still have that around if you 
like. Maybe it can give you some indication. I don't know really.

> Are there any benchmarks for this ?
> What is better, simh on Wintel, simh on InTux ? Or another host hardware ?
> What kind of PC is required to have a similar performance of a "real" VAX
> ?
> How does a multi-cpu host affect the performance of simh ?
> What's about IO bandwith (okay, lousy on a PC compared to a VAX, no ?) and
> network performance ? can the emulator handle 100MBit (or more) ?

Good questions. I can only give you one piece of advice. Make sure your 
host system is reliable first. I think that's probably more important. I 
know some kind of systems I wouldn't use. :-)

> PS: I am thinking to give the IBM guys the rattling shake and have a
> zSeries executing VMS !  Linux  running SIMH installed with VMS should do !
> To the heck with the performance, but I want to see the VMS $ prompt on an
> IBM terminal :-)
> Just think about telling this to some computer expert in the early eighties
> ...

Yeah, it's a good time for retrocomputing and simulation right now.

/andreas

-- 
A: Because it fouls the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?