Is that it simulated 18.9 VUPS ?? it must be, because I get about 7 VUPs on my ancient 1.5ghz P4 & about 14VUPS on a 2.8ghz celeron...<br>
<br>
Villy<br>
<br>
----- Original Message -----<br>
From: "Hittner, David T (IT)" <david.hittner@ngc.com><br>
Date: Wednesday, March 11, 2009 12:00<br>
Subject: Re: [Simh] Deathrow OpenVMS SIMH Node.. Hosted by OpenVMS?<br>
To: Bill Korendyk <bill@korendyk.net><br>
Cc: simh@trailing-edge.com<br>
<br>
> I'm not ruling out OpenVMS as a SIMH VAX platform. Far from it. <br>
> Use it<br>
> if you want to.<br>
> <br>
> Simulation of another CPU eats an astonishing number of host CPU <br>
> cyclesto perform the simulation work. <br>
> <br>
> I'm trying to set your (and other's) expectation level, based on my<br>
> simulation experiences. Context is important here - maybe it's <br>
> because I<br>
> already have some real VAXes, but I don't WANT another "smaller"<br>
> MicroVAX in my cluster. I already have a physical MicroVAX <br>
> 4000/VLC in<br>
> my cluster, and to be frank, it's performance sucks at 5.8 VUPS,<br>
> particularly when compared to my physical MicroVAX 4000-500 at <br>
> 22 VUPS.<br>
> <br>
> When I ran a SIMH VAX on my Alpha PWS 500au (which runs at 155 <br>
> VUPS), I<br>
> get 1.1 VUPS on the SIMH VAX. For comparison, when I ran the <br>
> SIMH VAX on<br>
> my 3.0 GHz PC, it simulated 1.89 VUPS.<br>
> <br>
> If you have newer iron like the multi-issue Alpha ESxx/GSxx <br>
> servers (or<br>
> even a DS10!!) just sitting around that you can run simulations <br>
> on, go<br>
> for it. If you are looking at _purchasing_ a low-cost platform <br>
> for home<br>
> use to run VAX software, I'm merely suggesting that you might <br>
> want to<br>
> take the cheaper path to Hobbyist VAX performance, which is a <br>
> fast PC<br>
> running SIMH VAX - or many times, just a used VAX.<br>
> <br>
> I'd be happy to look at your Alpha-based SIMH VAX VUPS numbers for<br>
> comparison with mine; maybe my measuring and estimating sucks. :-))<br>
> <br>
> FWIW, my low-end HP Integrity rx2620 [at work] running OpenVMS <br>
> clocks in<br>
> at 533 VUPS. I've never tried running SIMH it, but I'm sure it <br>
> would be<br>
> much better than the Alpha (at 155 VUPS); but then again, just <br>
> try to<br>
> buy an rx2620 on Ebay ... $$$. The semi-affordable, semi-supported<br>
> rx2600 is pretty reasonable on Ebay at $300-$2500, but is _much_ <br>
> slowerthan the rx2620.<br>
> <br>
> Dave<br>
> <br>
> > -----Original Message-----<br>
> <br>
> > From: Bill Korendyk [mailto:bill@korendyk.net] <br>
> > Sent: Wednesday, March 11, 2009 12:14 PM<br>
> > To: Hittner, David T (IS); simh@trailing-edge.com<br>
> > Subject: RE: [Simh] Deathrow OpenVMS SIMH Node.. Hosted by OpenVMS?<br>
> > <br>
> > Hi Dave,<br>
> > <br>
> > Thanks for your response.<br>
> > <br>
> > On Tue, 10 Mar 2009 13:13:06 -0500 "Hittner, David T (IT)" <br>
> > <david.hittner@ngc.com> wrote:<br>
> > > Bill,<br>
> > > <br>
> > > The reason you don't see any networking under OpenVMS Alpha <br>
> > is likely <br>
> > > because you didn't install the VMS PCAP Execlet. I didn't <br>
> see any <br>
> > > reference to you doing so in the message below.<br>
> > <br>
> > I believe the Execlet was installed; if I recall correctly, <br>
> > the behaviour and errors were very different before it was <br>
> installed.> I'll check the next time I have a chance to work on <br>
> this. In <br>
> > the meantime, additional hints are welcome ;-)<br>
> > <br>
> > > Regarding running SIMH VAX on OpenVMS Alpha: that would be a <br>
> gross <br>
> > > waste of the Alpha unless it had absolutely nothing else to <br>
> do. The <br>
> > > performance of SIMH on a recent PC will decimate SIMH <br>
> > performance on <br>
> > > OpenVMS Alpha. Same for any Itegrity machines that you or I <br>
> > can afford.<br>
> > > :-)<br>
> > <br>
> > Hmm. I fear he doth exaggerate slightly :-)<br>
> > <br>
> > > All Alphas were excellent then, but the lower-end ones are <br>
> somewhat <br>
> > > slow now compared to high-end PCs. Higher-end <br>
> > Alpha/Integrity machines <br>
> > > might perform better, but still - SIMH really chews up <br>
> > single-stream <br>
> > > CPU cycles, and a PC is much cheaper.<br>
> > <br>
> > Yes, newer boxes are faster (and cheaper). But I already have <br>
> > the Alpha and wanted to avoid building and configuring (yet) <br>
> > another box needing care and attention, and adding to the <br>
> power bill.<br>
> > And since my "old" PCs are "really" old, buying a new PC for <br>
> > this task seemed silly with a perfectly good Alpha to use. <br>
> > SimH/VAX works fine (albeit off the network) and my <br>
> > benchmarks show it running as well as a smaller MicroVAX. <br>
> > <br>
> > So I find the SimH/VAX hosted on OpenVMS is adequate, <br>
> > especially for a hobbyist environment. Saves time and money, <br>
> > runs solid, responds well, and the process (unless very <br>
> > CPU-intensive) is barely noticed (i.e. buried in all the <br>
> > other activity) on my lowly DS10 ;-) I'd sure like to <br>
> try it <br>
> > on an Integrity...<br>
> > <br>
> > So, let's not offhand rule out OpenVMS as a SimH host shall we <br>
> :-)<br>
> > <br>
> > \bill<br>
> > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~<br>
> > Bill <br>
> Korendyk Bill@korendyk.net "Just call me Bill."<br>
> > ---------------------------------------------------------------<br>
> -----<br>
> > "Where <br>
> am I going... and why am I in a handbasket?"<br>
> > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~<br>
> > <br>
> _______________________________________________<br>
> Simh mailing list<br>
> Simh@trailing-edge.com<br>
> http://mailman.trailing-edge.com/mailman/listinfo/simh<br>
>