X-Message-Number: 8166
Subject: Re: CryoNet #8159 - #8164 
Date: Fri, 02 May 1997 09:31:52 -0400
From: "Perry E. Metzger" <>

> From: John de Rivaz <>
> 
> I am wondering whether I am ignorant as well.
> 
> Surely if you have 32 MB free RAM in your computer and a simulation (of 
> another computer) requires 8MB of overhead, if you run it once then you have 
> 24MB to play with. If within the first simulation you run another which also 
> requires 8MB then you have only 16 MB left.

Nope.

Modern machines don't use their RAM that way. They use virtual memory
subsystems -- MMUs are used to permit the machine to fool itself into
thinking that secondary storage is RAM, and to share executable code
and data transparently between processes.

> From: 
> 
> 1. He complains that I refuse to answer his question,  whether I can prove I
> am in the "real" world. How tiresome and lacking in perception! My postings
> have OBVIOUSLY accepted--as a basis for discussion--the possibility that all
> of us are living in a simulation. I doubt it, because, for reasons I have
> repeatedly stated, I doubt that a simulation could have subjective
> experiences; but I don't deny the long-shot possibility, and my disussions
> have clearly assumed that it is possible. 

I didn't suggest that YOU were simulated, Mr. Ettinger. I suggested
the world you were observing was. You see, as I've repeatedly said,
I've taken your brain, disconnected it from your body, put it on life
support, and hooked you in to a sophisticated virtual reality
system.

> Whether "crash" was a well chosen word is at most, to use Metzger's own
> expression, a matter of "silly semantics." If the computer effectively stops
> or freezes up, that's close enough to a crash for me.

Crash encompasses "freeze". Why would such a machine "freeze up",
either?

As I said, if you don't believe one machine can efficiently simulate
another, don't believe me on faith -- go and buy a copy of
"SoftWindows" for your Macintosh.

> I pointed out that, for reasons previously discussed, if we could create a
> simulated world with simulated people, those simulations would almost surely
> produce large numbers of somewhat different subsimulations, etc. If we have
> just one real computer, clearly it could not support such a cascade without
> slowing down to near zero.

And I've mentioned repeatedly that this isn't true, for the obvious
reason that if we are dealing with a sufficiently detailed simulation,
the simulation won't care if its simulating a waterfall or a
microprocessor.

> Then he goes on, against the freeze-up problem, by saying you can't build an
> unlimited number of computers on earth. In one of his favorite phrases, So
> what? How is this relevant to my point? 

Well, you seem to believe the fact that a computer can't simulate an
infinite number of computers "proves" you can't build a simulation of
the universe -- except for the fact that you can't build an infinite
number of computers in the universe around you, right now, as things
stand.

> Metzger asks, "So? Why is this interesting?" Could anything be more obvious?
>  The general discussion concerned the possibility of real people "uploading"
> and carrying on their lives as simulations.

You still haven't addressed the question of whether you aren't a
biological brain attached to a simulated universe RIGHT NOW, TODAY.

> He then continues to attempt to defend his rejection of the subsim problem by
> saying  you can trade memory for time. That is exactly MY point. With more
> and more work to do, it takes longer and longer. The single, original
> computer MUST run slower and slower (relative to demand) as more and more
> simulations and subsimulations are generated; pretty soon (VERY soon) the
> whole thing essentially grinds to a halt.

There is a slight problem with your reasoning here -- you've missed
the distinction between "slow" and "halted". "essentially halted"
is a handwave.

> 3. I pointed out that, if simulated worlds with simulated people were really
> possible (which I don't believe), we should probably have zillions of
> subsimulated worlds,

There isn't space on the planet you are standing on right now for
zillions of computers, so why should a simulation be defective if it
doesn't simulate a large enough universe for "zillions" of computers?

> He asks, "Why?" Again, could anything be more obvious?

Lots of things are obvious to you, but not obvious to the rest of
us. Perhaps its your privileged frame of reference.

> Then he asks, should the fish in his "Fish & Sharks" virtual world have
> prayed to him? Certainly--if they were intelligent beings with feelings,

No, if they were intelligent creatures with feelings, praying to me
would have been pretty silly for them. I wouldn't have been
"listening" for prayers, for one thing. You see, just because you're
God doesn't mean you're omniscient.

> I haven't touched all of the bases,

or even any of them.

> but that's enough for now.

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=8166