Subject: Re: tangent: bounce buffers.
To: Curt Sampson <curt@portal.ca>
From: Tracy Di Marco White <gendalia@iastate.edu>
List: current-users
Date: 09/18/1996 13:29:41
}On 18 Sep 1996, Michael Graff wrote:
}> Tracy Di Marco White <gendalia@iastate.edu> writes:
}> > Or, RAM drops to an amazingly cheap price, while SCSI cards don't.
}> Or, you have old machines which would be fine for large name servers and
}> other memory intensive processes, but don't need fast disk.  They just
}> need bounce buffers.
}
}Well, let's not go overboard on this `old machines' bit. At least
}in my part of the world, RAM prices have plunged for 72-pin SIMMs,
}but not for 30-pin SIMMs. A 16MB 72-pin SIMM is $130 (Canadian
}dollars) or so; four 4MB 3-pin SIMMs run about $240. So let's do
}an upgrade comparison, starting with a low-end 486 motherboard,
}8MB of RAM, and an Adaptec 1542, and make a nameserver for an ISP
}or other medium-size site out of it:
<comparisons deleted>

Well, my current motherboard uses 72 and 30 pin simms.  So, I can add
32MB of RAM to my current 16MB of RAM, and have 48MB, for only $240
according to your prices.  Or I can sell the 16MB of 30 pin SIMMS
for $120 dollars (again according to your prices), and it'll only
cost me $120 to upgrade to 32MB.

I've already gotten mail suggesting I switch to OpenBSD.  Well, I'm 
not currently looking at this solution, but why can't we have a
machine dependent bounce buffer solution while we wait for a machine
independent solution.  It's not like we don't have other machine
dependent code.

Tracy J. Di Marco White           gendalia@iastate.edu