[Simh] VAX/VMS

Timothe Litt litt at ieee.org
Tue Feb 16 10:12:20 EST 2016


On 16-Feb-16 08:58, Clem Cole wrote:
>
> Also it is also interesting to consider that while the AT&T folks came
> off of Multics, a number of us university types that would work on
> earlier Unix came from TSS and MTS (one 360/67).   In fact, TSS is
> still the only system I ever used that lived in the debugger as your
> command system - which I always thought was a cool idea.  
>
ITS (MIT PDP-6/10) did that too.  I wasn't a frequent user, but the CLI
was DDT.

> Also, if you peeked inside a modern processor, you would discover they
> are dataflow engines and put together with all of the modern computer
> science; but there is about a 5% silicon tax paid for compatibility. 
> Clearly, my siblings at Intel believe it's worth tax and the customers
> seem to keep wanting it.
5% is probably fair as as an estimate of silicon area.

Actually, the real tax is in processor validation.  I worked briefly in
that group.  It costs more than 5% in manpower to keep making sure that
the compatibility modes (yes, 's') work.  The older modes have lots of
quirks and errata that have to be kept bug-compatible in the face of new
implementations.

The other taxes are more subtle; things like the memory ordering model
and i-stream modification have real performance costs.   So does the
limited opcode space, which has resulted in ever more opcode prefixing. 
Intel tries to minimize these with ever increasing cleverness, but
they're still there.  And probably always will be.

Alpha tried to wipe these out with one swell foop, but failed to grok
the software costs.  (I did point them out at the time, but the general
belief was that the migration tools would be good enough and that
customers would pay the price for the performance.  They weren't, and
they didn't.)


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.trailing-edge.com/pipermail/simh/attachments/20160216/e0b3e56e/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4994 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://mailman.trailing-edge.com/pipermail/simh/attachments/20160216/e0b3e56e/attachment.bin>


More information about the Simh mailing list