[Info-vax] Why it is a good idea that OpenVMS isn't on x86-64 just yet

[Johnny Billquist]

On 2016-01-19 16:40, John Reagan wrote:
> On Monday, January 18, 2016 at 5:12:55 PM UTC-5, Bob Gezelter wrote:
>> On Monday, January 18, 2016 at 5:06:06 PM UTC-5, Hans Vlems wrote:
>>> John, what on earth did you do wrong to deserve that???
>>> The TU58 was probably cheap, certainly less expensive than the RX02 on the 11/780 along with its LSI-11.
>>> The 11/750 had to be slower than the 11/780 because DEC had announced an entry level VAX. The gate array technology was rather new and I think engineering was pleasantly surprised with the end result. Marketing must have felt a different emotion...
>>> The 11/750 cpu ran slower than the 11/780. At 0.6 VUPS it was twice as fast as the 11/730. Its only advantage was its small size, roughly a pdp 11/40 cabinet, a lot smaller than the 12/750 and no a/c required.
>>> Two TU58 drives didn't really help either.
>>
>> The 11/730 was also, if I recall correctly, at a far lower price point. The lower investment made it attractive to ISVs developing software for the platform.
>>
>> A point worth remembering: Capital costs are an impediment to high risk development efforts.
>>
>> - Bob Gezelter, http://www.rlgsc.com
>
> I also think the 730 was the last machine to do H-floating in real hardware.  I remember seeing some crude benchmarks between the 730 and 750 for REAL*16 showing the 730 to be much faster.

No. The 86x0 machines were the last to implement pretty much everything 
in "hardware", including PDP-11 compatibility, and all the FP stuff.

	Johnny