>>>>>I am willing to bet 100 USD that VFP will run without problems on the Windows version which is 64-bit, made for Intel processors and sold by Microsoft in 2025. Will you bet against me?
>>>>
>>>>not arguing the assumption that vfp will [mostly, SMB2 et al.] run on 32-bit layers -
>>>>but unclear if in 2025 acurrent 64 bit version will be sold (not counting MSDN-type older licenses).
>>>>
>>>>32 bit had about 20 years lifetime,
>>>>64 bit surfaced 2005 as XP64 or as Server 2003 -
>>>>64 bit might be runnigt only in your watch by then ;-)
>>>
>>>The difference between 64-bit and 128-bit is far bigger than most people understands. I give 64-bit a least 25 years.
>>
>>IMO depends on how well multiple cores can be used this decade to further computing expirience.
>>clock speed enhancements via shrinking have slowed down tremendously ;-)
>>so perhaps larger registerand bus sizes might come in handy to speed up calculating deficits post 2020.
>
>I'm also a hardware man myself, so I know quite a bit of how a CPU is made, production wise. What makes me predict that a main stream 128-bit CPU won't be released in many years, is the total of many factors. I will mention a few of them.

>
• The relatively little gain compared to the complexity of having an even wider bus, both internally (inside the chip) and externally,
  >
• The complexity of the CPU itself. They are already facing the problem that they are not able to raise the clock frequency very much, and they are not able to reduce the width of the wiring inside the chip. The result will be a yield which is too low to justify production.
  >
• Even if they overcome the technical challenges, there simply are too few applications which will have any substantial advantage.
  >
• And last; the price of such a chip will be way too high because of the low yield and demand.

Really now. Everyone knows mainstream 128 bit cpus won't be available until at least a couple of years after one gets sent back in time to kill John Connor.
:)