[Info-vax] Itanium Kittson lives on?

[Neil Rieck]

On Wednesday, February 10, 2016 at 3:27:33 PM UTC-5, JF Mezei wrote:
> On 2016-02-10 12:14, johnson.eric at gmail.com wrote:
> > Text is a bit ambiguous. But sounds like it will be available if you know to ask for it.
> > 
> > http://www.pcworld.com/article/3031703/hardware/intels-itanium-to-live-an-as-hpe-commits-to-new-servers-with-the-chip.html
> 
> 
> The way I read this is that the original deal where HP would pay for X
> number of iterations that would end with Kittson was still in place.
> 
> I am still of the opinion that Kittson will be to Poulson what EV7z was
> to EV7. A speed bump, possibly the higher quality chips from the Poulson
> run marketed as Kittson because they are capable of higher clock rate.
> 
> Since the support clock starts at end of sales, HP could release Kittson
> today and keep it in the order books for a few years more.
> 
> On the other hand, if they have too much Poulson inventory left, they
> want to reduce that as much as they can before releasing kittson.

While I have no doubt that we will see OpenVMS on "x86-64 XEON" and that it will be successful, this is not the same as running OpenVMS on a consumer PC which may never happen. Meanwhile, while progress continues on all Intel product lines, it is not happening as fast as most people think.

http://www.extremetech.com/extreme/223022-the-myths-of-moores-law

Just the phrases "Moore's Law" and "Moore's Wall" send people's brains in different directions. For example, the original definition of Moore's Law stated that the number of transistors would double approximately every 18-months "for the same cost" (which many equate to price). If true, then we hit "Moore's Wall" more than 10 years ago. Meanwhile, if you only think about "Moore's Wall" from the a technical point of view where geometry shrinks will expose the gates to increased quantum tunneling etc then we've got a while to go.

But in the end it does not make sense to have two products competing with each other (see: http://www.r-5.org/files/books/ethology/experience/Merrill_R_Chapman-In_Search_of_Stupidity-EN.pdf ) so Itanium will eventually yield to "x86-64 XEON". In the next 10 years (perhaps up to 2025 and beyond) companies who run businesses will sleep a bit better knowing their Itanium processors will be unable to become infected by x86 viruses. (this does not exclude other types of attacks like "SQL-injections", "plain-text sniffing" to only name two of many)

Neil Rieck
Waterloo, Ontario, Canada.