Re: HP admits discontinued IA64 workstations
From: David Froble (davef_at_tsoft-inc.com)
Date: 09/30/04
- Next message: David Froble: "Re: Tiny pickings for Itanium OEM's"
- Previous message: David Froble: "Re: As seen in WSJ"
- In reply to: JF Mezei: "Re: HP admits discontinued IA64 workstations"
- Next in thread: Nigel Barker: "Re: HP admits discontinued IA64 workstations"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Wed, 29 Sep 2004 22:44:29 -0400
JF Mezei wrote:
> David Froble wrote:
>
>>I don't think you can pin this one on the VMS engineers. They had their course
>>dictated to them, and they/re trying to make the best they can of the situation.
>>
>
> And from the point of view of their responsabilities, they seemd to have done
> an excellent job at porting VMS to that IA64 thing and minimizing the
> technical impacts of migration. And they get an A+ for that.
Ok, and you could have led with that remark. They are doing a good job within
the framework they have to work with. If you refer back to your post, you'll
see that it appears that you were blaming 'VMS engineers'. I'm glad you
reversed that particular statement. Good for you.
> However, when they say that because you just need to recompile with barely any
> changes and this migration is as easy as pie, they seem to forget that
> customers have due processes to follow in order to migrate production from one
> machine to another and such processes are far more involved than just
> recompiling. In many cases, they don,t even recompile because software is
> furnished by a 3rd party.
>
Well, if you have an application, and by that I mean you have source code and
build the application yourself (the corporate/royal 'yourself'), then having a
good chance of doing another 'build' of the executables and moving the data to a
new system is pretty good.
I do have some reservations about the entire issue. Some rather old
applications, or applications using some rather old techniques, one such being
D-float, then you have a significant problem. Even going to Alpha incurred this
issue, where D-float was first converted to G-float, losing 3 bits of precision,
doing an operation, and then converting back to D-float. This particular issue
seemed to fall on deaf ears at the time. I mention this as an example. While
there are those who don't agree with the practice, there are many applications
that used D-float for monetary data. Hey, it's worked for 25 years, care to
argue with such success?
As for third party applications, either the third party does the port and
testing, or you're probably not going to get the application running on IA-64.
In the latter case, moving to another application is a significant expense, and
no amount of re-compile and go, or anything else, will mitigate that. Those are
the people who most likely will show HP the count of one (finger) and since
forced to change applications will choose a more responsible vendor. I like
that term. What Compaq did is truly irresponsible behavior. They did it solely
for their own reasons, and didn't factor in the cost to the customers. As said
before, it was probably in part done because of the PC/Microsoft mentality and
the lack of comprehension of commitment to long term customers.
I'll claim that the following is probably wildly conservative. The cost to
customers is probably more than the cost of continuing Alpha for the next 15
years. Plug that one into your cost/benefit calculations curly.
Dave
-- David Froble Tel: 724-529-0450 Dave Froble Enterprises, Inc. Fax: 724-529-0596 DFE Ultralights, Inc. E-Mail: davef@tsoft-inc.com 170 Grimplin Road Vanderbilt, PA 15486
- Next message: David Froble: "Re: Tiny pickings for Itanium OEM's"
- Previous message: David Froble: "Re: As seen in WSJ"
- In reply to: JF Mezei: "Re: HP admits discontinued IA64 workstations"
- Next in thread: Nigel Barker: "Re: HP admits discontinued IA64 workstations"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
