OSNews

Because they are giving the customers what they have been asking for?

Customers aren't asking for it. You give your customers a line of workstations (or servers) that are clear and related to each other. You don't muck about by offering what 'you think' customers want because you really want them to be using SPARC.

Because if they sell Opterons as well as SPARC they sell more machines and make more money?

In what way? What makes Opterons outsell SPARC? And if it's outselling SPARC then sell Opterons exclusively. It's obviously a selling point. All you've told me there is that Opterons outsell SPARC.

Because the different CPUs have different strengths and are useful in different situations?

Not on workstations, as I've pointed out. SPARC is a high-end processor, and that's its only domain now. If Sun want to sell it as a workstation and convince customers then they need to make it clear to customers what they are selling. If customers are scratching their heads about what you sell, and you are confusing them with talk about 'choice' and the non-existent 'different strengths', then they simply go elsewhere.

Because you want to be able to run Windows and Redhat and Solaris all on the same box?

That's a selling point for selling Opterons (and a weak one at that considering Sun want you to use Solaris), so what's the point of SPARC?

Look, we all know that Sun don't really want to sell anything that doesn't have SPARC in it, so they've got one of two options. Commit totally to what is actually selling (Opterons), or make SPARC an actual viable option on workstations, or many servers, and stick with it. The market has spoken on that front, and that is 'what customers want'.

Customers aren't asking for it. You give your customers a line of workstations (or servers) that are clear and related to each other. You don't muck about by offering what 'you think' customers want because you really want them to be using SPARC.

Don't talk nonsense. Just skim the threads here when the Ultra 20 got released-- many people were asking for a SPARC-based workstation from Sun, simply because many, including myself, have a SPARC workstation that badly needs replacing (my aging Ultra 5). Now, for what do I use an UltraSPARC? Reliability. I want UltraSPARC workstations from Sun because they do not die. They do not break. They are as solid as a rock.

Of course Sun wants to sell SPARC machines. But do you really think they would go through ALL the trouble of setting up, designing (they even hired a world-famous server designer to design Opteron servers), multiple Opteron product lines, JUST, to, after months and months, release one SPARC line of workstations?

Go fool yourself, Segedendum. The rest of us ain't buying it.

Look, we all know that Sun don't really want to sell anything that doesn't have SPARC in it,

They bought a company that designed Opteron servers then proceeded to sell them at a market competitive price. They negotiated with AMD to get a special Sun version of the AMD CPUs manufactured. Sounds like a company who wants to sell something which isn't SPARC.

so they've got one of two options. Commit totally to what is actually selling (Opterons), or make SPARC an actual viable option on workstations, or many servers, and stick with it.

Or option 3. Sell Opteron and SPARC systems, with both being viable options.

The market has spoken on that front, and that is 'what customers want'.

I think the market is speaking by buying both SPARC and Opteron systems from Sun.

Sun may be looking to dump SPARC altogether, but I doubt it. Even if they were they wouldn't do it until Opteron (or some other chip) could replace their high end servers which won't happen for some time.

I lament the almost lack of decent, cheap SPARC workstations from Sun, I think it harms their business becuase people grow up with Linux on x86 instead. Sun have, finally, decided that they ought to be making some money from the huge x86 market. I'm glad they did.