Joe,
DEC is clearly a technology loser, but not in the narrow sense that you wish to define it.
DEC certainly created faster chips than Intel did at the same point in time - a superior achievement. It did so, in part, by not being bound by a prior architecture.
Intel chose to bind itself to compatibility with its past. This was a risky approach, for it was no doubt clear to Intel that it would take much more effort to overcome the deficiencies of an older architecture, and slowly evolve it to a more efficient one, than it would be to abandon it and try migrating everyone by source compatibility. But Intel made the technology judgement that compatibility was more important than raw performance, within the bounds of what the performance difference was likely to be. They were right, and continue to reap the rewards of this decision.
Intel further believed strongly in elasticity of demand and economies of scale, enough so that they placed huge investments in manufacture, which is now an engine of their success.
For volumes, Intel virtually ignored the scientific/engineering market in favor of the commercial desktop. It has only been in the past few years that Intel has launched attacks outside that base. This showed an understanding of the technology needs of a huge unmet demand base.
Finally, Intel made its name virtually synonymous with processors, by a well-run advertising campaign.
A company that produces a processor that is fast, but doesn't have a software base, can't/isn't manufactured in economical volumes and/or isn't demanded by customers because it isn't correctly marketed or because the market really didn't exist, is a company that is a "technology loser". When/if they rectify these failings, they have a chance to be a "technology winner", if they manage to do them better than the other guy.
Best regards,
Arno |
