Posted on 09/29/2002 7:15:38 AM PDT by dennisw
Win one for the Gipper! God Bless You Reagan, We Will Never Forget Your Great Service and Leadership - We here on FR will carry on your great work with diligence. Thanks for the Memories and Inspiration!
If power efficiency does indeed trump processing speed, everything that Intel and Hewlett-Packard have done ....could now be a handicap. The chip ...is not even a contender in the Google universe. "We're incredibly, incredibly power sensitive..." Dr. Schmidt said.
Another example of what happens when the guys who think they are smarter than everyone else forget about the whole system. Every top end computer users struggle with bills for electricity and construction costs for buildings and cooling.
Unfortunately, only a few of the major computer makers currentl y offer an AMD option. However smaller makers and us geeks who custom build our systems have known for years that AMD gives better performance at a cheaper price.
This is the problem. The Itanium is too much of a radical change. Assembly language is a mess and standard compiler optimizations don't work well with the chip. None of this would matter if the chip was blazingly fast. Unfortunately, the bet on VLIW itself failed! VLIW doesn't provide the boost over x86 that was expected. Combine that with the physical complexity of the chip, which limits clock speed, and you have a disaster.
Intel has done this before. Some old-timers out there might remember the Intel i860. It was supposed to be the great break from x86. It flopped.
Bonus question: how is the i860 related to the name "Windows NT"?
Delaying or cancelling new systems....could be the starty of a trend if the industry stays at 2002 levels
Anyone know where the Google complex is located?
I would bet it is in California. Hence the concern regarding the power! Grayout Davis strikes again!
No mention in the article of the great IBM attempt know as FS - Future System - that was to be a complete break from the highly successful 360/370 that cratered because of its requirement to reprogram everything!
No mention of the stealth companies trying other approaches such as Transmeta and the VIA C3 . Perhaps a later article will widen the focus, there is much happening!
Um. Currently, there's no reason for them to buy. Everything works over at Google.
5 years from now, sure, Google might be looking at upgrading but now? Why bother?
What a dopey article.
It could almost be titled , HP bets the company ( and its merger with Compact ) on Itanium.
They have a much bigger problem than Intel if Itanium doesn't find good market acceptance!
"we're so smart, we make computer's"
since day one.
Now this facades exposure has reached critical mass as so many have seen these "wizards" as myopic dingbats lost in a dingy on a sea of change.
The Itanium does have an x86 execution mode, but due to the constrained clock rate, it's probably slower than a Pentium.
It's unfortunate that Intel will be stuck with their 1970s-era x86 architecture for a long time to come.
| Intel has done this before. Some old-timers out there might remember the Intel i860. It was supposed to be the great break from x86. It flopped.
This chip pre-dated the Pentium by several years. In a scary (for Intel) parallel to the Itanium, the 432 was going to take us into the world of 32-bit computing. It was supposed to be the successor to the 8086. But like the Itanium, it required all new software. We all know what happened next: the 432 fell on its butt (a shame, really, because it truly was an incredibly advanced architecture for its day) and a turn-the-crank-one-more-time-on-the-x86 called 'Pentium' became the market's choice and Intel's flagship processor. Watch for the same thing to happen again. It takes a huge hardware price/performance difference to overcome software inertia. That is just not here, either in the chip itself or -- as Mr. Google points out -- in total cost of ownership. Watch for the souped-up X86's, whether AMD's or Yamhill, to mow the Itanium down as soon as they are available. |
| "The real challenge to the Itanium may have less to do with marketing and design and more to do with a collapsing economy," said Michael Shulman, an analyst at ChangeWave, a research firm in Potomac, Md.
The chip may well fail in the market -- I think it will -- but Intel has deep enough pockets that they can easily weather a cyclical downturn. What a jerk this guy is. |
It will be at least another year before IBM's 64-bit GigaProcessor will be available in Apple's servers and professional systems. Apple is testing it now, but I'd estimate a product won't be delivered until about July 2004. Perhaps more information will be available after the Microprocessor Forum conference in a couple of weeks.
The 32-bit and 64-bit Motorola G5s should be available in early 2003, but I expect Apple will deploy only the 32-bit version initially.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.
They have done this before, but it wasn't the i860. That was the RISC chip, wasn't it? The earlier -- and very similar -- "breakthrough technology" debacle was the iAPX432, which when announced was going to be The Future Of Computing.