February 20, 2004

Snap! Goes The Intel!

http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-125-1-1-618817-5926-1

After over a year resisting the 64-bit trend pioneered by AMD, Intel has announced that it has implemented 64-bit extensions to its 32-bit processor line. This gets them into the 64-bit market on nearly the same level as AMD, namely, offering CPUs which [unlike the ITANIUM] offer backward compatibility with 32-bit operating systems.

Intel is still insisting that 64-bit systems have no place on the desktop, and is currenly limiting this extension technology to its high-end 'Xeon' server chip. The move to 64-bit computing has been discussed at length in this blog, and I still think Intel is mistaken [but I acknowledge they make more money and have more expertise than I do].

An analysis of this development, indicating that Intel is well-positioned to take advantage of the deep factors pushing for the transition to 64-bit computing [increased security is one of them] as well as the problems to come in creating 64-bit drivers, can be found here:

http://www.eweek.com/article2/0,4149,1528428,00.asp

Another look at the 64-bit Xeon chips, and how they relate to the existing 64-bit Itanium can be found here:

http://www.esj.com/news/article.asp?EditorialsID=864

A page displaying a series of articles on 64-bit computing, discussing AMD, Intel, IBM, and the whole background of 64-bit computing can be found here:

http://www.eweek.com/category2/0,1738,1425416,00.asp

Rather ironically, an analyst has concluded that Intel's 64-bit processor extensions were reverse-engineered from AMD's:

http://www.extremetech.com/article2/0,1558,1562294,00.asp

Posted by jho at 11:23 AM | Comments (1)

November 26, 2003

October 20, 2003

Willing Intel

http://www.newscientist.com/news/news.jsp?id=ns99994215

The portmanteau word "Wintel" has served as a shorthand acknowledgement of the throughough interdependence between Intel's hardware and Microsoft's software. So significant has this relationship been that despite a bias in this blog about announcements about deliveralbles Real Soon Now, I am mentioning this one: a CPU chip which will allow different OS to run together with the same ease as different applications run under one OS today.

I would be the last to underestimate the value of this, the threat it poses to Microsoft, or the degree of difficulty in actually delivering on this promise.

Posted by jho at 05:33 PM | Comments (0)

October 15, 2003

64-Dust-Biting Apple

http://www.pcworld.com/resource/printable/article/0,aid,112749,00.asp

Article which explores several important aspects of 64-bitness, including which hardware systems are ready, which operating systems are ready, the limited state of play with available applications, and what the future is likely to bring. What struck me so forceably is the reminder that it took 15 years between the introduction of the first 32-bit capable Intel processor and the rollout of a 32-bit operating system.

A number of factors suggests the delay in moving to 64-bits will be much less, and just like the 386 chips, the overall performance of the 64-bit chip even on 32-bit software may be sufficiently attractive in itself to guarantee it a market.

The other item worth remark in this article -- in a [admittedly crude] faceoff between Athlon-64s using the FX technology [the high end Athlon] and the Apple G5, the Athlon-64 boxes wielded a tin of the proverbial substance on most tests, and were 10% cheaper to boot.

Posted by jho at 02:46 PM | Comments (3)