Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Paul Engel who wrote (70512)	1/5/1999 1:56:00 AM
From: nihil	Read Replies (1) \| Respond to of 186894

Paul -- Isn't it true a mpu will run one OS better than it will run another? Doesn't i486 run Win 3.1 better than it runs on Win98? Isn't it true that one OS can be more easily implemented on one cpu (Linux on PII) than it can on on another (Linux on 8086?). Would one expect that a team upgrading and extending an existing OS might be able to get a better outcome if the designers of the next generation of the mpu were cooperative and collaborative. I remember spending $4.5 million of Uncle Sam's money trying to get some software to run on Illiac IV -- a 64 processor parallel processor (which cost $30 million more). I know nothing about how to do it, but it seemed at the time if the designers had spent more time thinking about data paths and creating redundant communications, enhancing or optimizing the set of hardwired commans, the software job would have been much easier. But software and hardware weren't talking and were in different orgs. It took a long time for parallel processing to recover. Software people tell me that x86 is a dog that always seems to be having its last day. It is said to have about 40% overhead that requires Intel and the others to work exra hard to compete with RISC. Good things our hardware guys are 40% smarter. Too bad about Microsoft. If they don't chose to implement the OS on your mip, you are dead. You can't do it. If MS makes a lousy version for you, you are finished in the windows mrkt.

To: Paul Engel who wrote (70512)	1/5/1999 2:26:00 AM
From: stak	Read Replies (1) \| Respond to of 186894

Paul, >>How does a CPU maker directly support LINUX ?<< That's a great question, given that Intel doesn't usually tread outside of its area of expertise(Mpus). I think we'll see some good articles speculating on just exactly that question in the very near future... Just very quickly, my suggestions would be to put a team of people onto developing a version of Linux that is optimized for Intel IA-32 and IA-64 chips. With Linux's open source code this is a possiblity. Intel already has considerable experience working with the O/S companies to optimize the performance of x86 with the O/S. Intel must have said more than a few times, " If only we had more control of the O/S".(As was the case with the Pentium Pro,when the O/S didn't sync up well with the currently available O/S). The expertise with the hardware could mesh well with the software side. Or, if Intel doesn't want to directly step on Microsoft's O/S turf then they could set aside a large chunk of money for development of applications that would run on top of the O/S. Everything from business apps to games etc. The Pentium MMX instruction set didn't do much for Intel due to a lack of software that could take advantage of MMX. I fear much the same fate for the KNI instruction set. Intel can take the bull by the horns and show the world just what KNI is all about. There's no rule that says Intel can't get into consumer software. Hell, Microsoft has got lots of hardware products and they're adding more all the time. Microsoft didn't mind putting $150 million into Apple in '97. They're going to make great money on the stock appreciation and on the revenue from the office apps. Plus they gave credibilty to the iMac by backing Mac software for a guaranteed 5 years. Also, Microsoft didn't mind backing off from Chromeffects either. stak