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 : How high will Microsoft fly? -- Ignore unavailable to you. Want to Upgrade?

To: cheryl williamson who wrote (8294)	6/4/1998 2:19:00 PM
From: mozek	Read Replies (2) \| Respond to of 74651

Thanks for the response Cheryl. Check with your CT @MSFT, there were a number of "undocumented features" to MS-DOS that were not divulged to the public in an official form until MSFT published the MS-DOS spec, after it had outlived its usefulness I might add, for $130. They can be found in "DOS Programmer's Reference 2nd Edition" by Terry Dettman & Jim Kyle published by Que Corporation. I'm well aware of this book. I got a copy of it when it first came out. I also reverse engineered many undocumented features of DOS before they were ever published. Since I exploited them for TSRs and a mutlitasking system, I learned ways of adapting my commercial programs to the nature of undocumented APIs, changing from version to version in function and spec. At the time, I didn't work for Microsoft, but it made sense to me that these would be undocumented. They were by their nature too wired into a specific OS version's architecture and not very useful for applications. My question still remains. Do you know of any specific API that was exploited by a Microsoft application to the detriment of any third party application? In my years of systems development both outside and inside of Microsoft, I have heard this complaint many times, but I have yet to learn of any specific example. Thanks, Mike

To: cheryl williamson who wrote (8294)	6/4/1998 2:42:00 PM
From: Logos	Read Replies (6) \| Respond to of 74651

RE: <<I've got news for you, SUNW's low-end servers and workstations are now competitively priced with high-end PC's running NT ... SUNW Enterprise 450 servers, which are competitively priced w/high-end CPQ 8-way multiprocessor PCs still outperform them by a factor of 2-to-1.>> Would you please provide the benchmarks or the source for this? I'm currently writing a report for my company on the server market and I was under the impression that NT servers had the most TPM-C bang for the buck. Any info would be appreciated. Logos

To: cheryl williamson who wrote (8294)	6/4/1998 10:06:00 PM
From: J Krnjeu	Respond to of 74651

Ms. cheryl williamson, <<<The fact this book exists at all is a perfect illustration of how MSFT began to leverage its OS in order to sell their own applications. Contrast that with Unix, which has always been fully documented, and in fact, the source code for BSD Unix is free to the public.>>> I was reading a UNIX book today and it stated that there are so many variations of UNIX, that there is no standard. HP's, Sun's, DEC's unix only run on their own box and can not run on a competitors box. They handle basic routines differently making them unique. As far as the documentation, make sure it's for the variation you want. Thank you JK I really think that the days of UNIX being a standard

To: cheryl williamson who wrote (8294)	6/4/1998 10:08:00 PM
From: mozek	Read Replies (2) \| Respond to of 74651

Cheryl, Thanks for the news. I did a little checking and found some information that might surprise you. My source was the Transaction Processing Performance Council, a 3rd party industry organization located at tpc.org on the web. You said: I've got news for you, SUNW's low-end servers and workstations are now competitively priced with high-end PC's running NT. SUNW Enterprise 450 servers, which are competitively priced w/high-end CPQ 8-way multiprocessor PCs still outperform them by a factor of 2-to-1. Interestingly, a Compaq 200mhz 4-way Proliant 5500 server running Microsoft SQL server (I know it's not the fastest database yet :-) ran slightly faster in absolute terms than a Sun Enterprise 450. The TPC-C throughput for the Compaq was 11748. The Sun throughput was 11559. Since you say that a Compaq 8-way is half as fast as a Sun 450, are you saying that adding 4 more processors slows things down? I suppose I'd just use 4 then. Another interesting fact is that the price/performance ratio for the Compaq is $26.61 while the Sun (which was the cheapest in Sun's lineup) comes out to $56.60, making Sun's least expensive offering more than twice as expensive as Compaq's. From there, Sun's prices range up to the $200+ level. It seems that Sun's highest performance offering tested in absolute terms, the Ultra 6000, barely edges out Compaq's Tandem Server Cluster running NT at a substantially higher price. Here's a summary of all the tests: tpc.org Here's the Sun 450: tpc.org Here's the Compaq 4-CPU Proliant 5500: tpc.org You can look up the others as you like. Thanks, Mike