To: Joe NYC who wrote (176669 ) 1/27/2004 2:00:13 PM From: The Duke of URLĀ© Read Replies (2) | Respond to of 186894 Some interesting observations concerning the analogies of the 286/386 (16-32) transition vis-a-vis the 32-64 implemention method of the Opteron: This is from Mike Schmit, of MF, but rudedog is the real Kunta Kinte in this area: mschmit Number: of 46831 Subject: Re: For fans of 64 bits... Date: 1/26/04 8:21 PM Post New | Post Reply | Reply Later | Create Poll Report this Post | Recommend it! Recommendations: 3 (in response to) Intel's 386 didn't have any software created for it for years, and it still sold by the millions. IIRC, 386 came out in late 1985 or early 1986, and the first real "mass-market" product that used some features (not all) of the 386 was Microsoft Windows 3.0, released in 1990. What you say is true...in that the only real "product" that general users knew by name was probably Windows 3.0. However, long before that, there were drivers as part of DOS and lots of 3rd party drivers for 386 code. These allowed access to more RAM than 640K and were pre-installed on virtually every 386/486 by the OEMs right from the start. All the big name memory hog programs at the time, such as Lotus 1-2-3 and dBase II/III ~required these memory manager programs. So while you could run programs like 1-2-3 with just 640K of memory being "available" you were severely restricted. Virtually everyone in the 1988 time frame, for example, bought systems with 2 or 4 MB of RAM and a 386 memory manager in order to be able to use 1-2-3/dBase with decent performance and large (for the time) spreadsheet/database sizes. It should be noted that Lotus, Intel and Microsoft created the LIM/EMS standard for extending memory back in 1985 to enable hardware add-on memory boards to go beyond 640K for 8088 and 286 systems...because people were desperately asking for it back in 1984. With the 386 it became a trivial thing to do all this in software with a driver and just put more than 1MB in the normal address space. Mike