Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!olivea!mintaka!charm.LCS.MIT.EDU!mikew From: mikew@charm.LCS.MIT.EDU (Michael B. Williams) Newsgroups: comp.windows.ms Subject: EUREKA! (Was Re: Performance degradation in 386 enhanced mode) Message-ID: <1991Jan21.211522.18732@mintaka.lcs.mit.edu> Date: 21 Jan 91 21:15:22 GMT References: <4491@mindlink.UUCP> Sender: daemon@mintaka.lcs.mit.edu (Lucifer Maleficius) Reply-To: mikew@charm.LCS.MIT.EDU (Michael B. Williams) Organization: Massachusetts Institute of Technology Lines: 79 In article <4491@mindlink.UUCP>, a752@mindlink.UUCP (Bruce Dunn) writes: |> |> From recent responses, it is apparent that there is some confusion about |> the problem which is being experienced. The basic problem is as follows: |> |> On some machines, but not all, tasks running in 386 enhanced mode are much |> slower than in standard mode. This can be seen by using the computational |> index from the Norton Utilities in both standard mode and 386 enhanced mode. |> It can also be seen for some programs such as the Microsoft Flight simulator, |> which become extremely sluggish in 386 enhanced mode but are ok in standard |> mode. ... |> I have the problem on a 16 MHz 386 (not a 386 SX). I wonder if the |> problem is because my memory above 1 megabyte is not on the motherboard, but is |> on a memory card on the bus which can only be accessed at bus speed. My |> hypothesis is that when run in standard mode, Windows runs mainly in the 1 |> megabyte of easily accessible memory on the motherboard, but when I ask for |> enhanced mode much more use is made of the memory which can only be accessed |> through the bus. Would people who have the problem please mail me |> information on whether their memory is all on the motherboard, or whether part |> is accessible only through the bus. Bruce, after months of enduring countless (and incorrect) explanations of multitaksing in 386 mode as the cause of this problem, I think that you may have finally found the REAL cause: >>> AT-bus memory! <<<< Here's the story: About a year ago, I was running Windows/386 2.11 with 2 MB of memory on the motherboard of my SX machine. I had no problems running DOS apps quickly (as long as I ran them full screen), except that with only 2 megs I could only run one of them at a time. (Windows veterans will recall that Win 2 did not implement swap space.) So, I bought a BocaRAM AT + card with 2 MB of memory, figured out how to run it at 0 wait states, and everything ran fine. As I recall, DOS apps ran as fast as they always had. Then came along Windows 3. After I installed it, I could run as many DOS apps as I wanted, but they ALL ran *very* slowly--too slow, in fact, to be useful. I tried running them in exclusive mode, full screen, tweaking this, and tweaking that, but DOS apps were still too slow t use. The only solution was to fall back to Standard Mode, where I've been ever since. But recall that I added 2 megs before I every installed Win 3--so I've never run Win3 on my bare-bones 2 MB machine. Would that make a difference? As it turns out, it makes all the difference in the world. I left the actual figures at home, but here are the results of running SI with and without the memory: 2 MB of memory on motherboard, only 386 Mode Standard Mode Full Screen 15.6 16.8 2 MB of memory on motherboard + 2 MB in a slot 386 Mode Standard Mode Full Screen 4.6 16.8 Windowed 3.0 n/a hDC App's memory viewer showed that the Windows program took twice as much base memory (that's twice as much of the 2MB on the motherboard) when I have 4 MB as when I have 2 MB. So it seems that by the time I get around to opening a DOS window, Windows has already gobbled up the fast 2 MB on the motherboard, and all I can get is the slower 2 MB of memory (as Bruce theorized). Whew. It's a relief to finally (believe that I) know the cause to the problem. Now I wish I had a solution! My guess is that this doesn't happen in Standard mode because Windows swaps its lower portion (the part in the base 640K) out to make room for the DOS window, whereas in 386 Mode, Windows uses the remapping ability of the 386 to make the DOS window *think* it's running in the lower 640K. Can anyone verify the above? If this turns out to be true, I congratulate Bruce for being the first person to suggest an explanation for this 8-month-old problem!!! ________________________________________________________________________ Michael B. Williams \ 1-2-3-4, KICK THE LAWSUITS OUT THE DOOR MIT NE43-532 \ 5-6-7-8, INNOVATE DON'T LITIGATE Laboratory for Computer Science \ 9-A-B-C, INTERFACES SHOULD BE FREE 545 Technology Square \ D-E-F-0, LOOK AND FEEL HAS GOT TO GO! Cambridge, MA 02139 -------------------------------------- (617) 253-5983 Internet: mikew@athena.mit.edu CompuServe: 73667,3264