tom@tandon.UUCP (Tom Friel) (11/16/89)
In our Tandon booth at Comdex, I was fortunate enough to be the "man" for our new 486 systems. We are located right next to Quarterdeck, makers of DesqView. Late Monday night, they came over to play with one of our machines, which thanks to Toshiba's gift of 4MB SIMMS, had 40MB of memory (not a typo!) in it. The Q-deck guys wanted to see how fast it was, and had a great time. I'm posting because we had a result no one could explain well. We run Landmark Speedbar at 117-118Mhz (good, I'm told). When we ran Speedbar in a DesqView window, it showed over 150Mhz. Unfortunately, I didn't see this myself. How can this happen? The one thing I DID do myself last week was to run Speedbar within a VP/ix window on 386/ix and the results were down under 100MHZ -- obviously some overhead. So what's happening with Speedbar and DesqView? My guys here are off on other stuff and can't investigate. Help! By the way, Xwindows is a gas on a 486! I'll bet you all can't wait! -- Tom Friel | Tandon Computer Corporation | (805) | tom@quad1.UUCP | 609 Science Drive Moorpark, CA 93021 | 378-7881 | tom@quad.com | Moorpark, CA 93021 |__________| UUCP: ..psivax!quad1!tandon!tom ..psivax!quad1!tom
PICHER@MAINE.BITNET (Michael W. Picher) (11/20/89)
Tom,
Apparently how Landmark works is that it measures a 386's speed relative
to a 386/20 with 1 wait state. So, if you have a say Micronics 386/20
motherboard running with 0 wait states you will see a landmark speed
rating of about 24 Mhz or so (at least this is how I have had it
explained to me). Now, suppose you run in a Desqview window... You are
running the processor in the protected mode and it will run faster
(but a speed rating of 100 Mhz under 386/ix???, maybe with a 386/33
of if you were talking about running 386/ix on the 486). Anyway, I do
believe that because you were in the protected mode you were getting
a higher speed rating than in DOS.
Mike
Michael W. Picher, / ** ** MicroLab / LexIkon Microsystems
Vice President / ** ** 333 Water Street
/ *** ** Augusta, Maine 04330
Picher@Maine.Bitnet / ** **** (207) 623-4012
/ ** Consultants and PC Compatible Manufacturersdavidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (11/22/89)
In article <2554PICHER@MAINE> PICHER@MAINE.BITNET (Michael W. Picher) writes: | Now, suppose you run in a Desqview window... You are | running the processor in the protected mode and it will run faster | (but a speed rating of 100 Mhz under 386/ix???, maybe with a 386/33 | of if you were talking about running 386/ix on the 486). Anyway, I do | believe that because you were in the protected mode you were getting | a higher speed rating than in DOS. By any chance do you mean "virtual 8086 mode" rather than "Protected mode?" The ix/386 native stuff runs protected, but the DOS emulation runs virtual86. I'm not sure about the Desqview, I thought it used QEMM to emulate LIMS, but I don't have the details fresh in my memory. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
madd@world.std.com (jim frost) (11/22/89)
PICHER@MAINE.BITNET (Michael W. Picher) writes: >Now, suppose you run in a Desqview window... You are >running the processor in the protected mode and it will run faster You are mistaken; protected mode will slow the processor down, as it must do access checking. If you are doing memory mapping, things will run even slower since you have to access the mapping tables before doing the real memory access. jim frost software tool & die madd@std.com
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (11/22/89)
In article <1989Nov21.220611.4804@world.std.com> madd@world.std.com (jim frost) writes: | You are mistaken; protected mode will slow the processor down, as it | must do access checking. If you are doing memory mapping, things will | run even slower since you have to access the mapping tables before | doing the real memory access. Since protected mode always applies a segment lookup to each memory access, I don't know what you mean by "if you are running memory mapping." When running in DOS under UNIX environments the clock tick is simulated, so if that is being used for timing the results can be off. Since the system never ADDS ticks, just delays them the number of ticks will be low, making the speed of programs appear faster. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
gilmore@vax1.acs.udel.EDU (Scott Gilmore) (11/22/89)
In article <1989Nov21.220611.4804@world.std.com> madd@world.std.com (jim frost) writes: >PICHER@MAINE.BITNET (Michael W. Picher) writes: >>Now, suppose you run in a Desqview window... You are >>running the processor in the protected mode and it will run faster > >You are mistaken; protected mode will slow the processor down, as it >must do access checking. ... >jim frost >software tool & die >madd@std.com Not so, says the latest issue of PC Magazine. According to PC, the new optimizations in the 486 design mostly affect protected mode operation. They found that the projected speedup factor of 2 or more from a 386-33 to a 486-25 was not realized for DOS applications, where they found only a speedup of roughly 10%. Some of the 486-based machine manufacturers (such as ALR and HP) claim that the projected speedup becomes reality when running in protected mode. PC Mag. did not test the machines in protected mode themselves. I suppose the best test would be to benchmark true 386 32bit applications on both processors. Anyone with the hardware and software to do it? ------------------------------------------------------------------------------ Scott Gilmore Research Assistant University of Delaware gilmore@vax1.acs.udel.edu Center for Composite Materials gilmore@oscar.ccm.udel.edu Dept. of Mechanical Engineering gilmore@minnie.me.udel.edu ------------------------------------------------------------------------------
PICHER@MAINE.BITNET (Michael W. Picher) (11/27/89)
========================================================================= Path: maine!cunyvm!psuvm!psuvax1!brutus.cs.uiuc.edu!samsung!uunet!crdgw1!crdos1!davidsen From: davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) Newsgroups: comp.sys.ibm.pc Subject: Re: Tandon 486, DesqView, Landmark Speedbar question Message-ID: <1686@crdos1.crd.ge.COM> Date: 21 Nov 89 20:36:21 GMT Organization: GE Corp R&D Center, Schenectady NY Reply-To: davidsen@crdos1.UUCP (bill davidsen) References: <8@tandon.UUCP> <2554PICHER@MAINE> Distribution: usa Lines: 18 Bill Davidsen (davidsen@crdos1.crd.GE.COM) writes: | By any chance do you mean "virtual 8086 mode" rather than "Protected |mode?" The ix/386 native stuff runs protected, but the DOS emulation |runs virtual86. I'm not sure about the Desqview, I thought it used QEMM |to emulate LIMS, but I don't have the details fresh in my memory. I don't know how 386/ix runs, but Desqview can use QEMM to emulate LIMS if you like. I am not overly familliar with DV and how it works yet but I figured it was running in the protected mode for a speed increase. I tried running Landmark under DV and here are some results: Results are for an Informtech 386SX/20 motherboard with Pheonix BIOS. Landmark under DOS 3.3 ==> 24.3 Landmark under DV (active window) ==> 30.3 Landmark under DV (background task) ==> 34.0 What I don't understand is the speed increase gained by being a background task. Anybody? Mike Michael W. Picher, / ** ** MicroLab / LexIkon Microsystems Vice President / ** ** 333 Water Street / *** ** Augusta, Maine 04330 Picher@Maine.Bitnet / ** **** (207) 623-4012 / ** Consultants and PC Compatible Manufacturers
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (11/28/89)
In article <2670PICHER@MAINE> PICHER@MAINE.BITNET (Michael W. Picher) writes: | Results are for an Informtech 386SX/20 motherboard with Pheonix BIOS. | | Landmark under DOS 3.3 ==> 24.3 | Landmark under DV (active window) ==> 30.3 | Landmark under DV (background task) ==> 34.0 | | What I don't understand is the speed increase gained by being a | background task. Anybody? Many benchmarks use the timer tick to count "real time." Running in some modes you may lose timer ticks, and therefore appear faster. I'm not sure that this is what's happening with this b.m. but I have seen it with others. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon