--------------------------------------- SL> No, that's a jab at what he said. Processors are designed to run at a SL> specific speed. If they can't handle that speed, THEN they are underclocked SL> until they are reliable. SL> SL> That other guy just said they are all made the same and then tested down to SL> their reliable speed. SL> SL> My CPU runs perfectly at 120 Mhz (it's marked 100) under everything xcept SL> OS2. Hmm.. How can you say it "runs perfectly" when it fails when overclocked running OS/2? Doesn't this say more about the way each of these operating systems are using the hardware rather than the hardware capabilities itself? Are you sure I (or you) couldn't design a DOS based "test" that would show the same kinds of "failures" when it is overclocked? My aunt used to have a car that "ran perfectly" on only 7 out of 8 cylinders too. Even after I "fixed" it (you should have seen the shape of ALL her plugs) she didn't notice any "improvement", but I sure did. Denis All opinions are my very own, IBM has no claim upon them . . . --- Maximus/2 3.01 * Origin: T-Board - (604) 277-4574 (1:153/908) --------------- FIDO MESSAGE AREA==> TOPIC: 129 OP. SYS DEBATE Ref: E2L00118Date: 02/16/97 From: DENIS TONN Time: 10:15am \/To: SCOTT LITTLE (Read 3 times) Subj: win no operarting sy Original from Scott Little to Rich Veraa on 02-14-1997 Original Subject: win no operarting sy --------------------------------------- RV> What benefit would a 32 bit CHKDSK confer that's not provided in the RV> 16 bit version? SL> Speed? Porting the current 16 bit code to 32 bit would make very little difference in speed. It needs a redesign and rewrite. A redesign would make a LOT more difference even if it stayed 16 bit. The rewrite could be done in 32 bit, but is not the primary factor in speed differences. 32 bit != faster than 16 bit in all cases. Denis All opinions are my very own, IBM has no claim upon them . . . --- Maximus/2 3.01 * Origin: T-Board - (604) 277-4574 (1:153/908) --------------- FIDO MESSAGE AREA==> TOPIC: 129 OP. SYS DEBATE Ref: E2L00119Date: 02/16/97 From: DENIS TONN Time: 10:29am \/To: SCOTT LITTLE (Read 3 times) Subj: Test Results Original from Scott Little to Mike Ruskai on 02-14-1997 Original Subject: Test Results --------------------------------------- MR> Superior in that it is both faster and more reliable. I guarantee MR> that if I were to install and configure OS/2, it would be faster than MR> Win95 on your system. SL> Ok. When I get your optimisation recommendations, I want to to dial SL> [intl code] + 61-2-9534-1702. Page me. I will put the test machine online SL> under OS/2. You can then call that and test it. Then I will change it to SL> Win95, and you can test that. Trust me, win95 is faster, esp. on screen SL> writes. Let me get this straight. You mean you judged the "speed" of the BBS based on LOCAL SCREEN WRITES? There is something definately wrong with this methodolgy. Perhaps you will want to rerun the "tests" using a proper methodolgy. Denis All opinions are my very own, IBM has no claim upon them . . . --- Maximus/2 3.01 * Origin: T-Board - (604) 277-4574 (1:153/908) --------------- FIDO MESSAGE AREA==> TOPIC: 129 OP. SYS DEBATE Ref: E2L00120Date: 02/16/97 From: DENIS TONN Time: 12:06pm \/To: ELLIOTT GOODMAN (Read 3 times) Subj: Hot Keys Original from Elliott Goodman to Denis Tonn on 02-14-1997 Original Subject: Hot Keys --------------------------------------- EG> BTW, a separate question, while I got your ear: running on my laptop at EG> 640 x 480 x 16, whenever I open up a windowed program, like BlueWave for EG> instance, it always opens offset so I have to drag it up or down and off EG> to the side so I can see the whole window. If I use the Shift-maximize EG> option (I think it is) to have all windows open maximized, they all EG> start centered but at the top. Any way of having windows open centered EG> side to side but not at the top of the screen? No. Size is the only thing you can set for the "default" opening of a windowed session. The location is still under system (and application) control. It *is* possible for OS/2 windowed sessions to control the location of the window. Setting the default to "maximized" will always place the window in the upper left corner of the screen. 4OS2 has some additional options on the "start" command that will allow you to select the x,y position of the window. This might solve your query. Denis All opinions are my very own, IBM has no claim upon them . . --- Maximus/2 3.01 * Origin: T-Board - (604) 277-4574 (1:153/908) --------------- FIDO MESSAGE AREA==> TOPIC: 129 OP. SYS DEBATE Ref: E2L00121Date: 02/16/97 From: DENIS TONN Time: 12:33pm \/To: SCOTT LITTLE (Read 3 times) Subj: Test Results Original from Scott Little to Mike Ruskai on 02-14-1997 Original Subject: Test Results --------------------------------------- MR> It would seem that Win95 is much more forgiving of hardware than OS/2 MR> (or WinNT, Unix, etc.). SL> Linux worked. I was compiling kernels and running FTP servers and all sorts SL> of things. OS2 didn't even boot properly. SL> It worked OK fully tweaked under Linux (as well as Win95). I haven't SL> tried NT on it. They both worked like a charm. OS2 didn't. MR> That doesn't change the fact that OS/2 didn't fail - the hardware did. SL> OS2 caused the hardware to fail. If OS2 hadn't tried to be smart the hardware SL> would be working fine. You mean that your hardware has worked *SO FAR*, don't you? I'll bet if you install a new app (or driver) under Win95 that then "fails" you will want to blame the new code too, right? I fail to understand why you want to lay the blame for this on OS/2's doorstep. It is clearly a hardware problem brought on by your attempts at overclocking. Any competent technician or engineer will tell you so. If you don't want to believe me, head over to the DR_DEBUG echo or the various Intel related ones and express your idea that an Operating System is at fault when it doesn't "handle" your attempts at overclocking the CPU and system. Let me know exactly what the "OS" (or application or driver) shouldn't "do" on a system that is overclocked, while working fine on the same system running at it's rated speed. Oh, and just to make your search easier, I know of a couple of things that device drivers need to do when running on faster (or slower) CPU's, but these particular areas are handled fine on OS/2. If they were not, OS/2 couldn't run on a 200mz PentiumPro which it obviously does just fine. You can eliminate any consideration of processor speed related I/O timing loops in your search. As a final consideration, I should tell you that Bill Wolff had the same "complaint" about OS/2, to wit: that it somehow used improper "instruction sequences" that "caused" marginal hardware to fail. I asked him back then to supply examples, and he has yet to do so. Maybe you should get together with him and try to find possible "bad instruction sequences" that an Operating System should not use instead of just commiserating with him on the "BS" in my postings. Denis All opinions are my very own, IBM has no claim upon them . . . --- Maximus/2 3.01 * Origin: T-Board - (604) 277-4574 (1:153/908) --------------- FIDO MESSAGE AREA==> TOPIC: 129 OP. SYS DEBATE Ref: E2M00000Date: 02/14/97 From: LAWRENCE LUCIER Time: 02:33pm \/To: MIKE ZELESKI (Read 3 times) Subj: CLI MZ> Problem is Bill; You should not even try to use HPFS with an 8 meg system. Ran Warp 3 on a 386/40 with 8 meg plus combination of 1 FAT and 4 HPFS partitions without any problems for close to 2 years. In fact, the only time I have had a disk problem was when I first installed warp (toasted the FAT partition) but I suspect the problem was due more to my own ineptitude than the OS itself! --- Sqed/32 1.13b2/r15155 * Origin: T-Shirts 'N Genes BBS - (250)748-3408 v32b v42b XA CM (1:340/204)