Thread View: alt.games.doom
5 messages
5 total messages
Started by at... John C.Kahoon
Wed, 19 Jan 1994 11:02
Re: Math Co-Processor and Doom
Author: at... John C.Kahoon
Date: Wed, 19 Jan 1994 11:02
Date: Wed, 19 Jan 1994 11:02
7 lines
895 bytes
895 bytes
The shareware version of doom does NOT use an FPU .The registerd version, well I'm not sure but, I think the only reason it runs slowly is because it's ver. 1.1 .I do know that if you have less than 8 meg of ram it won't run to good. Any way, I'm running a 486 33sx w/8meg (no fpu) and I'm real happy with running a full screen at low res. It looks like I'm behind the DOOM cam.
Re: Math Co-Processor and Doom
Author: sco... B. Scott Bilas
Date: Wed, 19 Jan 1994 14:57
Date: Wed, 19 Jan 1994 14:57
55 lines
3581 bytes
3581 bytes
In article <2hjn87$2...@vixen.cso.uiuc.edu> me...@uxa.cso.uiuc.edu (One Lain Meyer) writes: >From: me...@uxa.cso.uiuc.edu (One Lain Meyer) >Subject: Re: Math Co-Processor and Doom >Date: 19 Jan 1994 16:31:03 GMT >at...@cleveland.Freenet.Edu (John C.Kahoon) writes: >>The shareware version of doom does NOT use an FPU .The registerd version, >>well I'm not sure but, I think the only reason it runs slowly is because >>it's ver. 1.1 .I do know that if you have less than 8 meg of ram it won't >>run to good. Any way, I'm running a 486 33sx w/8meg (no fpu) and I'm real >>happy with running a full screen at low res. It looks like I'm behind the >>DOOM cam. I do the same -- full screen low res. And I've got a 33DX. It's mostly bearable unless I get into some heavy action in multiplayer mode. But as for the FPU use, I can't imagine why anybody would use floating point graphics routines. Judging from the speed of Doom I'd bet that it's mostly 32-bit integer math. Take a look at any version of Fractint for DOS for a good example of the speed difference. That's pure integer math (up to a certain resolution). And it's faster than any floating point fractal programs. However, if Doom truly does use the FPU, I'd love to see it run on a Pentium system. The pipelined Pentium FPU is supposed to be 4 times faster than a 486-66! >i'm playing the shareware version 1.1 on a 386 DX 40 with only 4 meg. i'm >using the screen size 2 smaller than full size with minimum detail... >works just wonderful...no choppiness what-so-ever. and yes, i have seen it >played on a 486 DX whatever, and the playability/smoothness doesn't suffer >much at all... I don't know, I think there's a big difference. You should see my friend's DX66 with local bus ATI GUP card compared to my 33 with non-LB Speedstar Pro. Talk about smooth... >oh and also...you haven't played DOOM until you've played it with a Gravis >pumping out 4 digital channels (plus the music) into a 200 watts/channel amp >driving a pair of speakers with 15 inch ported woofers. crank it all the way up >and you feel DOOM. guess it would be like playing it out in california. =) >check it out...you'll like it. Done! Nothing beats a Gravis (or, I suppose, a Roland + a PAS-16 or SB16). Most of the time (when the neighbors yell), what works beautifully is plugging some good monitor headphones into the amp output of the Gravis. Doom isn't supposed to support 3D sound for the Gravis, but it sure _sounds_ that way. When an imp throws a fireball at you, you can almost hear the thing whizz by your head and hit the wall behind you. By the way, try it with a surround processor. It's weird. >-OLM ___________________________________________________________________ Scott Bilas | Beware of Wild Bill and his SPA Nazis! Terris Engineering Company | sco...@iastate.edu
Re: Math Co-Processor and Doom
Author: me... One Lain Meyer
Date: Wed, 19 Jan 1994 16:31
Date: Wed, 19 Jan 1994 16:31
23 lines
1637 bytes
1637 bytes
at...@cleveland.Freenet.Edu (John C.Kahoon) writes: >The shareware version of doom does NOT use an FPU .The registerd version, >well I'm not sure but, I think the only reason it runs slowly is because >it's ver. 1.1 .I do know that if you have less than 8 meg of ram it won't >run to good. Any way, I'm running a 486 33sx w/8meg (no fpu) and I'm real >happy with running a full screen at low res. It looks like I'm behind the >DOOM cam. i'm playing the shareware version 1.1 on a 386 DX 40 with only 4 meg. i'm using the screen size 2 smaller than full size with minimum detail... works just wonderful...no choppiness what-so-ever. and yes, i have seen it played on a 486 DX whatever, and the playability/smoothness doesn't suffer much at all... oh and also...you haven't played DOOM until you've played it with a Gravis pumping out 4 digital channels (plus the music) into a 200 watts/channel amp driving a pair of speakers with 15 inch ported woofers. crank it all the way up and you feel DOOM. guess it would be like playing it out in california. =) check it out...you'll like it. -OLM
Re: Math Co-Processor and Doom
Author: haym... Ed Haymore
Date: Thu, 20 Jan 1994 16:20
Date: Thu, 20 Jan 1994 16:20
26 lines
2085 bytes
2085 bytes
B. Scott Bilas (sco...@iastate.edu) wrote: | I do the same -- full screen low res. And I've got a 33DX. It's mostly | bearable unless I get into some heavy action in multiplayer mode. But as for | the FPU use, I can't imagine why anybody would use floating point graphics | routines. Judging from the speed of Doom I'd bet that it's mostly 32-bit | integer math. Take a look at any version of Fractint for DOS for a good | example of the speed difference. That's pure integer math (up to a certain | resolution). And it's faster than any floating point fractal programs. | However, if Doom truly does use the FPU, I'd love to see it run on a Pentium | system. The pipelined Pentium FPU is supposed to be 4 times faster than a | 486-66! On my 486DX, there isn't all that much difference in calculation time between integer versus floating point on Fractint; perhaps a 25% or so slowdown. If you're doing floating point emulation in software, I agree you'll see a very significant slowdown -- but having the FP hardware helps out a lot. (And Intel even admits that the floating point unit on the 486's is "dog meat".) BTW, an earlier poster in this thread made reference to the size of the 486 internal cache being 1k. It's actually 8k -- although don't the Cyrix 486's have a smaller cache? -- Ed Haymore | AA6EJ e...@byu.edu | Es perillos abocar-se.
Re: Math Co-Processor and Doom
Author: mke...
Date: Fri, 21 Jan 1994 00:08
Date: Fri, 21 Jan 1994 00:08
21 lines
1542 bytes
1542 bytes
My brother has a 386DX-40 with the Co-Processor and I bumped it two notches with low detail, but it _does_ got dog slow on Episode One, Level Three, and Ultra- violence. When open the door where you see the health bonus, it just drops drastically!In article <2hjn87$2...@vixen.cso.uiuc.edu>, me...@uxa.cso.uiuc.edu (One Lain Meyer) writes: |> at...@cleveland.Freenet.Edu (John C.Kahoon) writes: |> [stuff deleted] |> |> i'm playing the shareware version 1.1 on a 386 DX 40 with only 4 meg. i'm |> using the screen size 2 smaller than full size with minimum detail... |> works just wonderful...no choppiness what-so-ever. and yes, i have seen it |> played on a 486 DX whatever, and the playability/smoothness doesn't suffer |> much at all... |> I don't know how come your 386DX-40 doesn't get bogged down! -Michael Doom for the Apple //e (a novelty item), how about it ID?
Thread Navigation
This is a paginated view of messages in the thread with full content displayed inline.
Messages are displayed in chronological order, with the original post highlighted in green.
Use pagination controls to navigate through all messages in large threads.
Back to All Threads