smsmith@hpuxa.ircc.ohio-state.edu (Stephen M. Smith) (03/19/91)
OK, all you QEMM gurus, here's one for you:
I'm trying to get the performance out of my ProDesigner II VGA
card that I would like. If I load my system without QEMM,
and include the line DEVICE=C:\VGA\SPEED_UP.SYS in my config.sys
file (SPEED_UP.SYS is the utility for video BIOS shadowing which came
with my Orchid VGA card), I then can get an unbelievable 31,000+ CPS
rating on my video card (tested with QaPlus diagnostics).
Problem: If I load QEMM, I can only get HALF that speed. What's
going on?
Things I've tried:
1) Let QEMM shadow the BIOS with the ROM parameter. Doesn't help.
2) Load SPEED_UP.SYS in high memory with QEMM. Doesn't help.
3) Load QEMM with the no video ram option, then load SPEED_UP.SYS
into high memory. System hangs.
4) Load SPEED_UP.SYS in low memory before starting QEMM in my
config.sys file. Still doesn't help. The only difference is
now I've got 32k less for DOS applications.
Summary:
In other words, I can only get 31,000+ CPS speed if I use NO
EMM (I've tried it with Windows too and it didn't help) and only
if I use SPEED_UP.SYS in low memory.
I'm wondering if for some reason the QaPlus diagnostics don't test
correctly under various conditions, and that my test results are
worthless. I'm also wondering is other things are affecting the
results, like my 64k memory cache. Should my tests be run with
the memory cache off? (It's horribly slow if I do!!)
Can somebody help me?
Stephen M. Smith \ + /
<smsmith@hpuxa. \+++++/ " #*&<-[89s]*(k#$@-_=//a2$]'+=.(2_&*%>,,@
ircc.ohio-state. \ + / {7%*@,..":27g)-=,#*:.#,/6&1*.4-,l@#9:-) "
edu> \ + /
BTW, WYSInaWYG \ + / --witty.saying.ARCphil@brahms.amd.com (Phil Ngai) (03/20/91)
smsmith@hpuxa.ircc.ohio-state.edu (Stephen M. Smith) writes: >with my Orchid VGA card), I then can get an unbelievable 31,000+ CPS >rating on my video card (tested with QaPlus diagnostics). Gosh, you must be a fast reader. What's the point of all this? -- The Macintosh makes it easy to do sloppy work.
rdippold@gdansk.qualcomm.com (Ron Dippold) (03/22/91)
In article <1991Mar19.205736.17884@amd.com> phil@brahms.amd.com (Phil Ngai) writes: >smsmith@hpuxa.ircc.ohio-state.edu (Stephen M. Smith) writes: >>with my Orchid VGA card), I then can get an unbelievable 31,000+ CPS >>rating on my video card (tested with QaPlus diagnostics). > >Gosh, you must be a fast reader. >What's the point of all this? It means that in actual graphics modes that graphics will go much faster than they would on a machine with an average VGA card. Less waiting around for graphs, etc. to draw.
phil@brahms.amd.com (Phil Ngai) (03/23/91)
rdippold@gdansk.qualcomm.com (Ron Dippold) writes: >In article <1991Mar19.205736.17884@amd.com> phil@brahms.amd.com (Phil Ngai) writes: >>smsmith@hpuxa.ircc.ohio-state.edu (Stephen M. Smith) writes: >>>with my Orchid VGA card), I then can get an unbelievable 31,000+ CPS >>>rating on my video card (tested with QaPlus diagnostics). >> >>Gosh, you must be a fast reader. >>What's the point of all this? >It means that in actual graphics modes that graphics will go much faster than >they would on a machine with an average VGA card. Less waiting around for >graphs, etc. to draw. Not at all. There is little to no relationship between graphics mode speed and text mode speed. How long would it take to write a 640x480x256 screen at 31,000 bytes/second? I think I figured about 10 seconds. 20 seconds at 15,000 bytes/second. Hm. -- Help! I just got a Macintosh. Anyone got a magnifying glass?
rdippold@cancun.qualcomm.com (Ron Dippold) (03/26/91)
In article <1991Mar22.181213.11232@amd.com> phil@brahms.amd.com (Phil Ngai) writes: >rdippold@gdansk.qualcomm.com (Ron Dippold) writes: > >>In article <1991Mar19.205736.17884@amd.com> phil@brahms.amd.com (Phil Ngai) writes: >>>smsmith@hpuxa.ircc.ohio-state.edu (Stephen M. Smith) writes: >>>>with my Orchid VGA card), I then can get an unbelievable 31,000+ CPS >>>>rating on my video card (tested with QaPlus diagnostics). >>> >>>Gosh, you must be a fast reader. >>>What's the point of all this? > >>It means that in actual graphics modes that graphics will go much faster than >>they would on a machine with an average VGA card. Less waiting around for >>graphs, etc. to draw. > >Not at all. There is little to no relationship between graphics mode >speed and text mode speed. > >How long would it take to write a 640x480x256 screen at 31,000 bytes/second? >I think I figured about 10 seconds. 20 seconds at 15,000 bytes/second. I'm not saying that the two speeds are the same, but that the text speed of a card is usually a good indication of the graphics speed of the card. They give you some idea of how fast you can move data to the card, how many video wait states you have to put up with, etc. If he gets a blazingly fast text speed, it's a good indication that the graphics speed will also be superior.