home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ukma!darwin.sura.net!gatekeeper.es.dupont.com!pocis.ponca.dupont.com!fieldls.ponca.dupont.com!fieldls
- Newsgroups: comp.os.os2.misc
- Subject: Re: CONTRIBUTE your system video scrolling performance
- Message-ID: <921116132634@fieldls.ponca.dupont.com>
- From: fieldls@ponews.ponca.dupont.com (Lee S. Fields)
- Date: Mon, 16 Nov 1992 13:26:34
- References: <1dmi7iINNrdg@darkstar.UCSC.EDU>
- Nntp-Posting-Host: fieldls.ponca.dupont.com
- NNTP-Software: PC/TCP NNTPLines: 64
- Lines: 64
-
- In article <1dmi7iINNrdg@darkstar.UCSC.EDU> david@maxwell.ucsc.edu (David Darknell) writes:
-
- >> Ever wondered how your OS/2 system compares to others for screen
- >> scrolling, full screen and windowed? Since there is a dearth of
- >> benchmarks that specifically test the capabilities of OS/2 I put
- >> together a screen scrolling benchmark program (in REXX) you can run on
- >> your system!
- >>
- >> Once you run the benchmark, send the results to me
- >> (david@maxwell.ucsc.edu) and I will compile and post the results
- >> back on the net for all to see!
- >>
- >>
- >> *****cut here*****
- >>
- >> /*
- >> ** vs.cmd
- >> **
- >> ** This is a new and revised video speed test
- >> ** that does not rely on the existance of a
- >> ** particular OS/2 README file.
- >> **
- >> ** This file copies 200 carriage return/line feeds
- >> ** to your screen 4 separate times. Each time
- >> ** the elapsed time is calculated, and after the
- >> ** fourth time a running average is calculated.
- >> ** This allegedly eliminates the deleterious effects
- >> ** of cache refill. It also I might add, is MUCH
- >> ** faster with SLOW video systems like mine. The
- >> ** 26 minutes plus copying the OS/2 README file for
- >> ** real slow systems has been a real complaint from my
- >> ** original post.
- >> **
- >> ** email your results to: david@maxwell.ucsc.edu
- >> **
- >> ** Along with the two times, be sure to include:
- >> ** Your CPU Type: (386SX,386DX,486SX,486DX,486DX2,586)
- >> ** Your CPU speed: (in Megahertz)
- >> ** Your BUS type: (ISA,EISA,MCA,LOCAL-VL,LOCAL-PCI,LOCAL-PROPRIETARY)
- >> ** Your BUS clock speed (in Megahertz). This should be the speed of
- >> ** the local bus if you have one. Your graphics card (i.e. tseng 4K,
- >> ** Paradise, ...) and the graphics resolution of your desktop. For
- >> ** example 1024x768x256 or 1024x768x16.
- >> **
- >> ** I will post the results once I have accumulated a good
- >> ** cross section of different system types!
- >> **
- >> ** David Darknell 8 Nov, 1992
- >> */
- >>
-
- [code deleted]
-
- The font and window size should also be reported. On my system, the
- windowed scrolling speed is directly proportional to the amount of
- display space the window occupies. A 25 line window with the 16x8
- font on a 1024x768x256 display is about 6 times slower than a full
- screen session. A 50 line window with the same font runs 11 times
- slower.
-
- If you really want to skew the times, run it in a 2 line window with an
- 8x8 font. Then it runs at close to full screen speed. ;-)
-
-
-