home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.ibm.pc.hardware
- Path: sparky!uunet!cs.utexas.edu!swrinde!gatech!concert!uvaarpa!murdoch!galen.med.Virginia.EDU!hh7r
- From: hh7r@galen.med.Virginia.EDU (Hsiang-Hsin Hsiung)
- Subject: Performance boost or woe ? (ABOUT BUS CLOCK SPEED)
- Message-ID: <1993Jan8.182606.21215@murdoch.acc.Virginia.EDU>
- Sender: usenet@murdoch.acc.Virginia.EDU
- Organization: University of Virginia
- References: <1993Jan7.194352.20642@murdoch.acc.Virginia.EDU>
- Distribution: usa
- Date: Fri, 8 Jan 1993 18:26:06 GMT
- Lines: 44
-
- I posted a article "Is there a program to test BUS CLOCK SPEED?"
- yesterday and got some reponses through e-mail. For those of you who are
- interested, I have not received any hint about "a program", nor have I solved
- the puzzle about the "2" in CLK2I/5. There are some interest comments about
- the bus speed though. It seems that some people believe an increased bus
- speed will not improve the overall performance too much but only cause trouble.
- So, I did some benchmark tests to see if there are performance improvements.
-
- Test machine: 486DX/50, 256k cache, 8MB RAM.
- Peripherals : 16-bit IDE controller card, Seagate 3283A HD,
- Diamond Stealth vidoe card.
- Test speed : 8.33Mhz, 10Mhz, 12.5Mhz.
-
- Test results:
-
- 1. EGA/VGA Text Mode Tests (PC Magazine Labs Benchmark 6.0):
-
- i) 10Mhz : 35% faster than 8.33Mhz in average.
- ii) 12.5Mhz : 61% faster than 8.33Mhz in average.
-
- 2. EGA/VGA Graphics Mode Tests (PC Magazine Labs Benchmark 6.0):
-
- i) 10Mhz : 15% faster than 8.33Mhz in average.
- ii) 12.5Mhz : 23% faster than 8.33Mhz in average.
-
- 3. Graphics Winmark (PC Magazine Labs WinBench 3.1):
-
- i) 10Mhz : 14% faster than 8.33Mhz in average.
- ii) 12.5Mhz : 23% faster than 8.33Mhz in average.
-
- 4. Hard disk maximum thruput (Comptest 2.57):
-
- i) 10Mhz : 16% faster than 8.33Mhz.
- ii) 12.5Mhz : 36% faster than 8.33Mhz.
-
- 5. Hard disk data transfer rate (Norton Utilities 6.0):
-
- i) 10Mhz : 18% faster than 8.33Mhz.
- ii) 12.5Mhz : 23% faster than 8.33Mhz.
-
- I understand there are potential problems associated with the
- improvements on system performance. But, what are these problems? Does
- anybody have experiments against the "significant" performance boost?
- Thanks.
-