P-II 450 BENCHMARKS TO 527MHz
Thanks for your patience.
Here are the WinBench 98 CPUMark32 results. They were run on an Abit BX6 v2.0 with 64Mb CL2 SDRAM. Adding a second 64Mb lowered performance. A single 128Mb module may match the single 64Mb.
The CL numbers represent selectable settings, not the RAM itself. CAS Latency is the best indicator of performance. CL3 RAM might not run beyond 100MHz. Tcore was typically 2F higher at CL3. Tcore range was 95F - 100F.
(Dual fan HS w/thermal paste, Case fan: intake, PS fan: exhaust.)
The SEL 100/66# settings also are selectable. Abit co-designer Oskar Wu has unofficially disclosed that the Lo setting results in overclocking the CPU's L2 cache.
(MORE RESULTS IN SEPARATE POST)
CPU SPEED Vcore SEL CL3 CL2
450 (4.5x100) 2.0v Hi 1090 1090
450 (4.5x100) 2.0v Lo 1130 1130
504 (4.5x112) 2.2v Hi 1220 1220
504 (4.5x112) 2.2v Lo 1260 1260
527 (4.5x117) 2.3v Hi 1280 1270
527 (4.5x117) 2.3v Lo 1310 1320
[This message has been edited by Roy (edited 02-13-99).]
That is definatly smoking at 527mhz. Nice to see some test figures on PII's, seems like all we see is the Celeron @ 450 tests. Kinda dispells the myth that the Celeron 300A @ 450 is as fast as a true PII 450 (at 450) as your test results are a good 100+ over what I have tested several Celeron (at 450) systems at, including my own.
Interesting that the performance increases so much with a change from hi to low on the sel 100/66. I haven't noticed much difference on the Celerons that I have tested, must be because they already have full speed L2's
Makes me want to go get one also!
Agreed! I just found out about '99 yesterday and have not had time to download it. Even when I do, finding four hours to run the series of 12 tests will take a while.
I'm thinking bout o/cing my PII 450. Im using an Abit bx6-2 and PC100 128mb as one single dimm. Have a globalwin vek12 dual fan attached to the cpu and one Case fan that blows air out. Do you think there's a possiblity to overclock to 527mhz or even (oh my god!) 558mhz? Am I gonna fry anything by trying? As you have tried 527, would you recommend any additional cooling? Also, how bout stability? How long has your PII been running stable @ 527mhz?
Thanks for helping me out. It's real hard to get any decent info on overclocking 450's. You always get replies saying how stupid you are buying a pII instead of a celery. Well, folks here are the benchmarks that prove that PII still kicks!
Thanks for asking.
I say Go for it! Unless you know your RAM is CL2, set it for 3 (try 2 later). It should be OK. I tested some "premium 66" RAM that was tested only to 100. It ran stable at 112, worked at 117, but messed up my W98. Work your way up and watch for glitches.
The case fan probably is at the bottom, the PS fan at the top. Hot air (not heat) rises. If necessary, remount the PS fan to exhaust and the case fan for intake. Thermal paste is important, there should be a thin layer on the entire area of contact.
I've had no problems at 127, and that's running some tough graphics renderings after many hours of cooking along at 98F. If you go to reset the Temp Warning, you'll find that the Abit default is 122F! Tests at 558 would get all the way to the W98 desktop before lockup. I'm going to add some cooling and run some more tests. You'll see it here if I succeed.
hey ! since u guys were talking about benchmark results , i wanna add some question
a celery @464a issit equivalent to a PII450 ?
as Norton benchmark gives only about PII300
how come ? issit like what u guys called as biased towards certain processor ?
I had posted a msg at the main menu there . U guys can look further detail there.
Offhand I would say no.
There are measureable reasons for the difference in price between Pentium IIs (Deschutes core) and Celerons (Mendocino core).
It's important not to compare apples and oranges. DManly's comments imply he used the same benchmark test as I did. If you could get WinBench 98 and run it with your system, then you could compare your results with mine (@450).
Eric points out that DManly and I used an obsolete benchmark. Let's all go to www.pcbenchmarks.com/ and download the new CPUMark99 Then we can all compare oranges.
If all goes well, I'll post the results, probably as a New Topic, sometime Friday or Saturday.
OR MAYBE NOT! INVESTIGATION IN PROGRESS, MORE TO COME!
[This message has been edited by Roy (edited 02-05-99).]
When you get your CPUMark99 results, I would greatly appreciate it if you would come to PC Benchmarks and contribute your results to the database. Many thanks! Oh, and in answer to antidote's question, there is almost no difference between a Celeron and a Pentium II at the same clock speed on most tasks.
Note that CPUMark32 is outdated, and was biased towards some processors. CPUMark99 would be a better comparison for the P2-CeleronA argument in terms of Windows app performance.
Eric, okay, tested my Celeron 300A @ 450 results! (36.7) Seems that I am on the low side at least according to other posts.
And here are some numbers to chew on . . .
winben98 winben99 wintune98
cpu fpu cpu cpu fpu mem
c450a 958 2410 38.3 1320 527 814
p2-450 1120 2290 33.7 1312 526 756
k6-392 940 1310 26.6 921 472 650
We need to find a way to post tables here. Scott???
I have questions. What is your PCI divider setting at that speed? Also, which harddrive are you using? Thanks.
PCI - 1/4x117=29.25
HD - WD 6.4GB
Roy: Why are you running your PCI at 29.25? That slows the system down, why not run it at 39 Mhz and take the little increase in overall speed? We all use to run the buss at 83 mhz and pci at 41.5 so 39 should work fine! Just a thought so I figured I would ask. Jim
New Security Features Planned for Firefox 4
Another Laptop Theft Exposes 21K Patients' Data
Oracle Hits to Road to Pitch Data Center Plans
Microsoft Preps Array of Windows Patches
Microsoft Nears IE9 Beta With Final Preview
Simplified Analytics Improve CRM, BI Tools
Android Passes RIM as Top Mobile OS in 2Q
VMware Updates Hyperic System Management
File Monitoring Key to Enterprise Security
LinkedIn Snaps Up SaaS Player mSpoke