Wintune, Sandra or Winbench?

[daveleau]

I have used both Sandra and Wintune in the past few days to benchmark the changes in my system and am getting mixed reviews on each program. What do you think is best?

I have heard that Sandra's memory tests do not perform operations that the system would normally do and thus are not reliable.
I have heard that Wintune is not a good CPU bench and gets erratic readings when using Rain.
I haven't used or heard of any problems with Winbench but haven't heard of anyone using it either. I am currently downloading it.

Thanks for your input-
Dave

[bigslammer]

winbench is a solid bechmark. it's just a large file though

[pickel]

Wintune 98 seemed to work better for me than Sandra, but then I don't have a big hotrod
rig. It was simpler and easier to use but probably not as comprehensive.

Ed_S, I got totally different results between
Wintune 97 and Wintune 98.
Wintune 97 Wintune 98
clock 689 351
you had mentioned about the reading being
doubled in w 97
667 MIPS 865MIPS
202 MFLOPS 404MFLOPS
cashed sp 49Mbs 57Mbs
Etc.
How come such a variation? Got any ideas?

the pickel


[This message has been edited by pickel (edited 11-03-1999).]

[AuraEdge]

Wintune is not easy to read results and compare to eithers.
For Winbench, the CPUmarks is very popular and easy to read. The most popular and compared is Sandra.
I probably will use sandra the most, winbench second, and wintune last (if ever)

[Ed_S]

I still prefer Wintune 97 for this since it allows you to save each test run to a database, with full details. Wintune 98 doesn't give enough detailed info, and Sandra makes you look at separate pages for everything.(unless there's a way I haven't seen...)

Anyway, I like to be able to easily see & compare several tests of cpu, video, hd, and memory all on one page when making changes. This is the only benchmark program I've found that does this.

Others probably do more extensive testing, though.