|
|||||||
| Notices |
 |
|
|
LinkBack | Thread Tools |
Jun 25, 2004, 04:00 PM
|
#1 |
|
Styleless Wonder
Join Date: Jun 2002
Location: Ottawa, Ontario
Posts: 6,049
 |
Going to write a Benchmarking FAQ..
I'm itching to write a benchmarking FAQ.. going through the forums and stuff, I've noticed some people don't know how to benchmark properly. (Yes, running it once through doesn't cut it in my opinion).. It'll incorporate some questions and answers about how many times to bench..which benchmarks determine what...and some stability checking stuff. Good idea? Bad idea? I don't want to be wasting my time if it isn't going to be used
 It'll be great if it was a collaboration as well..get some questions for it and some answers.. UPDATE: I hope to put this up in HTML format as well.. As you can see it's very rough right now and no where near completion.  If you have any comments on it, or if you'd like to help out with layout or ANYTHING. Don't hesitate to post or PM me. This is just a sample!Download Last edited by No_Style; Jul 28, 2004 at 10:57 AM. |
|
|
|
Jun 25, 2004, 06:01 PM
|
#2 |
|
Driverheaven brewmaster
Join Date: Oct 2002
Location: British Columbia, Canada
Posts: 4,835
|
It will be used for sure, just make sure to include real world benchies, not just synthetic ones
And yes, you need to run them about 3 times and take the avg. |
|
|
|
|
Jun 26, 2004, 05:42 AM
|
#3 |
|
Mostly lurking lately....
Join Date: Jun 2002
Location: U.S.A.
Posts: 1,983
|
Sounds good to me. I say, Go For IT.
|
|
|
|
|
Jun 26, 2004, 05:54 AM
|
#4 |
|
DH's Latest Mac Convert
|
go for it
|
|
|
|
|
Jun 26, 2004, 06:02 AM
|
#5 |
|
DriverHeaven Extreme Member
Join Date: May 2003
Location: Sweden
Posts: 4,116
|
Yes go for it my man.
|
|
|
|
|
Jun 26, 2004, 10:04 AM
|
#6 |
|
Styleless Wonder
Join Date: Jun 2002
Location: Ottawa, Ontario
Posts: 6,049
|
Thanks for the vote of confidence
 I'll be working on it over the next week or so.... Some things came up and need my immediate attention |
|
|
|
|
Jun 26, 2004, 10:09 AM
|
#7 |
|
DH's Latest Mac Convert
|
write in when you are ready, i would like to read this
|
|
|
|
|
Jun 26, 2004, 11:47 AM
|
#8 |
|
DH's Dormant Dragon
|
write it up... there are alot of people that would like to know
|
|
|
|
Jun 26, 2004, 11:57 AM
|
#9 |
|
ein Krieger
Join Date: Jan 2004
Location: Clemson Uni
Posts: 3,127
|
good idea...i think an OC faq would be more benefitial to some people though...(lol does one already exist?)
|
|
|
|
|
Jun 27, 2004, 12:22 AM
|
#10 | |
|
DriverHeaven Senior Member
Join Date: Aug 2003
Location: Not Vegas, anymore :(
Posts: 1,370
|
Quote:
|
|
|
|
|
|
Jun 27, 2004, 04:17 AM
|
#11 |
|
DH's Dormant Dragon
|
indeed... i try to be as thorough as possible.. reason it's taking me AGES to get done this bloody major review i've been working on for months.... hours....
|
|
|
|
|
Jul 3, 2004, 09:04 PM
|
#12 |
|
Styleless Wonder
Join Date: Jun 2002
Location: Ottawa, Ontario
Posts: 6,049
|
Nearly done a rough draft..
I'm nearly done a rough draft of my FAQ/Guide..
I'd appreciate some help on it though so.. if anybody wants to proofread, add suggestions, help design or contribute.. I'll accept the first two people to post in this thread with their intention to help. I want to cover a lot in this and I want it to go right as well. There will be revisions made for sure, and I hope to update it on a regular basis or as needed. So if you're knowledgeable in benchmarking and it's practical uses besides just "measuring performance".. don't hesitate to post Last edited by No_Style; Jul 3, 2004 at 09:15 PM. |
|
|
|
|
Jul 3, 2004, 09:06 PM
|
#13 |
|
Mr. Nobody
Join Date: Jul 2003
Location: OmniPresent Nightwatcher
Posts: 5,933
|
Its a very good idea bud, go for it!!!
|
|
|
|
|
Jul 3, 2004, 10:00 PM
|
#14 |
|
Roxy Music
|
Re: Nearly done a rough draft..
I think you may be going down the right track based on your comments.
The best use for benchmarking a system is to know that something has changed over time. CPU marks go down could be a sign of services running that you are not aware of, maybe a virus. Knowing the performace gain when adding a new video card can only be achived by having good data before the upgrade. There is also the stability factor. If your system is having problems, random crashes etc... Running benchmarks may help to pin-point the problem. OpenGL benchmarks are fine but DirectX benchmarks crash. Since benchmarks stress the system they are a valuable tool beyond my system get 10FPS more than someone else. Network benchmarks are not covered much in any guide. Using ttcp for example is an easy way to see how well an internal home network is performing. That kind of information is very useful. Greg QUOTE]Originally posted by No_Style I'm nearly done a rough draft of my FAQ/Guide.. I'd appreciate some help on it though so.. if anybody wants to proofread, add suggestions, help design or contribute.. I'll accept the first two people to post in this thread with their intention to help. I want to cover a lot in this and I want it to go right as well. There will be revisions made for sure, and I hope to update it on a regular basis or as needed. So if you're knowledgeable in benchmarking and it's practical uses besides just "measuring performance".. don't hesitate to post [/quote] |
|
|
|
|
Jul 28, 2004, 10:55 AM
|
#15 |
|
Styleless Wonder
Join Date: Jun 2002
Location: Ottawa, Ontario
Posts: 6,049
|
This is a rough, very rough draft.
I've got more work to do on it, but I thought I'd post what I have so far for people to see.
I hope to put this up in HTML format as well.. As you can see it's very rough right now and no where near completion. If you have any comments on it, or if you'd like to help out with layout or ANYTHING. Don't hesitate to post or PM me. Download |
|
|
|
