PDA

View Full Version : [Release] A64 Overclocking Helper X5.0


AgentGOD
01-09-2006, 06:34 PM
------------------------------------------
Athlon 64 Overclocking Helper Version X5.0
------------------------------------------
Created by Ray (AgentGOD)

Written in C++

Purpose: To display all the possible combinations of clock speeds the user enters.
It is designed to be very flexible and displays a chart of settings.

Homepage:
www.ultimate-filez.com


Program in usage:
http://img344.imageshack.us/img344/3856/a64helperoc1ov.jpg

I'm just trying to help out the community, and I also made this for my own usage. This program should save alot of time and energy looking through all the combinations, as it displays all the possible combinations in the range.

Suggestions/Bug Issues for the program may be posted here, and may be updated into the next version.
Good luck overclocking, and have fun.

Download Link:
http://www.ultimate-filez.com/A64_OC_Help.zip

Updated to X5.0
Changes:
-Refined some code
-Added output to text

Updated to X4.0
Changes:
-More user friendly
-More settings
-Now supports overclock range
-More accurate memory dividers

Frequently Asked Questions
Q: What is this 'HTT' business about?
A: It is AMD's HyperTransport bus. the HTT is the external clock on new AMD processors (AMD64) and replaces the traditional FSB.

Q: What is a memory divider?
A: It takes the HTT clock and divides it accordingly to the divider

STickGuy
01-09-2006, 11:15 PM
One good program here

before
01-09-2006, 11:44 PM
Nice prog mate... ;)

metro.cl
01-10-2006, 12:07 AM
nice one i'll try this one, maybe you could do something similar with vcore, like volts close to 1.6v and the prog will list the closests ones

before
01-10-2006, 12:18 AM
I'm seconding metro on his request...

Evil_Spork
01-10-2006, 05:56 AM
awesome. tried it out real fast.

could you make a version for us dothan people also?
we can use this one buts its a little bit harder.

also feedback: get rid of half multi's. they dont work right on A64s so they shouldnt be used/suggested. ;)

5days
01-10-2006, 06:43 AM
also feedback: get rid of half multi's. they dont work right on A64s so they shouldnt be used/suggested. ;)
I've always wanted to know what people meant by 'not work right'?

Are you talking about the divider issues or something for complex or spanning?

DesertShooter
01-10-2006, 10:08 AM
Nice tool :)

AgentGOD
01-10-2006, 11:49 AM
Thank you all for your support :clap:

shimmishim
01-10-2006, 04:03 PM
can you add mem divider option for ddr360 or 180 MHz for those with DFI boards for the next revision?

Merconium
01-10-2006, 07:55 PM
Thanks!

Evil_Spork
01-12-2006, 06:11 AM
I've always wanted to know what people meant by 'not work right'?

Are you talking about the divider issues or something for complex or spanning?
10.5x200 will result in a memory clock of 190mhz

this picture says it all.
http://evilspork.net/es/compstuff/A64Divider.jpg

ozzimark
01-12-2006, 06:30 AM
10.5x200 will result in a memory clock of 190mhz
that's more of a design feature than anything, due to the way the memory dividers from the cpu are set up. once you get over the fact that it rounds up, decreasing memory speed in some cases (which are 100% predictable) you'll notice that there is nothing really wrong with the 1/2 multi's

5days
01-12-2006, 06:31 AM
10.5x200 will result in a memory clock of 190mhz

this picture says it all.
http://evilspork.net/es/compstuff/A64Divider.jpg
Okay your statement of 'not working' is a bit misleading.

Its just a bad code in the CPU.

The OP's program CAN be programmed to predict these so called 'errors' though.

HARDCORECLOCKER
01-12-2006, 06:48 AM
:D AGENT, U are a real gaaaaaaaaangsta - this program rocks !!! :clap:

:toast:

periquitos
01-12-2006, 11:11 AM
very nice tool!!!

Evil_Spork
01-12-2006, 02:39 PM
that's more of a design feature than anything, due to the way the memory dividers from the cpu are set up. once you get over the fact that it rounds up, decreasing memory speed in some cases (which are 100% predictable) you'll notice that there is nothing really wrong with the 1/2 multi's
yeah.. they just dont give the same mem speed one would expect, thus are not the best choice. i think they might read wrong in cpuz and such too.. i dont know this for sure though

Okay your statement of 'not working' is a bit misleading.

Its just a bad code in the CPU.

The OP's program CAN be programmed to predict these so called 'errors' though.
it can be programmed for it. its just a bit of a pain in the ass to program though. but it can be done.

MaxxxRacer
01-12-2006, 02:42 PM
AgentGOD,

Any chance that you could give us the 183 and other dividers that the DFI nf4 has?

RichUK
01-12-2006, 03:23 PM
Nice Work!!!

butter_fry
01-12-2006, 09:01 PM
the dividers you've provided are good for most boards. But indeed, I'd like to see the 140, 150, 183 included in there.

chinkgai
01-12-2006, 11:58 PM
0mgOmg!! i've been inf3cted!!! dont d/l it!!! vi12us!!!!







jk

teh program is teh l33t. nice job agent.

rick_fx
01-13-2006, 12:01 AM
Handy little program :thumbsup:

SET
01-13-2006, 12:09 AM
Thanks AgentGOD

Great contribution :woot: