Liquid3D

I beeaving problems lately with every system I overcloci. Pushing these rigs is continaully corrupting HDD's and creating delays to my tests and reaking gneral havoc in my lfe.

I thought the issu of orruption was resolved with locked buses? What happend to this? Wasn't there a time when we didn't have to worry about HD corruption because the 33/66 buses were locked? Along comes SATA and that goes out the window?

For my recet horro show the Gigabyte P965-DS3 (Rev.1) the system began with the oddest problems, I would power it up and experience rapid on/off states as th system seemed to have some interal struggle with BIOS settings.

Al I woul do is change th setting to manual, not overclok anaything just change the settinsg to manual, such as PCI 100MHz, GFSB 266, memory 1067MHz etc. First I experinced video corruption which led me to belaive I damged my 7950GX2 re-intsalling it? Or perhaps i wasn't seated corectly, I also though the external; 12V 6-pin wass sharted because hsi screen looked like damaged hardware voltage related.

Weirdest thing I upped the PCIe by 0.01V and it seemed to resolve this, until the next time.

Now it's trying to run ANY kit of memory I have at 1067MHz while running the CPU at 266FSB. I wouldn't mind if it were 450FSB but this is simpyl insane. Of course I only hav one kit of offical 1067MHz memory and obe stick fried from a bad BIOS (903) from the Asus M2N32-SLI board. The fact this BIOS has damged systems is documented.

The most intriguing part is that memory run under that BIOS went on to damag other systems. Most people say this is impsible, well tell that to m buddy who ha the memory and had to RMA his mobo, and he knows as muh about memeory as anyone. It also wouldn't boot in any of his rigs and wouldn't pss mem-test yet when I checked it it would pass mem-test on my Inte system but not on my AM2 system Mystery deepens.

ANyway back to the Gigabyte P965-DS3 (Rev.1) I've been Googling the proble for two days and have read all sorts of stuff. or example some people experinced the rapid on/off power-up phenomenon and found removing the 4-pin power connector allowed them to run the system?

I'm using BIOS F7 I tried F12 and F7 seems more stable.

jmke

Overclocking not working can be real frustrating, I share your pain

__________________

geoffrey

I can't give you an answer on the questions you've asked here, I ran into mutch issue's aswell with multiple mainboards and even by understanding which part is causings issue's I couldn't resolve the problem I was being confronted with.
One question that really sticks to my mind is... how trusty is todays software? By sofware I mean the procedure that is being followed everytime your system is being powered up. If power-good, then reset this or goto register x and ...
I'm no engineer, though how is it possible that a board is not booting after CMOS clear. How hard is it to program a hardware reset? How hard is it to really lock a bus at 100MHz if it is being requested, how hard is it to set the Command Rate to 1T if your chipset supports it. How hard is it to just make shure every tiny part in the Basic Input Output System is working like it should???

For todays mainboards it seems like 'perfect' is either a dream or a marketing slogan, we just have to stick to decent hardware which all come with their own kind off problems. Overclocking has been made easier with all options we find in todays BIOS's, though 1 part I really want to see improve with newer boards is proper functions of each setting on each board.

Sidney

Simply put, overclocking is all about hardware and not the operators. It is all done by engineers behind the scene. AMD, Intel, Nvda will not have to limit what their products could deliver if they sell overclocked products.

Car manufacturers will be selling cars that could do 160MPH at a great low price if they don't have to provide warranty and prevent lawsuits when brakes and gearbox failed, engines overheat, fan belt melting away, engines die within days ..... etc.

There is more pain in overclocking than pleasure; it is not what computing is all about to begin with.

However, I wouldn't mind overclocking the bandwidth from my DSL or making the air "thicker" to speed up Wifi or my bluetooth .... can it be done?

__________________
lazyman

Opteron 165 (2) @2.85 1.42 vcore AMD Stock HSF + Chill Vent II

jmke

Yes
http://icrontic.com/forum/showthread.php?t=4517

DISCLAIMER: If you toast your router, cry to someone who cares. This hack is to be used at your own will, and I take no responsibility for you smoking your $112 802.11G WAP. Short-media also takes no responsibility. REMEMBER THAT THIS WILL VOID YOUR WARRANTY.

I repeat this clearly:
-You're doing this on your own accord
-I am not responsible for damages
-Short-Media is not responsible for damages
-Your warranty will be voided

PREREQUISITES: The newest 9/9/03 firmware, WRT54G by Linksys.

PRELIMINARY INFO: The WRT54G runs Linux kernel 2.4.5 on a 125MHz MIPS processor with 16mb RAM. Approximate 9MB free if you would ever wish to execute your own code. If you want to know how to do that, let me know.

THINGS TO REMEMBER: This hack must be applied again if you (A) Turn your router off and back on (B) Soft-reset the router (C) Factory reset the router. The command functions like a linux command on a temporary file, reboot and it's gone.

Log into the router through your browser and proceed to the ping test window by going to administration -> Diagnostics. Click the ping test.

Follow the picture below. Where it says "TXPWR 84" change the value to 48. Eighty-four is too high a mW rating without additional cooling, and this value provides unstable signals.

The Linksys WRT54G comes with 19mW signal, which is complete crap. Due to FCC regulations, Linksys can't ship it with a signal strength much higher than this. They can't stop you though!

`/usr/sbin/wl -i eth2 txpwr 48` is the string to be placed in the WRT54G ping test interface. Place it in, and hit ping. Your signal strength is now double, and you have successfully overclocked your router. Also, go to the advanced section on your WRT54G and enable "Frame burst." This is what netgear and several other 54G providers are calling turbo mode.

__________________

Sidney

Why not a signal booster; kick that mother from 19mW to 4 watts like CB radio a linear amp will do the job

I bought a signal booster last year, never got it installed.

__________________
lazyman

Opteron 165 (2) @2.85 1.42 vcore AMD Stock HSF + Chill Vent II

jmke

that's possible too, you can overclock wireless range and hook up larger booster

__________________

Liquid3D

I'm sorry I don't reply to some of my own posts I actually came across this post while Googlig the P965-DS3 for power stage phts betwen rvision.

I always look for the emails that "someone replied to your 3-page post at such&such.com" I have to fid and enable that feature heere or I' mising out on your input and empathy.

Thank you. As with life Im looking for those "big answers" you know the ones which help you resolve thing complicated with simple procedures haha.

The software theory is a very good point, innate commands such as: if-any-minor-problems-then-goto-frustrate-stage are more common. And In the BIOS there are also settings such as "HALT on any error" or something like that I don't want to re-boot now just to get it perfect.

Anyway thank you all yiu really help me and that's why I've been posting here more often.