Is my Duron really this crappy ?

[CHHAS]

I unlocked my new Duron 650 today, but I can't even get a post at multiplier x7 or better. I tried setting the multiplier at x6, and that worked, and I could boot at 650 MHz with multiplier set at x7 before I 'took the lead' to it ...

I tried boosting the voltage bit by bit to 1.85V, still nothing.

I can overclock the chip to 715 MHz by setting the FSB to 110 MHz without boosting the voltage, so I would assume that I would be able to boot at 700.

I haven't got a clue as to what's going on here, any suggestions ?

[Himself]

Make sure you have really unlocked your cpu, do it again if possible, use a continuity tester to make sure the pins are shorted, also check that you are not bridging sets of pins together.

It could also be a mb/bios issue.

[CHHAS]

My board is a Soltek SL-75KAV with the newest BIOS, there should no problems there.

I'm certain that I haven't bridged the bridges ( ) but I can't be certain that all the pins are connected by enough lead. I assumed that I would either get an error, or a still locked CPU if they weren't, but I'll try to check that again.

On an up-note, my CPU temperature has dropped by 5C after replacing the 'rubber mat' with 'heat conducting goo'. I'm now running at 33C while doing SETI at 650 MHz.

[Sir Hitech]

<font face="Verdana, Arial, Helvetica" size="2">Originally posted by CHHAS:
I unlocked my new Duron 650 today, but I can't even get a post at multiplier x7 or better. I tried setting the multiplier at x6, and that worked, and I could boot at 650 MHz with multiplier set at x7 before I 'took the lead' to it ...

I tried boosting the voltage bit by bit to 1.85V, still nothing.

I can overclock the chip to 715 MHz by setting the FSB to 110 MHz without boosting the voltage, so I would assume that I would be able to boot at 700.

I haven't got a clue as to what's going on here, any suggestions ?</font>

CHHAS:

It took me five tries the first time I unlock my TB. Sharpen your H2 pencil or use 0.5mm mechanical lead. Go over each pair of bridges with 2-3 short straight line strokes and make sure the lead cover the entire bridge and not bleeding toward the neighboring bridge. If no working, use damp paper towel clean off the lead thoroughly and repeat the whole process again. Patience, my man! Its the Art of Overclocking... LOL

[CHHAS]

Hehe, I'll give it another go tomorrow, right now I'm in no condition to mess with the interior of my PC due to my attempts to cure my cold

http://forums.murc.ws/ubb/Forum9/HTML/001289.html

[CHHAS]

An update:

I managed to get my memory to run at 133 MHz, so I decided to do some FSB overclocking, I can only get to 110 MHz, any higher and it hangs while booting.
The core is 34C at idle, so temperature is not an issue.

[Greebe]

That's normal CHHAS

[CHHAS]

Even with a KT133A chipset ?

[Greebe]

yes

[DentyCracker]

6.5x133 1/3=866 2/3
maybe your chip doesn't like running at 266 2/3 MHz

[Sir Hitech]

<font face="Verdana, Arial, Helvetica" size="2">Originally posted by CHHAS:
Even with a KT133A chipset ?</font>

FWIK, you need 133/266 MHz DDR bus Athlon or Duron processors to take advantage of the KT133A boards. I have no idea how to distinguish the 133fsb chips from 100fsb chips. Maybe all the Duron and TB's are 133fsb compatible. None of the website reviews really clarifies this. Maybe Mike can enlighten us.



[This message has been edited by Sir Hitech (edited 18 January 2001).]

[Himself]

I'm not sure if KT133A reviews were done with official 133FSB cpus or not, I tend to doubt it. It could simply be that this paticular cpu doesn't like a faster FSB than 110 (220). It could also be that the +33MHz mode needs to be turned off because the ram can't do more than 143MHz. The FSB itself can probably go higher, but you would be underclocking your ram unless you cpu could do a FSB of 133 (266).

[Indiana]

It seems that you can drive any "normal" (100MHz FSB) TBird or Duron at 133MHz FSB with the KT133A, the FSB limitation apparently was due to the chipsets not the CPUs, there were reviews that stated that they used a 100MHz FSB CPU.
Of course this is true only as long as the CPU is able to run at a 33% oc'ed Speed. If not you can still LOWER the multiplier to make it run at the higher FSB.

For instance if you have an TBird 1GHz, it's unlikely that it will do the 1.33GHz you normally would get with a FSB of 133MHz. Maybe 1.2 are more realistic, so you can lower the multiplier to 9 and get 133*9=1197MHz, this setting will be faster than raising the mutliplier to 12 with the FSB at only 100MHz, especially due to the higher RAM-throughput.



[This message has been edited by Indiana (edited 18 January 2001).]

[Greebe]

let me reiterate my previous comment...

When OCing any cpu not locked, you start by running the bus the highest you can attain and then see if you can run it still faster by upping the multiplier.

This insures that you are gaining proformance in both bandwidth and clock speed.

With that said, if your cpu is already at it's maximum speed, increasing the bus will only lead to trouble.

[Indiana]

Yes, but it still is generally desirable to achieve the same speed with a higher FSB and a lower multiplier than vice versa.
Of course this is valid only if the CPU is the limiting factor and not the RAM, as it is with most higher rated CPUs.