The Plot Thickens

I’ve been poring over the article found at Anandtech and it has been truly frustrating.

This is what they said:

1. They found connectors on an AZ11 which could be used to change multipliers
2. They soldered pins to it
3. They found out the pins worked.
4. They thought that overclocking the Thunderbird wouldn’t be a problem, but that the Duron would be due to voltage.
5. The AZ11 didn’t have working voltage controls.
6. So they modified the voltage settings on the Duron by “connecting the dots”.
7. After all this was said and done, they tried overclocking both the Thunderbird and Duron, and couldn’t. They could underclock, but not overclock.

   One Piece That Doesn’t Seem To Fit

   These weren’t production CPUs; they were allegedly unlocked samples. So why are there multiplier locks?
   Doesn’t quite jive with AMD’s story.

   What does this mean?

   This isn’t conclusive

   While it’s a strong indicator that there is some sort of maximum multiplier lock in the AMD chip, the fact that the AZ11 cannot override it doesn’t mean that it can’t be overridden (nor does it mean it can).
   This isn’t exactly a feature of the AZ11. 🙂

   It’s quite possible that a motherboard needs BIOS or hardware capabilities to override the multiplier that just aren’t there anymore for the AZ11.

   Remember, a couple days ago, Insane Hardware had no problem overclocking a Duron on an A7V. Anandtech used preproduction samples and couldn’t. Only apparent difference is the mobo.
   If you read today or tomorrow or next week that someone getting an A7V or a KT7 can’t overclock, either, then start worrying.

   Email Ed