badge



We have 32 guests online

Login Form

Username

Password

Remember me
Password Reminder

Unlocking the Athlon II X3 435 Print E-mail
Written by Michael Schuette   
Nov 27, 2009 at 05:23 AM
Despite having different code names such as Toliman, Heka and Rana, AMD's triple core solutions are anything but a discreet die design. Rather, existing Deneb (Phenom II X4) and Propus (Athlon II X4) dies are tested and those with defects in one of the four cores are rebadged as triple core solutions. It is no secret though, that often enough, marketing and demand will create defective dies where the only defect is realistically the lack of the latter. In other words, many of the triple core CPUs offered at a budget price compared to their full-fledged brethren are wolves in a sheep skin.

To make matters more interesting, it turns out that a number of the Athlon II brand processors supposedly based on the Propus core are in fact Deneb processors featuring the 6 MB shared L3 cache, a feature that makes those CPUs more pliable for gaming purposes. The good news in this case is that the crippling inflicted by AMD appears to be done on the basis of register programming rather than blowing fuses and, therefore, with a few BIOS tricks, it may be reversible. To sum it up, chances are that an inexpensive Athlon II X3 processor may transform into a Stanley Beamish-like Phenom II X4 with the extra bonus of the pill not wearing off.

BIOS screenshot of the Athlon II X3 435 after enabling "Unleashing Mode"

The big question is, as always, is it going to work or not. The short answer is: maybe! As long as yields are good enough so that there is no fall-out of bad cores and demand for the triple core processors by the OEMs his high, chances are that many of the X3 435 and similar are going to be either fully functional Athlon II X4 or even Phenom II processors.

The Test:

We started out with a sample of the Athlon II X3 435 (media sample) and running it on the ASUS M4A79T using the 2002 (or later) BIOS revision (the latest we tested was the 2205 version).

In the "Advanced" menu of the CMOS Setup utility, select Advanced Core Calibration (ACC) and change it from "Disabled" to "Auto". This setting will unhide a separate tab called "Unleashing Mode" with the options "Enabled/Disabled". When "Enabled" is selected, it will give several options for triple core selection and, more importantly, it will unlock the L3 cache. A shortcut is provided by pressing "4" during POST to enable all four cores. Unfortunately, the ASUS BIOS does not allow to selectively enable all cores without activating the L3 cache.

Default configuration of the test CPU.

Did it Work?

To make a long story short, the sample we had could not be persuaded to make it into Windows7, regardless of which core we disabled. As soon as the "L3" cache was enabled, the system managed to get to the "Windows7"/"Previous OS" screen but from there on, all we got was a black screen. We tried to underclock the CPU by selecting multipliers as low as 5x to give a marginal core a chance to at least make it into Win7 at 1 GHz, to no avail.

At this point, it is not clear whether the "Unleash Mode" (CPU Unlock in MSI parlance) just misreports a non-existent L3 cache, which then causes the sytem to stall during boot or whether the L3 on our sample was actually present but crippled so badly that the CPU just couldn't make it into Windows. It doesn't matter, though, this is one sample and whether we have a true Propus die or whether it is a Deneb in disguise is inconsequential. What matters is that there is a potential for getting a bargain deal and there have been reports on Inpai.com.cn that at least some of the retail processors in circulation are in fact Phenom II CPUs.

Update

We confirmed this morning with AMD that the '435 sample we have in our hands is in fact a Deneb die. We also tried to down-clock the NB/L3 all the way to 1200 MHz to see whether this would allow to get into Windows but no matter what, the system kept stalling immediately after launching the boot process, indicating that the die was defective enough to not even work at low frequency.

What we don't want to do with this article is sending our readers on a wilde goose chase, trying to hunt down any possible data code of Athlon II X3 and potentially getting frustrated by lack of success. What we do want to do, though, is encourage those who settle for the X3 435 to try their luck and see how far they can get. Hey, you never know...

Special Thanks to our Reader "Justapost" for pointing us into the right direction.

Discuss this article in our forums

Last Updated ( Dec 06, 2009 at 12:55 PM )
<Previous Article   Next Article>