Phenom 9500 w/ MSI K9A2 Platinum

[justapost]

I'll show you a way to test if your NB is really changing.
Throw the HT to the same multi as NB. If NB really is changing, the HT will be changed to the same speed. If not, the HT will not change above 9x which is because 9x is the real operating NB multi.

Try it. If its working, then you'll be able to validate with CPUZ and all software will be in sync (you can see macci's ss as a good example how). Also performance should'nt drop with higher NB but increase.

Thx, only checked CPU-Z, so i did not realize the ref HT drop.
There is a msr register 0xc0010071. Bit 63:59 are the MaxNbFid. In my case 5 means 1800MHz max.

[Daveburt714]

I've come to the conclusion that AOD doesn't play all that well 64 Bit Vista...
Neither does CPU-Z

This may be old news, but if you set CPU Vid to 24 in p-states, it will boot up at 1.25v's everytime

28 = 1.20v
26 = 1.225v
24 = 1.25v

[KTE]

KTE are these first generation boards sufficient enough for quad core overclocking? In other words what kind of board taxing do you have to put on it to overclock quad cores?

Looking at the MSI/GBT/DFI/ASUS boards, I have mixed opinions.
The MSI CF versions isn't suitable for quads and oc'ing them yet IMO.
The M3A, I'm not sure about yet. Looking at justaposts' progress.
The rest look all fine to me apart from BIOS problems.

We need guys to measure MOSFET temps on their boards under full load at 1.4-ish VCore. That would help.
Measuring Vdroop and Vdrop at the inductor/MOSFET/SUper I/O would also help.
We also need guys to test stability at above 2.4GHz, like folding or a 48 hour stability run. That would also speak volumes.

My board always boots with CPU Vid @ 1.2... I guess I could try and play with it in p-states to try and force it to 1.25.

Usually this is a performance power state thing. If you leave everything on auto and stock, does it still boot 1.20VID?

Thx, only checked CPU-Z, so i did not realize the ref HT drop.
There is a msr register 0xc0010071. Bit 63:59 are the MaxNbFid. In my case 5 means 1800MHz max.

The HT ref. hasn't actually dropped, that's what I'm saying.
However, what has happened is, the CPU clock has stayed the same, the HT clock has stayed the same, the RAM clock has stayed the same and the NB clock is running at 9x.

This is a known occurrence with Phenom 9500/9600 that I've tested and all ones I've seen from others. Higher than 9x multi is no change although it causes software to start reading as though its downlocked and higher than stock multi is also a no change. That's typically because they are locked.

I've come to the conclusion that AOD doesn't play all that well 64 Bit Vista...
Neither does CPU-Z

I always stay clear of Vista 64-bit and 32-bit for testing new boards, new CPUs, new BIOSes, new apps. Because they all have patchy or bad support for it until much later and you never realize where the problem lies then. It's confusing.

This may be old news, but if you set CPU Vid to 24 in p-states, it will boot up at 1.25v's everytime

28 = 1.20v
26 = 1.225v
24 = 1.25v

Looks correct to me, IF you're in a low performance power state (low NB clock will do this). Experiementation is good, don't ya think?

Have you tried going lower than 24? Does it boot? What VID does it give you?

So far I've seen it's locked at 1.25VID max.

[Blacklash]

AoD is hit and miss with Vista x64. Changing voltages seems to be fine if you do them one at a time and altering CPU HT is a crap shoot. Attempting to load a profile via AOD = freeze every time for me. Vista x64 overclocks fine on my two Intel rigs, and I love it's sound stack and other features. I won't be giving it up to make an extra 200MHz on this CPU.