doh!!!!!!!
nevermind, the overclocking part was testing the memory subsystem versus raw FSB speed. There shouldn't be much of a diffrence anyway since, Doom 3 is heavy depended on graphics card and raw CPU speed.

What is the deal with Asus and Abit? Do they really have to name their boards with model numbers that are almost, but not quite, identical? I always confuse the two when I'm trying to remember which board is which. A8V vs AV8, K7V and KV7, A7N or AN7! WTF! Maybe that's the reason I bought an MSI board...

I'd like to eventually upgrade to an Athon 64 socket 939 system. What I'd like to know is if VIA chipsets still have a shoddy PCI bus or have they squashed that bug? I used to have a Giga-byte GA-7VAXP based on the KT400. I really never had a problem with the board until I got a 3ware Escalade 7506-4. Performance was just never that good. More recently I upgraded it to an Asus A7N8X Deluxe. Aside from the speed boost due to the dual channel RAM hard drive performance with the same exact controller and drives practically doubled.

So, how is PCI performance on these newer VIA chipsets when compared with other chipsets?

No useless, dead-in-a-week northbridge fan! My prayers have been answered!

*yawn*

You guys going to review any NF3-based S939 boards any time soon?

Nice review of what appears to be a very nice (especially for the price) board!

Now time for a slightly OT question:

Why is it that only Asus seems to be supporting ECC on their Athlon64 (both S754 and S939) boards? Yes, I know that S940/Opteron is the "official" server platform... but Athlon64 would be very nice for low-end servers if only they could take full advantage of ECC. Furthermore, as memory sizes and densities continue to climb, I believe ECC will become more important on the desktop as well.

Perhaps AMD is discouraging mobo makers from adding ECC support to their Athlon64 boards, to prevent Athlon64 from cannibalizing sales from the low end of the Opteron line?

When will we see these boards with PCI-Express slots?

How is the power management on this board?
It would be nice to see how it overclocked at high Vcores. I recall one recent board that had dedicated fan-ducted sinks for its mosfets for this purpose.

Yet another mainboard review.. YAWN. Where is that article about Tech Reports latest server upgrade?

Nice review, nice board. Can't really find anything negative to say about it..

Oh, maybe just to point out that Firewire is supposed to have 0% CPU utilisation since it's NOT a Hostbased (aka master - slaves(s)) protocol but rather a P2P protocol and each firewire controller takes care of everything. If you have a situation where there is CPU utilisation for Firewire transfers, it's cause whoever designed the Firewire chip in question made it a "Winchip" type where they offload some stuff through special drivers to the host CPU. I think the first version of Via's Firewire chip was such an abomination..

Adi

Yes, this is the kind of board that would go great with a A64 3500 and something like my Asetek Waterchill.

Yes, let's see.

Niiiiice review!

What about an overclock comparo between the Asus A8V and the Abit AV8?

They're pretty much the two boards I'm considering getting later this month as I piece my socket 939 system together, and overclocking is all. (to me anyways)

NM - I re-read the Asus article... same review really with two different overclock and board layout/features portions for each board. The Abit appears to overclock better than the Asus board. Hmmm. I think I know where my cash is going!