Well, the first review is out for one of the much anticipated KT266 chipset based mobos (the Gigabyte 7VTX) and it's a real bummer. So much for VIA's highly touted V-link bus architecture... Seems that VIA just can't get anything right the first time around!!!
That review really bummed me out. I had been delaying my new system purchasing decision until reading that. VIA really is pathetic (and I really should have known better than to think they could design a better chipset than AMD or ALi)!
Since that website had no skip to section button, here is the conclusion
The board design is quite similar to the 7DXR but without the RAID. GA-7VTX is a simple board which has pretty good overclocking features. It's FSB support of up to 200Mhz and jumper of 170Mhz hints that the KT266 chipset is as good as the KT133A in terms of overclocking.
Although it has all the tweaking like VI/O, VAGP, Multiplier and FSB, it lacks one important one which is CPU Vcore setting. Let's hope that in future revisions of this board, we will have that option. Most of the newer batch of CPU should be easily overclockable even at default voltage, so it might not be that important after all.
Not to forget the mentioning of good sound quality CT5880 chipset by Creative and the Dual BIOS. This board also support online BIOS updates. These are good features which is commendable.
This board is an early sample, so that might be the cause for the poor performance. I suspect it is the VIA chipset rather than the engineering design of Gigabyte. Our results are in line with most of the other websites which have reviewed the early samples of KT266.
AGP Pro Slot
Good overclocking features (Vcore adjustments non existant)
No Diagnostic LED
<font color=red>Computers are useless. They can only give you answers.</font color=red>
Looks like the guys over at lostcircuits.com wouldn't agree with the ALi boards being the most reliable. They had serious memory stability problems on the one they tested. The question arises as to whether they just got a bad board, but their review shouldn't be ignored.
Now in defense of the A7M266, I've had mine up and running stable for over a week. I've pushed it through some pretty heavy 3D action and only saw 1 random crash (that can't be duplicated). Everyone sees those from time to time. No problems with that VIA southbridge to this point.
I also heard it rumored that Tyan is releasing an all AMD based 760 board (both north and south bridge) but I don't know if that is true or not.
A scientist will never show any kindness for a theory which he did not start himself. – Mark Twain
My answer to that is an emphatic NO! As I have stated previously, the AMD760 chipset consists of the AMD761 Horthbridge and the VIA686B Southbridge. VIA? - yuck!!!
Regarding the ALiMAGiK 1 chipset, ASUS' A7A266 is a new board going through mobo infancy problems. On top of that, it's overly complicated and reduces DDR DIMM slots to 2 in order to make room for 3 SDR slots. IMHO, if you are going with a DDR board, why screw around with SDR SDRAM?
The Iwill KA266(-R) has 3 DDR DIMM slots, affording you 50% more DDR memory capacity! On top of that, it has had nearly 4 months of maturing, to the point where it is very stable.
Unfortunately, not everything’s perfect with the chipset.
Firstly, there seems to be a compatibility issue between the chipset and ATI cards. Iwill was quick to assure us that the problem lies with the MAGiK 1 chipset, not the board itself, which we assume to be true, since the board didn’t have any AGP-related troubles with other video cards. ALi and ATI are supposedly working closely to resolve the issue, and we would expect that it will be dealt with before the final revision of the chipset and boards are released. In the meantime, we stuck with a GeForce 2 in our testbed, and it ran without a hitch.
Secondly, the 266 MHz (133 MHz DDR) bus setting was not as stable as we would perhaps like. We’ve seen in the past that AMD’s EV6 bus is nowhere near as easy to ramp as Intel’s GTL+, and so it’s not surprising that ALi, this being its first Athlon chipset ever, is having a little trouble with the 266 MHz setting. The board was able to complete most of our benchmarks at 266 MHz, except for MadOnion’s 3Dmark2000 and BAPCo’s SYSMark2000, both of which failed to make a complete run despite several attempts. The rest of the benchmarks ran without a hitch. Again, this is an issue that we’re sure will be resolved before the final revision of the chipset is released.
All in all, these problems are fairly typical of early releases of just about any chipset, and shouldn’t cause too much alarm. There are always little bugs. So long as they are resolved in a timely manner, we don’t see cause for worry. If ALi can’t figure out how to solve the problems--then we worry.
I think it is fairly safe to assume that until AMD start mass-producing their own chipsets, we are going to be stuck with the crap combination of via southbridge and amd northbridge... where is the damn MSI K7 master when you need it?
Australian PMs are like steer horns; a point here, a point there, & a heap of bull in the middle.