Asus A7A266-E in the test: The 3rd generation of the ALi MAGiK 1
- 1 Preface
- The chipset: ALi MAGiK 1 - 'C1 Stepping'
- 2 Scope of delivery and documentation
- 3 Features
- 4 Test system
- 9 Conclusion A7A266-E
Since the 'C1 Stepping 'of the ALi MAGik 1 is not a completely new chip, but a revised version, b Dear, the basic structure of North and South Bay, as with VIAs KT266A, unchanged. Asus was thus able to keep the old layout of the A7A266, so changes are initially not found. Asus continues to use the AGP Pro slot, which, at least for the moment, does not bring any advantage for the average PC user. The same applies to the AMR, the audio modem riser, which is only used in the OEM sector, if at all. But at least the buyer does not have to accept one less PCI slot with the A7A266-E because of this, because the ALi MAGiK 1 supports only five of them out of the box. Another striking feature are the various RAM slots. In addition to only two DDR RAM banks, there are three more slots for the use of SDRAM on the board. Price-conscious upgraders who want to keep their old RAM for the time being could find a worthwhile interim solution with the ALi MAGiK 1. Other features are more in the details. So it occurred to usOnce again a capacitor standing very close to the CPU socket, which didn't get in the way of our GlobalWin WBK38, but still didn't look completely kosher. However, a look at AMD's specifications, the so-called 'keepout area', showed that it did not exceed the maximum height for this area and that every cooler that adheres to these standards should not have any problems with the A7A266-E. The position of the ATX power connector is also to be assessed as positive. Placing the cable on the edge of the mainboard prevents the cable from running past the CPU fan, the air flow remains unaffected.
Always something There is good news to report in the Stability division. In the end, we haven't come across a single mainboard in our entire mainboard test series whose reliability we should have commented negatively. It is no different with the A7A266-E. Apart from the 'reset' in Quake3Arena, which up to now every board has completed with too many benchmark runs set one after the other, the board ran cleanly and without a single flaw. The tuning of our Apacer CL2 bar also worked perfectly, even with the sharpest settings.
As usual from Asus, the A7A266-E can be tuned to your heart's content either with a jumper or in the BIOS. The multiplier in the BIOS can range from 5 - 13 andon the board can be bet from 5 - 12.5. The situation is similar with the front-side bus. The BIOS enables the clock to be adjusted in 1Mhz steps in an interval of 100-166Mhz, the board offers the settings 90, 100, 101, 120, 126, 133Mhz. The VDimm and core voltage are different. While the RAM voltage may only be set to 2.5 (default) or 2.6 volts on the board, only the BIOS offers the option of increasing the VCore. From 1.75 (default) to 1.85V can be freely selected in 0.025V steps according to your mood.
On the other hand, we discovered a nice but poorly documented property of the bio by chance. Just out of curiosity, the 'System Acceleration Mode' switch was set to enabled and a few tests were run. However, the increase in performance achieved in this way could hardly be attributed to a few tightened timings, so a look at the WCPUID tool should provide clarity. And lo and behold. Without a warning or notice, the BIOS had raised the front-side bus to 137Mhz and thus helped the CPU and memory to achieve a considerable performance boost. Not a bad idea in and of itself, but it should be noted that there are definitely memory or other hardware components that drop out at the slightest change in the FSB. In this case, the user only has to reach for 'CMOS-Clear'. For this reason, a description of this 'feature' would be extremely helpful and prevantive.
Note : Since Tomshardware.de found out in the summer of this year that Asus would prefer to use a performance-increasing 134Mhz when the 133Mhz front side bus was actually set, we dared to look into the WCPUID tool from H.Oda this time as well. We were able to test the A7V266-EWe are pleased to note that in the latest Bios version 1005 Beta the front-side bus ran at the specified 133Mhz for the first time. A rethink across the board? It doesn't seem like it is. Because even the latest BIOS of the A7A266-E (version 1007) continues to 'secretly' clock the front-side bus one MHz higher than it should. With a multiplier of 10, at least 0.7% more CPU power and RAM speed jump out (example 1333Mhz CPU).
On the next page: Test system