Seeing the AZ-11 for the first time gives the impression of a very large ATX sized board. At 9.6" of width, there seems to be much more circuit card real estate than most ATX boards currently seen on the market. The bottom of the board in particular has plenty of room for additional slot connectors (such as PCI or ISA) and the need for this real estate is questionable. FIC seems to have grown the PWB in the width wise direction instead of using the full height of the board. The AZ-11 comes with 5 PCI slots and 1 AGP slot, but the addition of another slot might help the marketability of this design. Thankfully FIC left off the ubiquitous AMR slot which typically goes unused by a majority of users. I’d recommend sizing up your ATX case before trying to shove this big guy into a space that might be a tight fit and not give optimal room to breath and be appropriately cooled.
The AZ-11 comes with three DIMM slots that are capable of housing up to 1.5GB of PC133 SDRAM and is equipped with ECC memory support. There are the standard two UDMA66 IDE channels and floppy interface. These connectors are located in a very nice location at the right of the board making cable routing easy and clean. Curiously there is only one COM port connector installed. A hole where the second COM port connector was unpopulated. There is also the standard keyboard, PS/2 port and dual USB connectors available.
The Socket A area is vacant enough for most fan/heatsink combos, but don’t expect to be able to use too big a fan given the close proximity of the numerous bulk capacitors around the socket. The biggest negative to the layout is the poorly located ATX power connector. Being actually below and on the left side of the socket, it doesn’t take too long to see your main power harness hanging over your processor fan unless you take care to tuck it out of the way.
The AZ-11 features VIA’s KT133 chipset which was specifically engineered for the Socket A CPU connector. The KT133 has essentially the same feature set as the KX133 used in many Slot A designs. After the news broke about the KX133 not actually being Thunderbird compatible, much confusion ensued about which Slot A board would work with the Thunderbird and which was rendered obsolete. If you were one of the lucky ones to avoid the KX133/Slot A boards by waiting, a KT133/Socket A board is the type of board I’d recommend looking for to house the new Thunderbird Athlon or Duron. The VIA KT133 chipset consists of a VT8363 system controller for the North Bridge and the VIA 686A Super South Bridge. The KT133 is also capable of supporting AGP 4X. Onboard AC97 sound is possible via the Sigmatel STAC9744T included on the mainboard. Line in, line out and mic in are all supported with external connectors built into the board.
The AZ-11 comes equipped with what FIC calls NOVUS II BIOS adjustments. This feature includes something called "Easy Key" which provides instant keyboard access to the BIOS adjustments for clock and default settings. By pressing TAB during setup, you can get past the FIC BIOS screen and do a number of things. Pressing CNTL-C leads you right to the Frequency/Voltage control page within the BIOS. Pressing CNTL-P automatically loads in optimized performance default settings into the BIOS. Pressing CNTRL-F loads up failsafe BIOS settings in case you need to back out of some aggressive settings to get things to work again.
You can create a custom LOGO with something called "LogoGenie" that can be displayed during bootup (a feature many OEMs are likely to take advantage of). There is something called BIOS Guardian which gives a rudimentary form of anti-virus protection for your BIOS. This will need to disabled if you plan of flashing the board’s BIOS. Clockometer is a little GUI program that allows users to adjust system clock speeds without entering the BIOS subprogram. A very unique feature called Audio Alert II can also be added as a manufacturing option. Audio Alert II gives on onboard voice error message when problems are encountered during the BOOT process. The PC speaker from the case will most likely be needed for this to work since previous versions only generated garbled sounds via the sound card outputs. This did not function on my evaluation sample.
There are a number of FSB speeds that can be setup on the AZ-11. The following listing shows both the FSB speeds settings available followed by the resultant PCI speed: 100/33, 102/34, 104/34, 106/35, 107/36, 108/36, 109/36, 110/37, 111/37, 112/37 and 133/33. This gives some degree of flexibility if overclocking is your passion. Frankly the small degree of speed increase possible from the 900MHz Thunderbird was not worth the effort (it’s already extremely fast). Some sites have been excited over the inclusion of some multiplier setting tweaks if you like to solder some pin headers onto the AZ-11 near the processor slot. The PWB had the pin headers unpopulated but they are available if your adventurous and know how to use a soldering iron. Obviously this isn’t something I’d recommend for business users.
Since several websites have published information about how to unlock the Socket A processors, the multiplier switches may actually have some function for the bolder users. While the settings are indicated in the manual, they are listed by processor speed rather than by multiplier. For some reason, not all of the multiplier settings have been documented, but the following table provides all of them:
Following the RWT tradition of thoroughly testing the board, I ran numerous iterations of benchmark software under both Windows98SE and Windows2000. I started out with Windows98SE and ran the batch mode tests of the ZD Winstone 99 1.3 and Content Creation Winstone 2000 over 8 hours each while rebooting between each run and didn’t see any hang up or problems. I also employed Passmark’s Burn In for over 24 hours without any hiccups.
I also applied Windows 2000 and reran all the benchmarks with some success, but occasionally I had the board hang during testing. The amount of problems I had were minor but still worth mentioning. My experience using Windows 2000 is limited and I’m not sure how flawlessly these benchmark programs usually work under Win2K. Typical usage didn’t seem to cause any problems.
Occasionally the ATX soft power switch did not work as expected. Just turning on the power supply without closing the soft power switch caused the board to fire up a number of times. This will hopefully be alleviated in a future revision. The performance of the board seems quite good scoring ahead of my Coppermine PIII 850MHz processor by a noticeable margin.
Be the first to discuss this article!