Abit AW9D-MAX: When "Beta" MAX is a good thing
by Gary Key on September 8, 2006 3:10 AM EST- Posted in
- Motherboards
Basic Features
Abit has delivered a well optioned but very performance oriented 975X board that should sell for around US $229 or under. While our BIOS is still beta we were surprised at the stability of the motherboard during our benchmarking. We will provide screenshots and a more in-depth look at the BIOS once we receive a shipping version. At this time the one glaring omission is the lack of advanced DRAM timing control settings and a 1333 memory strap that would certainly let this board overclock further. Abit only allows the basic four timings to be changed (tCAS, tRCD, tRP, tRAS) and for a board of this caliber we believe this is a mistake. The ability to increase the MCH voltage to 2.00V and memory to 2.65V is impressive considering the limits on the other 975X based boards. However, we wish the memory settings above 2.35V were available in .05V increments instead of .10V increments.
One of the main BIOS issues consisted of the inability of the board to lower the CPU multiplier on standard Core 2 Duo processors (and raise it on the Core 2 Extreme), a feature available in current Gigabyte and ASUS boards. The weirdest issue was that setting the PCI Express speed above 100Mhz would render our SATA drives inoperable in most instances. We had to hunt and peck for an acceptable increase in the PCI Express speed before our drives would be recognized. Our Seagate drives would work at 102 and 108 at certain times while our WD SE16 drives would only work at 105 with the Raptors not working at anything above 100MHz. In the end, none of the SATA drives would work above 100MHz consistently so we left this setting at the default.
We also had trouble overclocking the board at first unless we disabled the Abit EQ thermal controls. This held true when trying to increase our CPU or memory voltages at various times. If the system defaulted to standard EQ limit settings (memory voltage at 2.1V maximum) then we would have to disable the Abit EQ controls, set our increased voltages, and then enable EQ before we could overclock the system. Although this typically worked, the bios would sometimes lose its way and no longer accept the extended voltage settings we specified in the EQ utility. We ended up disabling EQ voltage monitoring altogether during overclocking. Our remaining issue was the bios was unforgiving with specific DDR2 modules at certain settings. We generally found that the 4:5 ratio would work at times with our other DDR2-800 test modules when a 1:1 ratio would not and vice versa. We certainly believe from our discussions with Abit that these BIOS issues will be fixed before retail release but the board basically works fine at this time.
Abit AW9D-MAX Specifications | |
Market Segment: | Performance Enthusiast |
CPU Interface: | Socket T (Socket 775) |
CPU Support: | LGA775-based Pentium 4, Celeron D, Pentium D, Pentium EE, Core 2 Duo |
Chipset: | Intel 975X + ICH7R |
Bus Speeds: | 133 to 600 in 1MHz Increments |
Memory Speeds: | SPD, 533, 667, 800 |
NorthBridge Strap: | CPU, 1066, 800, 533 |
PCIe Speeds: | Auto, 100MHz~200MHz |
PCI: | Fixed at 33.33 |
Core Voltage: | Base CPU V to 1.7250V in 0.0250V increments |
CPU Clock Multiplier: | Auto, 6x-11x in 1X increments if CPU is unlocked |
DRAM Voltage: | 1.75V ~ 2.65V in .05V or .10V increments, above 2.35V all increments are .10V. |
DRAM Timing Control: | SPD, 4 Options |
NB Voltage: | 1.50V ~ 2.00V in .01V increments |
Memory Slots: | Four 240-pin DDR2 DIMM Slots Dual-Channel Configuration Regular Unbuffered Memory to 8GB Total |
Expansion Slots: | 2 - PCIe X16 (x8 operation in multi-GPU setup) 2 - PCIe X1 1 - PCI Slot 2.3 1 - Audio Max Slot |
Onboard SATA/RAID: | 4 SATA 3Gbps Ports - Intel ICH7R (RAID 0,1,1+0,JBOD) 3 SATA 3Gbps Ports - Silicon Image 3132 1 e-SATA 3Gbps Port - Silicon Image 3132 |
Onboard IDE: | 1 ATA100/66/33 Port (2 drives) - Intel ICH7R |
Onboard USB 2.0/IEEE-1394: | 8 USB 2.0 Ports - 4 I/O Panel - 4 via Headers 2 Firewire 400 Ports by TI TSB43AB22A - via Headers |
Onboard LAN: | Gigabit Ethernet Controller - PCI-E Interface Realtek RTL 8111B |
Onboard Audio: | Realtek ALC882M HD-Audio 8-channel CODEC - Dolby Master Studio |
Power Connectors: | ATX 24-pin, 8-pin EATX 12V, 4-pin 12V Molex |
I/O Panel: | 1 x PS/2 Keyboard 1 x PS/2 Mouse 2 x RJ45 1 x eSATA 4 x USB 2.0/1.1 |
BIOS Revision: | AWARD W628 - Beta |
Abit has delivered a well optioned but very performance oriented 975X board that should sell for around US $229 or under. While our BIOS is still beta we were surprised at the stability of the motherboard during our benchmarking. We will provide screenshots and a more in-depth look at the BIOS once we receive a shipping version. At this time the one glaring omission is the lack of advanced DRAM timing control settings and a 1333 memory strap that would certainly let this board overclock further. Abit only allows the basic four timings to be changed (tCAS, tRCD, tRP, tRAS) and for a board of this caliber we believe this is a mistake. The ability to increase the MCH voltage to 2.00V and memory to 2.65V is impressive considering the limits on the other 975X based boards. However, we wish the memory settings above 2.35V were available in .05V increments instead of .10V increments.
One of the main BIOS issues consisted of the inability of the board to lower the CPU multiplier on standard Core 2 Duo processors (and raise it on the Core 2 Extreme), a feature available in current Gigabyte and ASUS boards. The weirdest issue was that setting the PCI Express speed above 100Mhz would render our SATA drives inoperable in most instances. We had to hunt and peck for an acceptable increase in the PCI Express speed before our drives would be recognized. Our Seagate drives would work at 102 and 108 at certain times while our WD SE16 drives would only work at 105 with the Raptors not working at anything above 100MHz. In the end, none of the SATA drives would work above 100MHz consistently so we left this setting at the default.
We also had trouble overclocking the board at first unless we disabled the Abit EQ thermal controls. This held true when trying to increase our CPU or memory voltages at various times. If the system defaulted to standard EQ limit settings (memory voltage at 2.1V maximum) then we would have to disable the Abit EQ controls, set our increased voltages, and then enable EQ before we could overclock the system. Although this typically worked, the bios would sometimes lose its way and no longer accept the extended voltage settings we specified in the EQ utility. We ended up disabling EQ voltage monitoring altogether during overclocking. Our remaining issue was the bios was unforgiving with specific DDR2 modules at certain settings. We generally found that the 4:5 ratio would work at times with our other DDR2-800 test modules when a 1:1 ratio would not and vice versa. We certainly believe from our discussions with Abit that these BIOS issues will be fixed before retail release but the board basically works fine at this time.
56 Comments
View All Comments
OvErHeAtInG - Sunday, September 10, 2006 - link
...probably to avoid cracking. Yes?yyrkoon - Monday, September 11, 2006 - link
Actually, Its a known issue with some ABIT motherboards, that if you put a metal screws in these holes, it will short the board out. Dont take my word for it though, look around on ABITs forums :)jackylman - Friday, September 8, 2006 - link
Abit also integrates additional cooper layers in between the PCB layers to aid in the extraction of heat from these areas.OvErHeAtInG - Friday, September 8, 2006 - link
While we're picking nits:The 24-pin ATX connector is conveniently located on the edge of the board in front of the number four DIMM slot. The 12-pin ATX connector is located at the edge of the first DIMM slot. The CPU fan header is located next to the 12-pin ATX plug and due to the size of the CPU area requires your heatsink/fan to be properly oriented if the cable is short.
Am I daft, or do you mean 8-pin instead of 12-pin? Sorry if this was already mentioned, I skimmed the comments.
jackylman - Saturday, September 9, 2006 - link
Both typo's fixed. Good job! ;)Gary Key - Sunday, September 10, 2006 - link
I apologize about those errors, using DNS on this article and still do not know how eight became twelve (eight in the charts and my type written text), flat missed catching the other one as that is one word that DNS does not like. ;-)joex444 - Friday, September 8, 2006 - link
you say one of the bios issues was not being able to change the multiplier down, then describe how you went from a 9x to an 8x multiplier... does it allow multiplier changing only with the new bios and the stock one didn't? i don't really care what the stock bios does if i'm going to flash it to the newest one anyways...Gary Key - Friday, September 8, 2006 - link
We could change the multiplier on the X6800 only. We used that chip as stated in the overclocking section to test at 8X and 6X. The issue with the X6800 is that you cannot raise it past 11x. :)johnsonx - Friday, September 8, 2006 - link
You gave us over a half dozen pictures of the board from every angle, but you couldn't toss us a picture of these 'overclocking stripes'? what the hell are they, and how do they work?
Gary Key - Friday, September 8, 2006 - link
A picture of the OC-Strips technology along with additional wording is available now.