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 Performance
The performance of the board was exemplary in the majority of our tests. We found the board to be a very consistent performer and extremely stable up to its overclocking limit. Unfortunately, we do not know what the true limit of this board will be until we see the production level BIOS. As with recent Abit boards we will see significant BIOS tuning for the performance oriented crowd after stability or incompatibility issues have been resolved.
This board supports up to 2.65V on the memory that will allows for a great amount of overclocking headroom, though we would like to see additional .05V increments over 2.35V. The BIOS we tested only allowed adjustment of four memory timings but we expect to see additional options as the BIOS matures. We expect great things out of this board if Abit is able to release a production level BIOS for this board that keeps its current performance levels, addresses the issues we found in early testing, and is able to improve overclock levels.
Overclocking
We were really surprised by our overclocking results on this board with a beta BIOS. Like other recent boards we had to gradually increase the FSB speeds while adjusting voltages and memory settings in order to reach this level. Our retail E6600 was able to boot into Windows XP at 9x438 but we consistently had benchmark failures until we backed off to 9x431. While it would have been easy to take screenshots at this level with several popular benchmarks, our voltages and memory settings were outside of the range we would consider safe for air cooling over any length of time. We backed our settings down to 9x423 where we achieved a level that was consistent with a machine that would be safe running 24/7. In fact, due to improved memory latencies offering better overall system performance our ideal setting was at 9x416 with 3-4-3-10 memory settings on this board.
We were able to boot into Windows at 448 FSB with our X6800 set at an 8X multiplier, pass all benchmarks at 8x440, and could POST at up to 460FSB. We tried our E6300 and were able to run at a system stable setting of 7x448. Our board, BIOS, and memory combination did not allow us to POST past 460FSB. Overall the board was the equal of or better than the ASUS P5W-DH in overclocking. Since the memory remained at a 1066 strap during overclocking on the Abit board we found the general performance was slightly better than the ASUS past 416FSB.
Test Setup
Test conditions were maintained the same, as much as possible, over the platforms tested. For better comparison standard test results (1280x1024) were run with a single EVGA 7900GTX on all platforms. Our high resolution game tests were run at settings of 1600x1200 4xAA/8xAF with the EVGA 7900GTX card. We had planned on presenting 7950GX2 numbers but noticed several anomalies in our test results with the ASUS board. We will update our performance results once we have verified our test results with ASUS. Our X1900XT CrossFire setup ran fine on each board with the 6.8 Catalyst driver set and we will have test results in our full review.
All Core 2 Duo benchmarking used a Corsair PC-6400C3 2GB kit at 3-4-3 timings at 2.2V. In recent months the memory market has moved from a 1GB kit to a 2BG kit being the common memory configuration. Our new DDR2 test standard is 2GB. We tested with a retail E6600 and will follow up in our next 975X article with the E6700 processor utilized in our standard test bed configuration. All results are reported in our charts and color-coded for easier identification of results.
The performance of the board was exemplary in the majority of our tests. We found the board to be a very consistent performer and extremely stable up to its overclocking limit. Unfortunately, we do not know what the true limit of this board will be until we see the production level BIOS. As with recent Abit boards we will see significant BIOS tuning for the performance oriented crowd after stability or incompatibility issues have been resolved.
This board supports up to 2.65V on the memory that will allows for a great amount of overclocking headroom, though we would like to see additional .05V increments over 2.35V. The BIOS we tested only allowed adjustment of four memory timings but we expect to see additional options as the BIOS matures. We expect great things out of this board if Abit is able to release a production level BIOS for this board that keeps its current performance levels, addresses the issues we found in early testing, and is able to improve overclock levels.
Overclocking
Abit AW9D-MAX Overclocking Testbed |
|
Processor: | Intel Core 2 Duo E6600 Dual Core, 2.4GHz, 4MB Unified Cache 1066FSB, 9x Multiplier |
CPU Voltage: | 1.55V (default 1.3V) |
Cooling: | Scythe Infinity Air Cooling |
Power Supply: | OCZ GameXStream 700W |
Memory: | Corsair Twin2X2048-PC2-6400C3 (2x1GB), 2.45V (Micron Memory Chips) |
Video Cards: | 1 x EVGA 7900GTX |
Hard Drive: | Seagate 320GB 7200RPM SATA2 16MB Buffer |
Maximum OC: (Standard Ratio) |
423x9 (4-4-4-10, 1:1), CPU 1.575V, MCH - 1.85V 3806MHz (+58%) 416x9 (3-4-3-10, 1:1), CPU 1.575V, MCH - 1.85V 3743MHz (+56%) |
Click to enlarge |
We were really surprised by our overclocking results on this board with a beta BIOS. Like other recent boards we had to gradually increase the FSB speeds while adjusting voltages and memory settings in order to reach this level. Our retail E6600 was able to boot into Windows XP at 9x438 but we consistently had benchmark failures until we backed off to 9x431. While it would have been easy to take screenshots at this level with several popular benchmarks, our voltages and memory settings were outside of the range we would consider safe for air cooling over any length of time. We backed our settings down to 9x423 where we achieved a level that was consistent with a machine that would be safe running 24/7. In fact, due to improved memory latencies offering better overall system performance our ideal setting was at 9x416 with 3-4-3-10 memory settings on this board.
We were able to boot into Windows at 448 FSB with our X6800 set at an 8X multiplier, pass all benchmarks at 8x440, and could POST at up to 460FSB. We tried our E6300 and were able to run at a system stable setting of 7x448. Our board, BIOS, and memory combination did not allow us to POST past 460FSB. Overall the board was the equal of or better than the ASUS P5W-DH in overclocking. Since the memory remained at a 1066 strap during overclocking on the Abit board we found the general performance was slightly better than the ASUS past 416FSB.
Test Setup
Abit AW9D-MAX Performance Test Configuration |
|
Processor: | Intel Core 2 Duo E6600 (2.4GHz, 4MB Unified Cache) |
RAM: | 2 x 1GB Corsair Twin2X2048-6400C3 Tested at DDR2-800 3-4-3-10 2.2V |
Hard Drive: | Seagate 320GB 7200RPM SATA2 16MB Buffer |
System Platform Drivers: | Intel - 8.0.1.1002 |
Video Cards: | 1 x EVGA 7900GTX |
Video Drivers: | NVIDIA 91.47 |
CPU Cooling: | Scythe Infinity |
Power Supply: | OCZ GameXstream 700W |
Motherboards: | ASUS P5W-DH Deluxe (Intel 975X) Bios 1305 Abit AW9D-MAX (Intel 975X) |
Operating System: | Windows XP Professional SP2 |
Test conditions were maintained the same, as much as possible, over the platforms tested. For better comparison standard test results (1280x1024) were run with a single EVGA 7900GTX on all platforms. Our high resolution game tests were run at settings of 1600x1200 4xAA/8xAF with the EVGA 7900GTX card. We had planned on presenting 7950GX2 numbers but noticed several anomalies in our test results with the ASUS board. We will update our performance results once we have verified our test results with ASUS. Our X1900XT CrossFire setup ran fine on each board with the 6.8 Catalyst driver set and we will have test results in our full review.
All Core 2 Duo benchmarking used a Corsair PC-6400C3 2GB kit at 3-4-3 timings at 2.2V. In recent months the memory market has moved from a 1GB kit to a 2BG kit being the common memory configuration. Our new DDR2 test standard is 2GB. We tested with a retail E6600 and will follow up in our next 975X article with the E6700 processor utilized in our standard test bed configuration. All results are reported in our charts and color-coded for easier identification of results.
56 Comments
View All Comments
yyrkoon - Thursday, September 21, 2006 - link
Heya Gary, seems this motherboard is availible from newegg in the U.S. currently (which also seems to have the lowest price, even lower than ZZF, and mwave currently), are they still using the beta BIOS marked as a production BIOS or what ?Also, I noticed my question concerning the SATA port multiplier compatability never realy got answered fully ;) However, I DO realize that you guys are probably very busy :)
Gary Key - Wednesday, September 27, 2006 - link
The 1.2 beta we used is now official. Should have another beta bios update late next week. I will get to the eSATA question this weekend. Have a new external SATA setup that will make for the perfect test.biggersteve - Monday, September 18, 2006 - link
Gary, can you comment on when you hope to publish the upcoming P965 shootout?Gary Key - Tuesday, September 19, 2006 - link
I hate to give a date as I have already moved the article out twice. It should be within a week, just received two boards that are both exclusives along with one more coming tomorrow that I will need to get into the article. Expect 12, maybe 13 boards and a novel size review.BadThad - Tuesday, September 12, 2006 - link
I'd like to see information on the board components used like capacitors. Any board is only as good as it's weakest component. Also, how many phases is the power control?Gary Key - Wednesday, September 13, 2006 - link
http://www.anandtech.com/mb/showdoc.aspx?i=2829&am...">Abit AW9D Features PageI know it is a boring page but most of your questions are answered in the second paragraph. :) The board features a four phase power regulation setup and solid aluminium electrolytic polymer capacitors. The manufacturer of the capacitors will vary at times so until we have a definite word from Abit on the subject we hazard to guess which ones they will use. However, the difference in the quality of solid aluminium electrolytic polymer capacitors between suppliers is minor at this time when comparing the quality of traditional aluminium electrolytic wet capacitors.
BadThad - Thursday, September 14, 2006 - link
DOH! OK, yea, I do a lot of skimming because I don't have much time....good article BTW! :)Why don't reviewers start pressuring the mfg's to make a REAL enthusiast MB? I want a board with no intregated sound, video nor LAN....no integrated ANYTHING dasmit!
yyrkoon - Friday, September 8, 2006 - link
I see ABIT is still using *some* non grounded motherboard retaining holes on thier boards, I cant help but wonder WHY they are doing this. I'm fairly sure my Asrock AM2NF4G-SATA2 board has all motherboard screw holes 'tinned' (and grounded?), but even looking at my old ABIT NF7-S2G board, there are atleast two screw holes that have no 'tinning'.Is this to help with noise, or is there something else I'm missing ?
SocrPlyr - Saturday, September 9, 2006 - link
That isn't for grounding. It is actually not connected to anything. the point of the metal around the screw holes is to prevent the PCB from cracking when the screws are tightenedyyrkoon - Saturday, September 9, 2006 - link
Thats funny, because its been known for a long time, that when using a ABIT motherboard, you DO NOT put metal screws in those holes . . .