Benchmark Setup
Our abit AB9 QuadGT motherboard fully supports the current range of socket 775 Intel processors. We chose an E6300 for testing as we feel this will be a very popular Core 2 Duo CPU choice with the P965 motherboards. (Granted that many people looking at a $200+ motherboard like this abit are likely to opt for a more expensive CPU as well, but with the more budget-oriented P965 boards the choice makes sense, and we continue to use this CPU for purposes of consistency.)
A 2GB memory configuration is now standard in the AT test bed as most enthusiasts are currently purchasing this amount of memory. Our choice of DDR2-800 memory from GEIL offered a very wide range of memory settings during our stock and overclocked test runs. Our memory timings are set based upon determining the best memory bandwidth via our test application results.
We are utilizing the MSI X1950XTX video card to ensure our 1280x1024 resolutions are not completely GPU bound for our motherboard test results. We did find in testing that applying a 4xAA/8xAF setting in most of today's latest games created a situation where the performance of the system starts becoming GPU limited. Our video tests are run at 1280x1024 resolution for this preview article at standard settings.
Synthetic and Application Performance
We are utilizing an abbreviated test suite for our first look at the abit AB9 QuadGT motherboard. We will present full test results in our next article that will look at new P965 boards from ASUS and Gigabyte.
At stock speeds the AB9 QuadGT motherboard posts the best WinRAR and Quake 4 scores, and it's SuperPI and Nero Recode scores are near the top as well. We had performance issues with Battlefield 2 at stock settings as the frame rates were consistently about 5fps below the other boards. The game also stuttered at various points in the benchmark and dropped online connections three out of five times. We finally traced the issue down to the BIOS and JMicron controller. After setting our memory latencies to 3-4-3-10 and loading the JMicron RAID driver instead the standard XP IDE driver our scores improved to 104.8.
Although not listed, we noticed a drop of about 200~300 points in PCMark 2005 when compared to our other P965 motherboards. The solution was once again to switch the JMicron controller from IDE to RAID and load the JMicron RAID driver set. Our scores improved from 5664 to 5962 while our disk intensive WinRAR and Nero Recode scores improved by a couple of percent.
In our overclocking tests the AB9 QuadGT was absolutely superb and offered the best overall performance even though the ASUS P5B-Deluxe has a 70MHz CPU speed advantage. This performance is a direct result of the 1067 strap settings and aggressive memory sub-timings. Stability was also excellent during overclocking, although we continued to have issues with BF2 until we changed the JMicron driver. Once the driver was changed our overclocked BF2 fps score improved to 151.3.
We have not encountered any performance issues with our Quad Core (QX6800), E6600, or E4300 in limited testing to date. We will provide a performance update with these processors in our next P965 comparison article. Overall, this board clearly performed very well in our limited testing.
Our abit AB9 QuadGT motherboard fully supports the current range of socket 775 Intel processors. We chose an E6300 for testing as we feel this will be a very popular Core 2 Duo CPU choice with the P965 motherboards. (Granted that many people looking at a $200+ motherboard like this abit are likely to opt for a more expensive CPU as well, but with the more budget-oriented P965 boards the choice makes sense, and we continue to use this CPU for purposes of consistency.)
Standard Test Bed Performance Test Configuration |
|
Processor: | Intel Core 2 Duo E6300 Dual Core, 1.86GHz, 2MB Unified Cache 1066FSB, 7x Multiplier |
RAM: | Geil PC2-6400 800MHz Plus (2x1GB - GX22GB6400PDC) (Micron Memory Chips) |
Hard Drive: | Seagate 320GB 7200RPM SATA2 16MB Buffer |
System Platform Drivers: | Intel - 8.1.1.1010 |
Video Cards: | 1 x MSI X1950XTX |
Video Drivers: | ATI Catalyst 6.11 |
CPU Cooling: | Scythe Infinity |
Power Supply: | OCZ GameXstream 700W |
Optical Drive: | Sony 18X AW-Q170A-B2 |
Case: | Cooler Master CM Stacker 830 |
Motherboards: | ASUS P5B-Deluxe WiFi (Intel P965 C2, 1.01G) - BIOS 0804 abit AB9-Pro (Intel P965 C1) - BIOS 1.5 abit AB9 QuadGT (Intel P965 C2) - BIOS 10 Biostar T-Force 965 Deluxe (Intel P965 C1) - BIOS IP96a803 Gigabyte GA-965P-DQ6 (Intel P965 C2) - BIOS F7 MSI P965 Platinum - (Intel P965 C2) - BIOS 1.2 |
Operating System: | Windows XP Professional SP2 |
. |
A 2GB memory configuration is now standard in the AT test bed as most enthusiasts are currently purchasing this amount of memory. Our choice of DDR2-800 memory from GEIL offered a very wide range of memory settings during our stock and overclocked test runs. Our memory timings are set based upon determining the best memory bandwidth via our test application results.
We are utilizing the MSI X1950XTX video card to ensure our 1280x1024 resolutions are not completely GPU bound for our motherboard test results. We did find in testing that applying a 4xAA/8xAF setting in most of today's latest games created a situation where the performance of the system starts becoming GPU limited. Our video tests are run at 1280x1024 resolution for this preview article at standard settings.
Synthetic and Application Performance
We are utilizing an abbreviated test suite for our first look at the abit AB9 QuadGT motherboard. We will present full test results in our next article that will look at new P965 boards from ASUS and Gigabyte.
Click to enlarge |
At stock speeds the AB9 QuadGT motherboard posts the best WinRAR and Quake 4 scores, and it's SuperPI and Nero Recode scores are near the top as well. We had performance issues with Battlefield 2 at stock settings as the frame rates were consistently about 5fps below the other boards. The game also stuttered at various points in the benchmark and dropped online connections three out of five times. We finally traced the issue down to the BIOS and JMicron controller. After setting our memory latencies to 3-4-3-10 and loading the JMicron RAID driver instead the standard XP IDE driver our scores improved to 104.8.
Although not listed, we noticed a drop of about 200~300 points in PCMark 2005 when compared to our other P965 motherboards. The solution was once again to switch the JMicron controller from IDE to RAID and load the JMicron RAID driver set. Our scores improved from 5664 to 5962 while our disk intensive WinRAR and Nero Recode scores improved by a couple of percent.
In our overclocking tests the AB9 QuadGT was absolutely superb and offered the best overall performance even though the ASUS P5B-Deluxe has a 70MHz CPU speed advantage. This performance is a direct result of the 1067 strap settings and aggressive memory sub-timings. Stability was also excellent during overclocking, although we continued to have issues with BF2 until we changed the JMicron driver. Once the driver was changed our overclocked BF2 fps score improved to 151.3.
We have not encountered any performance issues with our Quad Core (QX6800), E6600, or E4300 in limited testing to date. We will provide a performance update with these processors in our next P965 comparison article. Overall, this board clearly performed very well in our limited testing.
41 Comments
View All Comments
yyrkoon - Monday, January 22, 2007 - link
Oh, Gary, and Jarred, if you're thinking we, your readers are being harsh on you, well perhaps we are to an extent, but speaking for myself, this is because we care, and often look forward to your articles. So think of this as constructive criticism, and not outright flaming, please.Gary Key - Tuesday, January 23, 2007 - link
You are not being harsh. I changed the article back to "Conductive Polymer Aluminum Solid Capacitors" as abit finally confirmed the majority of the capacitors on the board are this type (I looked up the capacitor part numbers before hand but edited the article back the other way). I will update it when they confirm the three capacitors that did not match, they are solid but I have two boards each with different part numbers/suppliers on those three items. At times the manufacturers want us to use their marketing terminology as they might change components during production lot runs based on engineering changes or spot market pricing or to use layman's terms on the website. My previous articles on the Gigabyte boards used the correct terminology based on the capacitors on the board and we had an enormous amount of email and forum traffic asking why we stated something different than on Gigabyte's website at the time. We dumbed it down a little this time after a discussion and should not have. ;) Thanks for the comments and we do listen.yyrkoon - Monday, January 22, 2007 - link
Like the man said, Solid State is used when referring to Integrated Circuits, not Capacitors. The end result, is that you end up 'looking' like a fool, when someone who knows better sees this ( and possibly spread a minor form if 'mis-information).Think about it like this, what is the difference between a NAS, and a SAN ? Would you call a SAN, a NAS, in the company of enterprise IT geeks ? Probably not, at least, not without causing some confusion, or being corrected several times in the process . . .
Operandi - Monday, January 22, 2007 - link
Well they (caps suitable for motherbards) are either AL electrolyte or solid polymer based.Electrolyte caps are the more common type an actually contain electrolytic fluid which can leak when the cap fails. Polymer caps avoid that problem and also last longer -- to the laymen that would be the key difference.
Gary Key - Monday, January 22, 2007 - link
I have used the terminology, "Conductive Polymer Aluminum Solid Capacitors" in past articles and received several emails asking why we do not use what the manufacturers state on their websites as that term was deemed confusing. :)Stele - Monday, January 22, 2007 - link
Part of the problem lies in the confusion about the various kinds of capacitors of that category.Amongst the kind that we're concerned with on motherboards, graphics cards etc, we have the most basic: the aluminium electrolytic capacitor. Because this type started out with liquid electrolyte, the word "liquid" is generally omitted from the name as it is understood.
Then came the solid aluminium electrolytic capacitor, which replaced the liquid electrolyte with, well, a solid one - usually based on aluminium oxide. These are the type that we're nowadays excited about - the lack of liquid makes them more resistant to blow-outs and electrolyte leakage/dry-out, especially under prolonged, high temperature use. For short, they're sometimes referred to as merely "solid electrolytic" ("aluminium" is sometimes left out because, as noted above, the solid electrolyte is usually based on aluminium oxide and hence is understood as such) capacitors. To call them "solid capacitors" isn't totally useful because most capacitors are indeed solid objects :)
In conductive polymer capacitors, on the other hand, the dielectric is made from polymer foil (e.g. polypropylene, polyester, polystyrene, polycarbonate) coated with a layer of metal deposited on the surface. It follows that the basic conductive polymer capacitor has no liquid electrolyte inside - indeed there is no electroylte as the dielectric is purely the polymer foil.
However, manufacturers can and do mix in aluminium electrolyte with the polymer foil to improve certain performance characteristics; the electrolyte is often solid (rather than liquid) aluminium-based, hence "conductive polymer aluminium solid electrolytic capacitor".
As such, the exact names can mean quite distinct types of capacitors, and are not merely loose permutations of words like "electrolytic", "aluminium", "solid" and "polymer"... so if one wants to accurately describe a capacitor being used, one would need to double-check exactly what dielectric is being used in that capacitor :)
Marlin1975 - Monday, January 22, 2007 - link
Thats a understatement. The only thing i don't like about my 965 board is lack of IDE and the use of the jmicron junk.
One of the reasons I am waiting for more 650i boards and the ?last? ati chipset for Intel chips.
LoneWolf15 - Monday, January 22, 2007 - link
It's one of the reasons I'd still chose i975x rather than i965 for a chipset. i975X still has native Intel IDE.Due to recent nVidia chipset/board issues, and past issues with heat production, I'm not sure I'd choose them for an Intel board either, so that leaves the i975X as the only chipset I'd be comfortable with.
Numb3rs - Thursday, January 25, 2007 - link
Honestly, what in a new build would require and IDE interface..? eSATA is important and I am glad it's included. Abit has always made mobos dropping legacy devices no longer used by enthusiasts. Look at their old Abit "MAX" boards.
Why, in 2007 are manufacturers still using serial and parrallel I/O's..? Remove them completly and free the backplane for more useful eSATA, USB..etc
LoneWolf15 - Friday, January 26, 2007 - link
Obviously, you've never configured a Cisco router through its console port (which usually requires a serial port). And perhaps I don't want my list of optical drives confined to only SATA (currently Plextor and LiteOn are my only options, and I don't want one due to price and the other due to writing quality reasons).There are also some known issues with using Symantec Ghost on the JMicron chipset. Unless I hear they are worked out, that's an important thing to me too, and so IDE is still important to me. Just because it isn't useful to you doesn't mean it isn't to a lot of others.