TMPGEnc MPEG Encoding

We will encode a 150mb AVI file to MPEG2 (a somewhat realistic chore as DVD's are MPEG-2). For the AVI to MPEG-2 I used a bitrate of 5000k/Sec, as this is the midrange for a DVD, which is typically between 1000k/Sec to 10,000k/Sec. I used a frame size of 720x480 (DVD Std) and 16:9 NTSC. Note that lower scores are better.

Lower times are better

The ASUS motherboard was outperformed by the AI7 in these tests, not by much, but enough if you do lot of video encoding.

Business Winstone 2004

This is a new test for VL, and from what I could see running, well flying by, as the test's proceeded, appears to be a real world attempt. This benchmark launches MS Office applications creating web pages and moving pictures and text in-between word, excel and power point. Something one might actually do (well, maybe with the exception of Front Page).

At the stock 2.4GHz speed, the AI7 outperforms the P4C800-E, by one point.

Unreal Tournament 2003

UT2K3 s a real system killer and can bring many systems to its knees. We used the [H]ardocp UT2003 Benchmarking utility version 2.1, which are excellent tools in testing various resolutions and detail levels. We selected the CPU test, which uses the dm-inferno map.

The AI7 performed well in UT2003 tests, outperforming the P4C800-E by just over 1% at overclocked speeds, but trailing at stock speeds by just under 2%.

Subsystem Testing

The first thing we'll check is the audio. We downloaded and installed Audio Winbench to test its CPU utilization. CPU utilization was fairly high throughout the DirectSound3D tests. CPU utilization never got to 15%, but it did average in the 8% - 11% range, which is a lot higher than the <2% averages we've seen with the nForce 2. It's not the end of the world, given that the slowest CPU you would ideally be using is a 2.4GHz "C", but we would have liked to have seen lower numbers this late into the Realtek chipsets lifecycle.

This is a synthetic benchmark though, and since I know all of you enjoy a game or two, let's see how the sound will affect UT 2003 performance.

The [H]ardocp Tool has an option to enable and disable sound during testing. Tests will be done with the same hardware configuration as the rest of the benchmarks, except we'll only be displaying the Pentium 4 2.4GHz "C" numbers.

UT 2003, Minimum Detail, 640x480 Resolution

It's plain to see that the CPU takes quite a hit when enabling the onboard sound. Keep in mind though that the settings used are not going to be ideal if eye candy is important to you. Lets see what happens when we setup UT2003 to how a typical gamer would use it.

UT 2003, Maximum Detail, 1280x1024 Resolution

When it comes down to it, at high resolution and detail levels, the onboard sound's CPU utilization will not be a factor at all.

In terms of sound quality, I found gaming to be very enjoyable. I watched some snippets from "Pirates of the Caribbean", listened to a little mix of music (Jazz to Pop to Alternative to Heavy Metal) and found the sound quality to be excellent. Not that coming out of the included onboard S/PDIF jack to my Wireless Acoustic Research Headphones had anything to do with it.

Hard Drive Performance

With the introduction of the more cost effective 7200rpm SATA drives, I snatched up a Hitachi Deskstar with a 8mb buffer (yes, it is the old IBM Deathstar, but supposedly that has been fixed), this should give everyone a better feeling for what performance increase, if any, there is between similar ATA-100 and SATA drives. For this test, we use HD Tach Version 2.61.

PATA	SATA

These results speak for themselves; the SATA drive more than outperforms the ATA-100. At nearly 10,000kps difference and the 7200 RPM SATA drives at close to the same price as ATA-100, I won't be buying any more ATA-100's, you?

Network Performance

We used DU Meter to test the networking speed, and Windows Task Manager for CPU usage. We copied a variety of install files, varying in size of 300kb to as much as 150MB per file from the AI7 machine, to my NF7 box. In between the two I tested with a crossover cable initially and than an Extreme Networks Summit 200-24 (oh the advantage of working for an integrator :p). I used the switch because I wanted to make sure that one of the NIC's was not limiting the other, which it appeared they were. The Extreme Networks Switch has built in buffers, therefore you would be hard pressed to overdrive the 100mb port. Here are the results from both an upload and download (as seen from the AI7) of the same files.

Upload	Download

As you can see, the AI7 appears to be taking a slightly larger hit on the download side, although the throughput was higher. The CPU utilization reached as high as 18% on the close to 10mb/Sec download, and 12% on the 8mb/Sec upload. These numbers are not as bad as a stand alone NIC, but we have to wonder if there is a driver issue here that is getting us a little to close to that 20% mark.

Final Words

ABIT has once again given us a great board to play with without making us take out a loan. With all of the built in features and overclockability, this motherboard is anything but cheap. I enjoyed the layout and attention to detail to things as minute as positioning of the headers and memory slots to reduce interference and allow someone as messy as I am to route cables neatly.

There aren't many problems we see with this board, other than the relatively high CPU usage with the Realtek NIC. Intel's CSA solution would have been a better choice, and provide Gigabit performance to go with it. Though we appreciate the features of the µGuru, the software based ABIT OC Guru isn't the best choice for serious overclockers. Thankfully, ABIT's Softmenu is as solid as ever, and works great in this area.

I was impressed to see this implementation of the Springdale chipset actually outperform the Canterwood based ASUS in the majority of the benchmarks, and the differences where it did not come out on top, where minimal at most. At , this would be a great purchase for the cost effective bargain hunter and enthusiasts alike. I had a Canterwood (an ABIT IC7) in my main rig before swapping it out for the AI7, and after using this board, it's safe to say that the Canterwood is not going back in.

Pros
Great Performance at a reasonable price
Everything is included, but Video, Memory and CPU
Well thought out layout of components and connectors (except floppy)
Great Overclockability

Cons
NIC uses a little more CPU resources than most boards in this class
ABIT OC Guru failed frequently
Those damn grounding pins on the I/O plate (ok, ok, I am getting picky)

Bottom Line: The ABIT AI7 is a good all around motherboard, it would be hard for me to pigeonhole the AI7 as a cost effective solution (it is), or to say it is a great performer (it is) or to even place it as a enthusiasts main board (it could be). The AI7 needs, no, it requires your attention if you are considering a new rig using the Intel P4 line.

If you have any comments, be sure to hit us up in our forums.

HOME