![[H]ardOCP](/images/blank.gif){kind=small}
- Date:
- Thursday , June 19, 2003
- Author:
- Morry Teitelman
- Editor:
- Kyle Bennett
- Google +1
GIGABYTE 8KNXP Ultra
The 8KNXP Ultra heads up GIGABYTE’s i875P line of motherboards. Does this board hold its own against other Canterwoods? Or does it really need to?
Subsystem Testing
Audio – CPU Utilization
The quality of a motherboard’s audio subsystem can be easily determined by measuring the CPU utilization under a variety of circumstances. A well designed audio subsystem should have minimal impact on the CPU while operating, thus freeing up the precious CPU cycles for other tasks such as scene rendering during an online death match. In order to best measure the real world CPU utilization of the audio subsystem, we use Ziff Davis’ Audio Winbench.
Going by the synthetic benchmark numbers alone, GIGABYTE seems to have chosen well by pairing up the Realtek based audio solution with the 8KNXP. The CPU utilization stayed around 4% during testing, with it spiking to almost 6% in a few cases. While in use, the audio subsystem should have minimal impact on other ongoing tasks. Please be aware that these results are based on a synthetic benchmark and may not accurately depict what you see on your home system.
Audio – Subjective Listening
Using a CPU utilization benchmark is an adequate way to measure the overall quality and performance of the audio subsystem from a system level, but cannot in any way measure the quality of audio produced by the subsystem. A good sound test requires a source that covers the full sound spectrum, from subtle harmonics to pounding bass. I chose the Tool Album Lateralus for this test. With this album, Tool provides a nice mix of subtle melody and pulse pounding rhythm.
I was very satisfied with the audio quality from the board. Not once did I detect any skipping, crackling, or hissing during the CD playback. Furthermore, both the high and low frequencies came across crisp and clear.
Audio – In Game Testing
In addition to CD or MP3 playback, users most often rely on the audio subsystem for gaming, whether it be stand alone first person shooter type or online death matching. To adequately test the quality of the audio subsystem during game type scenarios, I used a custom designed benchmarking script based on recorded bot matches developed by our own Brent Justice. The benchmark tool runs through three loops of each demo, one with sound disabled, one with normal 3D sound enabled, and the last with EAX 3D sound enabled.
As expected, the performance suffered more with sound enabled during the low quality visual settings tests than during the high quality visual settings tests. During the high quality tests, there was no difference between the scores with and without sound enabled meaning that the sound subsystem had no impact on the overall game play itself. In addition, the sound quality during the bot match was excellent, especially with EAX enabled. I detected no hissing, crackling, or skipping during the tests with the sound being delivered in a crisp and clear manner.
USB 2.0 / IEEE 1394
In order to adequately test the capabilities of the on board USB 2.0 and IEEE 1394 connections, we chose to use an ACOMDATA HD060U2FE-72-USB 2.0/FireWire HDD connected first to the USB port and after to the IEEE 1394 port in conjunction with TCD Labs’ HDTach program.
As expected, the USB drive had a RAT (Random Access Time) just below 20ms which is not bad for this type of device. The average read speed was about 22 MB/s with the write speed halving the read speed at 11 MB/s. Both average speeds fall short of previous results seen with the drive by almost 5 MB/s. However, the drive does get a good average CPU utilization coming in just over 10%. Normally, the CPU utilization for a USB drive is around 20%.
IDE/ATA Performance
System performance relies very heavily on three major subsystems, the CPU, the system memory, and the system IDE interfaces. In order to test the IDE performance of this board, I used TCD Labs’ HDTach program. My test bench currently uses Maxtor 40GB ATA 133 model 6E040L0 hard drives on the IDE headers. On the SATA headers, I have Seagate 80 GB Barracuda SATA hard drives installed in the test bench. For the SCSI connectors, I have a single Seagate Cheetah X15 36LP hard drive installed on my test bench. Multiple SATA drives were used for testing in a RAID 0 configuration on the ICH5R based ports. For testing on the ITE controller, multiple IDE drives were used in both RAID 0 and RAID 1 configurations. Testing was also conducted using a single SATA drive on the ICH5R controller, a single IDE drive on the ITE controller, a single SCSI drive on the Adaptec controller, and an IDE drive connected in a primary slave configuration.
It looks like the SATA RAID 0 arrays take the speed crown this time. With a 16k block size, the SATA RAID 0 array was unstoppable. With a RAT (Random Access Time) just under 13ms, it beat out the IDE array results by more than 2ms. Further, the SATA 16k array’s RBS (Read Burst Speed) was just shy of 140 MB/s which is no small feat. The IDE RAID 0 array on different cables did beat out the SATA read average by 2 MB/s. However, the SATA 16k array beat out the IDE different cable array by more than 6 MB/s on write tests. The IDE different cable array did have a bit lower CPU utilization than the SATA array, but neither array’s utilization is too bad. All in all, the results seem to say that a 16k block size on either controller is the way to go, as long as you use two different drive cables on the ITE controller.
Both the RAID 1 and JBOD arrays scored fairly evenly across the board. In both cases, the read and write averages were almost identical, coming in at 51 MB/s for read and 28 MB/s for write. Also, the RAT (Random Access Time), RBS (Read Burst Speed), and CPU utilization numbers were all almost identical. Note that a JBOD array is a RAID type array in which two drives are combined in to one larger logical entity. It is similar to a RAID 0 array, but the drives are spanned instead of striped.
The SCSI drive results are nothing short of amazing compared to ATA and SATA scores. One thing the test results do not get across is just how fast these drives are. With a RAT (Random Access Time) of 6ms, they are blazing fast. The average read and write results are ok, but nothing special compared to a standalone IDE drive. However, the CPU utilization of < 2% is amazing, considering the raw speed of the drive. Note that drive performance was the same no matter which on-board port the drive was connected to.
On the IDE side of things, both drives have similar results. Both drives scored an average of 51 MB/s for reads and 28 MB/s for writes, with RAT (Random Access Time) scores of a bit over 15ms as well. The big difference was in the RBS (Read Burst Speed) score, with the ITE based drive scoring about 10 MB/s higher than the primary slave drive. The SATA results were another story though. The standalone SATA drive trailed the IDE drives in both average read and write performance by over 10 MB/s, which is just terrible. Its CPU utilization was on par with that of the IDE drives, with its RAT (Random Access Time) score beating out the IDE drives by a good 3ms.
Network Utilization Tests
Hagel Technologies’ DU Meter software was used in conjunction with Windows Task Manager to measure the performance of the onboard Intel PRO/1000CT NIC. DU meter was used to measure bandwidth, with Windows TaskMan to monitor the CPU utilization on the test system. For the test itself, a 750MB archive file containing various sized .WMA audio files for the large file transfer test and a 750MB worth of various sized .WMA audio files for the small files transfer test were used in conjunction with an Intel Gigabit NIC on the host system, and a crossover cable to connect the host system to the test system. A crossover cable was used to rule out any possible bandwidth losses due to hub or switch passage.
The Intel GigE adapter continues to impress me. Both the upload and download speeds are fast, approaching the sustained speeds of a standalone IDE hard drive. The CPU utilization of 30% is a bit high compared to the IDE drive’s utilization, but well worth it in my opinion.
The large file transfer results mimic those of the small file transfer results with one caveat. The upload speed bested that of the small file test by 10 MB/s. These Intel GigE integrated adapters continue to rule the roost as far as network performance goes.
Test Systems
The following system configurations were used for the Sandra memory benchmark graph, as well as all graphs listed under the Application and Gaming Benchmark sections:
GIGABYTE 8KNXP Ultra (i875P) - Intel Pentium 4 3.0 GHz CPU (clocked at 12x250) and Intel Pentium 4 2.4 GHz CPU (clocked at 12x200) - 2 x 256MB Corsair XMS3200 - ATI Radeon 9700 Pro w/ ATI Catalyst 2.3 drivers - 40 GB Maxtor ATA133 HDD - Allied 400w PSU - WindowsXP w/SP1
NOTE: For all benchmark tests done on the GIGABYTE 8KNXP Ultra motherboard, the optimized default BIOS settings were loaded for the benchmark tests. The following BIOS settings were also engaged during the tests: CAS Latency Time = 2; RAS Precharge delay = 2; RAS-to-CAS delay = 3; Active Precharge delay = 5.
Intel D875PBZ (i875P) - Intel Pentium 4 3.0 GHz CPU (clocked at 15x200) - 2 x 256MB Corsair XMS3200 - ATI Radeon 9700 Pro w/ ATI Catalyst 2.3 drivers - 40 GB Maxtor ATA133 HDD - Allied 400w PSU - WindowsXP w/SP1
ASUS P4P800 Deluxe (i865PE) - Intel Pentium 4 3.0 GHz CPU (clocked at 12x250) - 2 x 256MB Corsair XMS3200 - ATI Radeon 9700 Pro w/ ATI Catalyst 2.3 drivers - 40 GB Maxtor ATA133 HDD - Allied 400w PSU - WindowsXP w/SP1
ABIT IS7-G (i865PE) - Intel Pentium 4 3.0 GHz CPU (clocked at 12x250) - 2 x 256MB Corsair XMS3200 - ATI Radeon 9700 Pro w/ ATI Catalyst 2.3 drivers - 40 GB Maxtor ATA133 HDD - Allied 400w PSU - WindowsXP w/SP1
ABIT BH7 (i845PE) - Intel Pentium 4 2.53 GHz CPU (clocked at 19x133 for 2.53 GHz speed) - 2 x 256MB Corsair XMS3200 - ATI Radeon 9700 Pro w/ ATI Catalyst 2.3 drivers - 40 GB Maxtor ATA133 HDD - Allied 400w PSU - WindowsXP w/SP1
ASUS A7N8X Barton (nForce2): AMD AthlonXP 3000+ CPU (clocked at 13x166); 2 x 256MB Corsair XMS3200; ATI Radeon 9700 Pro, onboard nForce2 10/100 NIC; 40 GB Maxtor ATA133 HDD, Zalman 300w PSU. WindowsXP w/SP1, ATi Catalyst 2.3.
Graphs are labeled as follows: Motherboard - CPU Clock - FSB Clock - Memory Clock
SiSoft Sandra Memory Bandwidth Benchmark
Note that all results above, with the exception of those from the BH7 board, were obtained running in Dual Channel mode.
The 8KNXP Ultra makes a strong showing against the 865PE based boards with a 1000MHz CPU FSB, but is unable to overcome ASUS’ Hyper Path or ABIT’s Game Accelerator memory technology. It seems that PAT may not be all that it was cracked up to be after all. At stock speed with the 2.4 GHz 800MHz FSB CPU, the 8KNXP fell behind the Intel 875 board by a bit of a margin.