![[H]ardOCP](/images/blank.gif){kind=small}
- Date:
- Thursday , June 05, 2008
- Author:
- Morry Teitelman
- Google +1
ASUS P5Q3 Deluxe
Based on the newly introduced Intel P45 chipset, the P5Q3 Deluxe board seems to have a good mix of power and features. ASUS looks to have a winner on their hands with this one, but does the P45 offer anything that the X35, X38, or X48 do not?
Subsystem Testing
NOTE: For all Subsystem Testing, an Intel LGA 775 E8500 Core2 Duo CPU with a 1333MHz FSB and 2 x 1GB Crucial Ballistix DDR3-2000 memory modules running at 1066MHz were used in testing.
Audio – Subjective Listening
One of the easiest ways to determine the quality of the audio subsystem is via a subjective sound test. Ideally, a sound test requires audio covering the entire spectrum, from subtle to intense. For this test, I chose to listen to the newly released Disturbed album Indestructible.
As expected, audio playback of the selected album was without issue using the board’s onboard audio subsystem, making for an enjoyable audio listening experience.
Audio – Microphone Port Testing
The MIC-IN input was tested using our standard Labtec Desk Mic 524 testing microphone. Spoken words were recorded and played back using Microsoft Sound Recorder, with the Microphone Boost option disabled and enabled. The Microphone Boost option is found within the Advanced menu under the Microphone section within the Volume Control menu.
The overall clarity and audibility of the recorded voice clip was the same with and without Microphone boost enabled, which is a sign of a good audio subsystem implementation. In both cases, no audio distortion was detected.
Drive Performance
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 to both ports. SATA and IDE drive tests were performed using Maxtor 40Gb ATA 133 model 6E040L0 hard drives on the IDE headers and Samsung 40GB SATA II hard drives on the SATA headers. The SATA drives were used for testing in a RAID 0 16k block size configuration on the Intel ICH10R Southbridge controller and in standalone mode on the Intel ICH10R Southbridge and Marvell controllers. Testing was also conducted using an IDE drive in a primary slave configuration on the Marvell controller. All drive benchmarks were done using the open source Iometer program
Overall, the SATA 2 based RAID 0 array took the performance crown. However, the standalone SATA drives were a very close second performance-wise to the RAID 0 array, with neither controller dominating. As for the external device tests, the IEEE 1394 drive showed a very slight performance advantage over the USB 2.0 disk. In all cases, the CPU utilization remained under 1%.
Network Utilization Tests
Hagel Technologies’ DU Meter software was used in conjunction with Windows Task Manager to measure the performance of the Marvell Yukon GigE NICs, and the ASUS 802.11n wireless NIC. Note that both Marvell Yukon GigE NICs performed on par with one another. DU meter was used to measure bandwidth, with Windows Task Manager 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 integrated 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 wireless testing was done by connecting to the host system though a configured 802.11n capable router at a 150Mbps data rate. Due to the bandwidth limitations of wireless and the router having gigabit wired ports, there was no concern of bandwidth loss due to router interference.
Marvell Yukon GigE controller
The large file transfer results left me speechless, with both the download and upload speeds clocking in at some of the fastest we’ve seen in the bunker. The average download speed came in at a blazing 43 MB/s, with the upload speed trailing it by 16 MB/s at an impressive 27 MB/s. More impressive was the CPU utilization, which did not break 20% during either test.
The small file transfer results were equally impressive, with the average upload speed besting that of download by almost 8 MB/s. Again, the CPU utilization did not stray above 20% during either test.
ASUS 802.11n wireless controller
The large file transfer results for the 802.11n controller were impressive, with the average upload besting that of download by almost 1 MB/s, coming in at a speed to over 7 MB/s. The CPU utilization remained at or below 20% for the duration of both tests. While the average file transfer rate numbers may seem low when compared with the wired GigE results, keep in mind that this was done without wires of any kind and that the average 802.11g speeds are approximately 1.5 MB/s.
The small file transfer results closely mimicked those of the large file test, with the average upload speed coming in slightly faster than that of download at just under 7 MB/s. As seen with the previous test, the CPU utilization again remained at or below an impressive 20%.
Test Systems
The following system configurations were used for the system benchmark graphs, as well as all graphs listed under the Application and Gaming Benchmarks sections:
Graphs are labeled as follows: Motherboard - CPU Clock - Memory Type
SiSoft Sandra Memory Bandwidth Benchmark
Note that all results above were obtained running the installed memory in Dual Channel mode.
The P5Q3 Deluxe takes a solid second place in our comparison. It is worth noting here that the ASUS P5Q3 is the only board in this graphing using 4GB of memory (2Up Configuration). Our testing has shown that in our specific testing suite that 4GB and 2GB makes little difference in scoring. We are going to scale to 4GB test configurations but this is one of our first boards to show.
We made the attempt earlier last year, but hit a brick wall when it came to test 4Up 1GB configurations as many boards simply have a problem with that many DIMMS consisting on that much memory. Even today, Intel will note that using 4 DIMMS on its X48 platform will make overclocking beyond 1600MHz DDR3 speeds nearly impossible. We have found this to be true of many motherboards and chipsets in the market.
SiSoft Sandra CPU Operation Benchmark
Here we see the P5Q3 Deluxe take last place among the Intel boards, matching performance almost identically with the other Intel P45 chipset based board. It would seem that the X48, X38, and even the 780i show to be much better contenders in this very synthetic benchmark. Had we not compared the two P45 boards side-by-side, I might has thought we had an issue, but with both P45 boards from two different builders scoring nearly identically, I would have to guess we are on the mark.
Super PI Mod v1.5 Benchmark
The P5Q3 Deluxe performs on par with the rest of the Intel systems, with no real speed differentiation between the lot of them. This is more of a reality based synthetic as well in that the processors are calculating Pi on different sets of integers.
wPrime
Again we see the P5Q3 Deluxe easily staying close with the leading Intel solutions in this multithreaded square root calculator having nothing to do with prime numbers as you might guess.