The 755A01-6EKRS may have one of the most awkward names around, but don’t let that discourage you. With Gigabit Ethernet, two Serial ATA RAID options, and a generally tweak-friendly BIOS, the 755A01 serves up plenty of perks for less than $90. Can this budget wonder keep up with more expensive and established Athlon 64 offerings? Read on to find out.
Before we take a pictorial tour around the 755A01, let’s take a peek at the board’s full spec sheet.
|CPU support||Socket 754-based Athlon 64 processors|
|PCI slots||5 32-bit/33MHz|
|AGP slots||1 AGP 4X/8X (1.5V only)|
|Memory||3 184-pin DIMM sockets
Maximum of 3GB of DDR266/333/400 SDRAM
|Storage I/O||Floppy disk
2 channels ATA/133
2 channels Serial ATA 150 via SiS964 south bridge with RAID 0,1 support
2 channels Serial ATA 150 via Sil 3112 SATA RAID controller with RAID 0, 1 support
|Audio||6-channel audio via SiS964 integrated audio and ALC655 codec|
|Ports||1 PS/2 keyboard
1 PS/2 mouse
4 USB 2.0 with headers for 4 more
1 IEEE 1394 port via VT6307 Firewire controller
1 RJ45 10/100/1000 Gigabit Ethernet via Realtek RTL8110S
1 front out
|Bus speeds||CPU: 200-232MHz in 1MHz increments|
|Bus dividers||SRC/ZCLK/AGP/PCI: 100/133/66/33, 100/150/75/37, 100/133/80/40|
|Monitoring||Voltage, fan status, and temperature monitoring|
In terms of features and peripherals, the 755A01 doesn’t bring anything new to the table. What makes the board impressive is the fact that it’s sporting Gigabit Ethernet and two Serial ATA RAID options for under $90. Throw in a little Firewire and some six-channel audio, and the 755A01 touches every peripheral base except ATA/133 RAID.
With so many integrated peripherals, it takes a little work to squeeze everything onto the board. Still, Foxconn has done a pretty good job with the 755A01’s layout. The board doesn’t look half-bad, either.
Other than an abundance of blue, the 755A01’s aesthetic doesn’t appear to follow much of a consistent theme. Not that I mind, though. Of the half dozen ATX cases I have in the Benchmarking Sweatshop, only one has a window. And really, the 755A01’s color scheme doesn’t have to be fashionable to be functional. With important ports and slots identified by unique colors, it’s easy to spot where everything is. I have to admit a certain attraction to the board’s celeste AGP slot and capacitors, even if it’s pure coincidence that the color matches Bianchi’s classic road bikes.
On a little less superficial front, the 755A01’s power connector placement could use some work. Both power plugs are a ways from the top edge of the board, creating potential for a mess of cables around the CPU socket. Careful routing, a sheathed power supply cable, and a couple of zip ties should keep cable clutter to a minimum, but it would take less work if the power plugs lined the top edge of the board.
Like most Athlon 64 boards, the 755A01’s CPU socket is surrounded by AMD’s new heat sink retention bracket. The bracket is screwed onto a metal plate that sits under the board, and the whole assembly can be removed to accommodate non-standard heatsink designs.
The 755A01 has loads of room around the AGP slot for longer graphics cards, and there’s more than enough clearance to swap out DIMMs without removing the AGP card. Unfortunately, I’m not so keen on the AGP slot’s retention mechanism, which can be awkward to reach with cards that have larger heat sinks, like the Radeon 9800 XT and GeForce FX 5950 Ultra. The retention mechanism holds AGP cards in place just fine, but my fingers are too stubby to disengage the clip easily.
Like just about every other Socket 754 Athlon 64 board, the 755A01 serves up three DIMM slots. Each slot supports a gig of memory, but the only 1GB DIMMs Foxconn has qualified for the board are DDR333 sticks. If you want to run DDR400 modules, you’ll have to roll the dice with unqualified memory sticks or settle for 512MB per slot.
In the storage department, the 755A01 clusters its ATA/133 and Serial ATA ports along the edge of the board where they’ll create the least amount of cable clutter. I’m not wild about the hot pink color for the SATA ports, though, if only because Foxconn doesn’t bundle matching cables with the board.
The 755A01’s port cluster is loaded with goodies, including PS/2 mouse and keyboard ports, parallel and serial ports, an Ethernet jack, a Firewire port, four USB ports, and a trio of audio plugs. There are on-board headers for an additional four USB ports and one more Firewire port, though Foxconn doesn’t include any header hardware in the box. Given the popularity of case-mounted peripheral ports and USB memory card readers, I’m inclined to forgive Foxconn for not including PCI back plate expansion ports with the board. Keep in mind that this is a sub-$90 package.
SiS755 in the house
The 755A01 is anchored by SiS’s 755 chipset, which is made up of the SiS755 north bridge and SiS964 south bridge.
Since the Athlon 64’s memory controller is integrated right onto the processor, there’s not much going on in the SiS755 north bridge. The chip features an AGP 8X interface and 16-bit/800MHz bidirectional HyperTransport processor link, but that’s pretty much it.
The north bridge hooks up with the SiS964 south bridge via SiS’s MuTIOL interconnect, which offers over 1GB/sec of bandwidth between the two chips. That should be plenty of bandwidth for the SiS964’s integrated peripherals, which include a two-channel ATA/133 controller, a two-port Serial ATA controller with support for RAID 0 and 1 arrays, eight USB 2.0 ports, and six-channel audio.
Output from the SiS964’s audio controller is fed through Realtek’s ALC655 codec, which can handle 16-bit audio at a sampling rate of 48kHz. Don’t expect much in the way of fidelity, though; the sound quality of integrated motherboard audio tends to be average at best.
To complement the SiS964’s integrated Serial ATA RAID controller, the 755A01 also comes with two additional SATA RAID ports courtesy of Silicon Image’s Sil3112. Unfortunately, while the SiS964’s Serial ATA ports can bask in the glory of direct access to the chipset’s MuTIOL interconnect, the Sil3112 is stuck sharing PCI bus bandwidth with other devices.
Speaking of other PCI devices, the Sil3112 shares the 755A01’s PCI bus with Realtek’s RTL8110S Gigabit Ethernet controller and VIA’s VT6307 Firewire chip. That’s three devices on the board’s shared 133MB/sec PCI bus already, setting the stage for potential bandwidth sharing issues when multiple peripheral devices are used at the same time, or when additional devices are added to the board’s PCI slots.
When I first dipped into the 755A01’s BIOS, I wasn’t expecting much. After all, one only expects so much tweaking potential from a budget board. However, the 755A01’s BIOS was a pleasant surprise.
With control over just about every memory timing option imaginable, the 755A01’s starts out well. The board’s memory tweaking options are similar to those offered by Abit’s KV8-MAX3, which is an excellent lead to follow in the BIOS department.
The 755A01’s tweaking options go deeper than just memory timings. The board’s BIOS also exposes an asynchronous AGP/PCI clock that should keep peripherals in-spec when running on an overclocked CPU bus. The BIOS offers CPU bus speeds between 200 and 232MHz in 1MHz increments, so the async AGP/PCI clock should come in handy. However, the BIOS’s total lack of voltage adjustment options ruins an otherwise hospitable overclocking environment. Foxconn plans to offer voltage adjustment options in future products, but it’s unclear whether such functionality will make its way into a BIOS update for the 755A01.
Unfortunately, the 755A01’s limited voltage options prevented me from getting my Athlon 64 3200+ running on a high enough CPU bus to test the board’s async AGP/PCI clock. The board doesn’t support CPU bus speeds below 200MHz, either, leaving me no way to verify that the async clock actually works.
The lack of voltage tweaking options aside, the 755A01’s BIOS finishes on a positive note with a user-configurable shutdown temperature and temperature-controlled “smart fan”. The smart fan is a particularly useful addition to the BIOS. In a perfect world, users would be able to set temperature thresholds and fan voltages for the smart fan, but Foxconn’s implementation is a good start.
Our testing methods
All tests were run three times, and their results were averaged, using the following test systems.
|Processor||Athlon 64 3200+ 2.0GHz|
|Front-side bus||HT 16-bit/800MHz downstream
HT 16-bit/800MHz upstream
|HT 16-bit/600MHz downstream
HT 8-bit/600MHz upstream
|HT 16-bit/800MHz downstream
HT 16-bit/800MHz upstream
|HT 16-bit/800MHz downstream
HT 16-bit/800MHz upstream
|Motherboard||Abit KV8-MAX3||Chaintech Zenith ZNF3-150||NVIDIA reference||Foxconn 755A01|
|North bridge||VIA K8T800||NVIDIA nForce3 150||NVIDIA nForce3 250GB||SiS755|
|South bridge||VIA VT8237||SiS964|
|Chipset drivers||VIA Hyperion 4.51||NVIDIA ForceWare 3.13||NVIDIA 4.08||AGP 188.8.131.520
|Memory size||512MB (1 DIMM)|
|Memory type||Corsair XMS3500 DDR SDRAM at 400MHz|
Western Digital Raptor WD360GD Serial ATA hard drive
|Graphics||ATI Radeon 9800 XT with Catalyst 4.3 drivers|
|OS||Microsoft Windows XP Professional|
|OS updates||Service Pack 1, DirectX 9.0b|
Today we’ll be testing the 755A01’s performance against retail nForce3 150 and K8T800 boards and NVIDIA’s reference platform for the new nForce3 250Gb.
We used the following versions of our test applications:
- SiSoft Sandra Standard 2003
- ZD Media Business Winstone 2002 1.0.1
- TCD Labs HD Tach v2.61
- Futuremark 3DMark03 Patch 340
- Quake III Arena v1.31
- Unreal Tournament 2003 demo
- RightMark Audio Analyzer 5.3
- RightMark 3D Sound 1.01
- Cinebench 2003
- Sphinx 3.3
The test systems’ Windows desktop was set at 1024×768 in 32-bit color at a 75Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests. Most of the 3D gaming tests used the high detail image quality settings, with the exception that the resolution was set to 640×480 in 32-bit color.
All the tests and methods we employed are publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.
The 755A01 takes top honors in Sandra’s memory bandwidth test and isn’t far off the lead in Cachemem’s bandwidth and latency tests. Overall, the Athlon 64’s on-die memory controller keeps the scores pretty close.
Disk controller performance
Our disk controller performance tests use a Maxtor 740X-6L 7,200RPM hard drive for “parallel” ATA (PATA) and a Western Digital Raptor WD360GD 10,000RPM hard drive for Serial ATA (SATA). Because we use different drives for PATA and SATA, scores aren’t comparable between the two. PATA scores should only be compared with each other. The same goes for SATA scores.
Moving to our disk subsystem tests, the 755A01’s read burst speed performance is somewhere between average and poor, depending on whether you use the SiS964 south bridge or the PCI-bound Sil 3112 SATA controller. However, the 755A01 finds redemption in HD Tach’s transfer rate tests, where the board’s Serial ATA performance is a cut above the competition, regardless of which SATA controller is used.
The 755A01’s Serial ATA controllers manage slightly faster access times than the competition, too. Unfortunately, the board’s SATA controllers also have higher CPU utilization than the rest of the pack.
Foxconn finds itself at the front of the pack in both Winstone benchmarks, though the 755A01 really only distances itself from the competition in the Multimedia Content Creation test.
The board does very well in our gaming tests, too. The 755A01 is a little behind the curve in Quake III Arena, but even then it’s neck and neck with Abit’s popular KV8-MAX3. That’s pretty good company to keep.
Scores are all pretty close in Cinebench….
Sphinx speech recognition
And in Sphinx.
The 755A01 performs relatively well in RightMark’s 3D audio CPU utilization tests. For some reason, the board crashes out of the 64-sound 3D Hardware test, but it’s otherwise competitive with VIA’s integrated audio.
For RightMark’s audio quality tests, I used a Terratec DMX 6fire 24/96 for recording. Analog output ports were used on all systems. To keep things simple, I’ve translated RightMark’s word-based quality scale to numbers. Higher scores reflect better audio quality, and the scale tops out at 6, which corresponds to an “Excellent” rating in RightMark.
Because our nForce3 250Gb reference board is an engineering sample that may use a different codec and audio components than retail products, we’ve left it out of audio quality tests.
The 755A01’s audio quality is only average, but for integrated motherboard audio, average is actually pretty good. Still, discrete 24-bit audio cards tend to provide far superior audio quality.
Our USB and Firewire transfer speed tests were conducted with a USB 2.0/Firewire external hard drive enclosure connected to a 7200RPM Maxtor 740X-6L hard drive. Since the nForce3 250Gb reference board doesn’t have a Firewire port, it’s riding the pine in our Firewire tests.
The 755A01’s USB transfer rates are disappointing, but it gets worse: CPU utilization during those slow transfers is more than 50% higher than the competition. Is USB really that hard to get right?
Fortunately, the 755A01 seems far more comfortable in our Firewire tests. Transfer rates are more competitive, and CPU utilization isn’t outrageous.
We evaluated Ethernet performance using the NTttcp tool from the Microsoft’s Windows DDK. The docs say this program “provides the customer with a multi-threaded, asynchronous performance benchmark for measuring achievable data transfer rate”. Sounds like what we’re after.
We used the following command line options on the server machine:
ntttcps -m 4,0,192.168.1.25 -a
..and the same basic thing on each of our test systems acting as clients:
ntttcpr -m 4,0,192.168.1.25 -a
Our server was a Windows XP Pro system based on Chaintech’s Zenith 9CJS motherboard with a Pentium 4 2.4GHz (800MHz front-side bus, Hyper-Threading enabled) and CSA-attached Gigabit Ethernet. A crossover CAT6 cable was used to connect the server to each system.
PCI-based Ethernet can’t hold a handle to the nForce3 250Gb’s on-chip GigE, but the 755A01’s Ethernet throughput is at least competitive with other PCI-based implementations.
Update 6/13/2005 We recently discovered that the ntttcp CPU utilization results included in this review were incorrect. The CPU utilization results have been removed, but they didn’t factor prominently into our overall conclusion, so that remains unchanged. A full explanation can be found here.
Unfortunately, I was unable to successfully overclock my Athlon 64 3200+ on the 755A01. I’ve had the same processor overclocked on other boards, but rarely without a little extra voltage. Unfortunately, the 755A01’s BIOS doesn’t allow adjustments in processor voltages.
If you plan to do any overclocking or regularly use high-bandwidth USB devices, skip the Foxconn 755A01. It’s that simple. Without voltage options, you’re overclocking with one hand tied behind your back, and the USB performance we observed really speaks for itself.
Voltage options and USB performance aside, the 755A01 is a pretty good board at a really great price. For only $87, you get four ports of Serial ATA RAID, Gigabit Ethernet, a great array of memory tweaking options, and generally excellent application and Serial ATA performance. That’s a whole lot of value for a Socket 754 motherboard, and perfect for budget-minded enthusiasts trying to scrape together enough cash for an Athlon 64 2800+.
Now, if only the entire board were draped in celeste.