Thinking about jumping on the Athlon 64 and SFF bandwagons? Well, Biostar has a small form factor system that will allow you to get on both bandwagons at the same time. It’s the iDEQ 200P, and it offers all the performance of its bigger brothers but at a fraction of the price. Did Biostar hit another SFF homerun or is it more of a stand-up double? Read the review to find out!
The small form factor PC (a.k.a. mini PC) market is in full swing now with new competitors and new products entering the market all the time. One company that must have had a past vision of this current boom in the fledgling SFF market is Biostar. Even though the company has only been in the market for about a year, Biostar has established itself as a leader and innovator. Biostar currently offers six different SFF systems in its iDEQ line. These systems are barebones offerings, including a stylish case and a motherboard. However, Biostar does not stop there. One of the more innovative features of the iDEQ line is its cooling system. Also, each iDEQ features pre-routed wires and cables in an effort to provide the consumer with a quick build.
As you have probably guessed, this review will cover one of the six current iDEQ systems offered by Biostar. Two other iDEQ systems have made their way into Bjorn3D reviewers’ hands, and each was highly praised and awarded with our highest honor – the Bjorn3D Golden Bear Award. Naturally, this made me excited to get my hands on one and see what all the fuss was about. However, as it turns out, I was in store for not quite the same uplifting experience as my fellow reviewers had with the iDEQ 200T and the iDEQ 200N. The iDEQ 200P is one of Biostar’s newest models in the iDEQ family, and it is the one we will be taking a look at today. It features support for AMD’s Athlon 64 processor by way of NVIDIA’s nForce3 150 chipset on Biostar’s own K8BNP motherboard.
Specifications & Features
For a complete list of specs and features, please check out Biostar’s iDEQ 200P web page. I’ve included most of the specs and features below.
|Supports up to AMD Athlon 64 3200+
|Two DDR400 Slots (for up to 2GB total memory)
|1 x AGP (8x)
1 x PCI
|I/O Ports (Front)
|2 x USB 2.0
1 x S/PDIF input
1 x Microphone
1 x Headphone
1 x IEEE1394A
|I/O Ports (Rear)
|2 x USB 2.0
1 x S/PDIF output
1 x IEEE 1394A
1 x LAN
2 x PS/2
2 x Serial
1 x Line-in
1 x Line-out
1 x Mic
|Accomodates up to 3 EIDE devices in Ultra DMA 133 (or lower) mode
|SATA RAID provided by VIA VT6420 chip
|6-channel Audio courtesy of Realtek ALC650
|Provided by VIA VT6307
|10/100/1000 mbps (Gigabit) LAN
|210 x 323 x 187 mm (W x D x H)
Each system in the iDEQ line provides the following features:
- Sliding front cover allows you to cover up any floppy or optical drive that does not match the color of the case
Push down on the tab at the top to reveal drives
- Cables and wires are neatly routed and bundled to minimize their affect on airflow
Cables neatly tucked away
More cables neatly tucked away
- A unique cooling system to keep the CPU and other components as cool as possible (Check it out here)
- Abundance of connectivity options (see table above for list)
Check it out! S/PDIF input
on the front panel!
The usual rear ports
- Removable 5.25″ and 3.5″ drive trays
The 3.5″ cage separates with one thumbscrew
In practice, most of these features actually were beneficial to my experience with the iDEQ 200P. For instance, the nice, clean wiring and cabling job made it easy to remove the primary IDE cable. I did this because the only hard drive in the system has a SATA interface, which meant the primary IDE cable was useless to me, especially since it is routed in such a way that it cannot be attached to an optical drive. That is fine, though, since the other IDE cable is routed to serve that purpose.
Being able to remove the drive trays certainly made installation of other components, as well as the drives, easier. Unfortunately placing the drive trays back into the case and attaching them was not nearly as easy as removing them, but that is at least somewhat expected since the iDEQ’s case is so small.
- iDEQ 200P barebones system
- Biostar K8NBP motherboard
- Black iDEQ SFF case
- Pre-routed cables for floppy and IDE devices
- Drivers/Utilities CD
- User´s Manual
- Installation Guide
- Thermal grease
- Power cord
- Serial ATA cable (with right-angle connector for drive connection to help it fit in the case)
- Serial ATA power cable (4-pin Molex-to-SATA adapter)
- Package of screws
This is what you get in the box
After removing the 200P from its box and packaging, the first thing I noticed was the very nice black, brushed aluminum finish. It’s definitely a nice touch and could make this stand out (even more than it already would) at a crowded LAN party. I hope that some of the pics above show this accurately.
I learned a hard lesson in SFF system building with this system! Planning ahead is crucial before starting the installation of hardware in a small system like the 200P. Obviously, one of the more appealing features of the 200P is its small footprint; however, this can also make installing hardware in it a much more stressful experience than it is with a mid-tower or larger system. Don’t get me wrong. Expecting anything else but a little more need for planning (and stress, if planning is inadequate) would be unrealistic. Basically, it comes down to a series of trade-offs. For example, with this system, installing a video card would probably be easier with the drive trays removed; however, replacing the drive trays may be more difficult because of the video card. You would have a similar story if you reversed the order. It’s a conundrum! The pictures below give you an idea of what you have to work with inside the iDEQ 200P’s chassis.
I think most people look at a system like the 200P and see LAN party written all over it. Of course, this leads to wondering if the upper echelon of video cards can fit into such a small case. The largest boards around seem to be based on NVIDIA’s GeForceFX 5950 Ultra GPU, so installing one of them would probably present the biggest challenge among video cards. If you’re wondering whether it fits, check out the list of hardware I installed in this system:
- AMD Athlon 64 3200+
- Corsair TWINX1024-3200LLPRO
- Reference GeForce FX 5950 Ultra
- Hitachi Deskstar 7K250 120GB Serial ATA 7200RPM Hard Drive w/8MB Buffer
- Pioneer DVD-ROM
- Sony floppy drive
That’s right. The two-slot-hogging 5950 fit into this SFF system; however, since it uses both of the only available slots, you cannot expect to be able to put a TV tuner or some other PCI card in there if you want to stick such a large video card in it. Also, please keep in mind that it did take quite a bit of work to get the board to fit. However, fitting any full-height AGP card in there is not exactly trivial. This may well be the most time-consuming effort in the installation process!
A key element in the iDEQ cooling system is the heatsink-and-fan unit for the CPU. It directs hot air from the CPU toward the back exhaust fan. That way both fans work in tandem to move that hot air out. I had no problem with installing the CPU and the cooling unit on it. I used Artic Silver 5 instead of the thermal grease supplied by Biostar.
A decent chunk of copper…
…a small fan…
…and a not-so-clean surface
And here’s a few pictures of what the system looks like with some hardware in it:
You may be thinking that the video card in the second picture is not a 5959 Ultra, and you are right. It is a 5600 Ultra! I took these pictures before trying to fit the 5950 in the case.
After installing all the hardware, I decided to install 32-bit Windows XP Pro (and Service Pack 1). Installing Windows on a SATA hard drive is a bit more complicated than doing so on a PATA drive. I have done this many times on other systems without a problem; however, the iDEQ 200P posed a problem I had never dealt with before (or I was just lucky those other times!). As usual, I proceeded through the Windows installation and pressed F6 to install third-party drivers (for the SATA controller). All went well, and the installation continued normally until the first reboot. It rebooted and restarted the installation from the beginning! After re-trying the installation many times and some troubleshooting in the BIOS, I figured out that in order to make the installation work, I had to enable the SATA ROM in the BIOS (under Integrated Peripherals -> Onboard Devices -> Onboard SATA ROM). After that was done, the installation proceeded without another hitch! I hope that professing my ignorance here will help someone avoid the grief that I experienced! Please keep in mind that I have performed Windows XP installations on SATA drives on at least two other occasions (one of them being a similar nForce3 system) without a single problem. Unfortunately, Biostar had no help on this issue that I could find in any user’s manual or BIOS manual or anywhere on their many websites.
I think I would be remiss if I didn’t mention a small gripe that I have that will plague anyone who has to clear the CMOS on this system due to setting a BIOS option incorrectly or too aggressively. The CMOS jumper is in a difficult to reach location. I was able to barely squeeze my fingers into the case just right to get to it and put it in the clear position, but I risked dropping it when I did it. I guess that would not have been too big of a deal, since a tweezers would probably allow me to pick it up. A tweezers might have made it easier to deal with the jumper from the beginning, but I was stubborn and just used my fingers. A lot of people probably would not be that lucky, and they will have to actually remove drives and/or expansion boards to get to the jumper.
To provide a broad range of comparisons for the iDEQ 200P, I decided that it would be best to compare the 200P to the previous Athlon 64 systems I have tested and the nForce2 system I compared to them. The system specs are detailed below. Before I present the benchmark numbers, I would like to cover some other aspects of performance that are important to someone thinking about purchasing the iDEQ 200P.
The main issue that concerns me is that only one memory benchmark completed during the entire time I was testing this system! It was a Sandra memory bandwidth benchmark, and every other time I ran it the system would either lock up hard (requiring a push of the reset button) or simply reboot the system. The PCMark04 memory test never completed successfully and exhibited the same instability. Please keep in mind that I was using Corsair memory that has not given me any troubles in three different systems. I tried using only one module, setting less aggressive memory timings, and increasing the memory voltage to 2.9V – none of these helped the situation! I must admit that I am puzzled by this, especially since I experienced no other problems with stability.
Another concern I have with the iDEQ is Biostar’s System Control Utility. This is a utility that is supposed to help with controlling the system and CPU fans. It allows users to select from three different modes of operation. According to the included Readme file, the ‘Normal’ mode is supposed to keep the system running at full power, and with “Auto Mode” selected it will find a good balance between fan speed and temperature to keep the system stable. Apparently, the “Quiet Mode” is meant to allow the user to select a certain fan speed and the utility can reduce system performance if necessary to keep the system and CPU temperatures within safe limits. Unfortunately, the Readme did not say how it reduced performance or what the threshold temperatures are or anything helpful like that. The final mode of operation is “Fuzzy Mode.” This mode is the default and is supposed to automatically find the best combination of system performance, CPU/system temperature, and CPU/system fan speeds. This seems like a pretty clever concept, but it did not seem to be working well, if at all.
The fan speeds never seemed to change on their own, at least not enough that I could hear an audible difference. Each of the modes of operation seem to start the fans running very slow (around 2100 RPM) and nearly inaudible. When the system and CPU fans reach their maximum speeds, which are around 4100 and 5200 RPM respectively, the system sounds as loud as my loudest system. The nearly inaudible state is very nice for surfing the web and other such activities that do not stress the system. Eventually, the idle CPU temperature would rise above 50 degrees Celsius, though. That just does not seem like a good thing to me. It seemed like the System Control Utility was not quite doing the job it advertised, unless the CPU temperature has to get much higher for the fan speed to be ramped up very high?!? Another issue I have with this utility is that it seems to be a resource hog. Looking at Windows Task Manager, I noticed that at any given moment SystemControl.exe was maintaining at least 6 percent CPU usage and often it jumped to 13 percent! That’s ridiculous for a utility that will always be running. I realize that in order to work as advertised it has to poll the sensors frequently, but I seriously doubt that this should require such a CPU load. To make sure this issue was not isolated to an older version of the software, I downloaded, installed, and ran the latest version (2.198). Unfortunately, it was still an issue with this version.
OK, now that I have cleared the air, I think it’s time to show how the iDEQ 200P stacks up against the competition.
The following elements of the all test systems were the same:
- GeForce FX 5950 Ultra
- Corsair TWINX1024-3200LLPRO (2x512MB DDR-SDRAM)
- Operation System: Windows XP with Service Pack 1
- Chipset Drivers: nForce Systems – 3.13, VIA System – 4.48
- Graphics Card Driver: Forceware 53.03
- DirectX Version: 9.0b
| Athlon 64 Test Systems:
| nForce2 Comparison Test System:
I realize there are a few more variables than just the motherboard in these test systems, but I think it is still helpful to do the comparison. I did the best I could with what I have. 🙂 Although I have issues with the System Control Utility, I decided that running most of the benchmarks with it running would provide a more realistic picture of how most users would have their systems set up.
Please keep in mind that the K8NBP motherboard in the iDEQ 200P utilizes NVIDIA’s nForce3 150 chipset.
To complete these benchmarks, I used FutureMark’s PCMark04 (v 1.0.0) and 3DMark03 (v 3.2.0), SiSoftware’s Sandra 2004 (v 2004.10.9.89), AquaMark3, Unreal Tournament 2003 Demo, and Gun Metal.
PCMark04 (version 1.0.0)
I decided to use the PCMark04 results to show how much of a resource hog Biostar’s System Control Utility really is. As you can see from the results that I’ve broken out into “w/ System Utility” and “w/o System Utility,” there is about a 3-5% decrease in performance when the utility is allowed to run in the background. That may not seem to be a big deal to some people, but I think a utility that is always running should be a little more efficient than that! The “Unstable” comments on the Memory Score section are because the system would never successfully complete the memory tests – reboots or lock-ups would happen each time.
SiSoftware Sandra (version 2004.10.9.89)
* Please note that I was able to get this benchmark to run without stability issues (e.g., lock-ups or reboots) only once. Therefore, I didn’t think reporting a score would be a true representation of my experience. For those who are interested, the scores for that one successful run were: 3046 MB/s for both the Int Buffered and the Float Buffered tests.
3DMark03 (version 3.2.0)
The default benchmark in AquaMark3, which is the only one that can be run on an unlicensed version, is the only one that provides GFX and CPU scores. It is the “official” AquaMark3 benchmark. The following scores were reported for the test systems:
- Biostar iDEQ 200P
- GFX – 5,794
- CPU – 8,454
- Gigabyte GA-K8VNXP
- GFX – 5,872
- CPU – 9,202
- Leadtek WinFast K8N Pro
- GFX – 5,835
- CPU – 9,230
- Leadtek nForce2 Ultra 400
- GFX – 5,956
- CPU – 7,398
Gun Metal Benchmark 1
It seems as if system performance does not have a great affect on Gun Metal peformance.
Unreal Tournament 2003 Demo
I ran the UT2K3 benchmark with the benchmarking utility from BensCustomCases.com, and a custom script file that sets up the maps with 12 bots and maximum detail. I think this is a good test for overall system performance.
With the help of PriceGrabber, I was able to find the Biostar iDEQ 200P for $288 and up. That may be a bit more than you would pay for a full-ATX nForce3-based motherboard and a nice case and power supply; however, you would not get it all in one nice package with the cables nicely routed for your benefit.
You definitely do get a lot in a little package with the iDEQ 200P. There’s SATA RAID, 6-channel audio, gigabit ethernet, and I/O ports galore! That’s impressive and probably appealing to a lot of people. A system like this would have been great for me a few years a go in college. It would not seem like a crazy idea to bring your PC home for the weekend or take it anywhere (can you say LAN party?) with you since it’s so small. Biostar even offeres a special bag to transport iDEQ systems!
If it weren’t for the few nagging issues I experienced with this system, I would be praising it without reservation. Unfortunately, those nagging issues were big enough that they cannot be ignored. The single completion of a synthetic memory benchmark is disturbing, and I still do not know what to think about it. What is really puzzling is that I did not experience any other negative issues with stability. The system was rock solid during an hour of UT2K3 deathmatching. Another issue that bothers me is that the System Control Utility seems to act very flaky, and it uses more CPU time than such a utility should. Even the nice features in the System Control Utility have serious issues. For example, you can manually set the speeds of the CPU and system fans; however, you cannot save these settings so that they are used at each boot-up. Other than these issues, the iDEQ 200P’s performance was on par with its peers. It even trumped the other systems in some benchmarks. Without the System Utility running, I think it would have a solid chance at being in the lead in nearly all the benchmarks I ran.
Update – March 15:
After further investigation, I was able to get the Sandra Memory Bandwidth benchmark to complete consistently with only DIMM slot 2 populated. However, the PCMark04 Memory test still locked up with this configuration. Nothing I did helped with the PCMark04 memory benchmark.