I normally don't like to rant, but quite frankly I am so upset with the quality of this particular motherboard, I want to save everyone else the trouble. What motherboard, you ask? It is none other than the Maximus III Gene
, a uATX board made by Asus. But first, let me get a few things strait.
- I have not overclocked anything (CPU, GPU, memory, etc.) on this motherboard. Everything is at stock clocks.
- I have nothing against Asus in particular. I am using their monitor, and I really like it.
- I purchased the motherboard in November of last year.
- I dual boot to both XP x64 and 7 x64, but only use 7 when I need DX10/11.
So, what are the problems? Well, where to begin...
Okay, to be fair, it's not all bad, it does have attached power/reset buttons. However, that's about where the good ends. My primary complaint is the PCI(e) slot design. As you may know, the Lynnfield processors integrate 16 PCIe 2.0 lanes on-die, moving lanes normally found on the north bridge into two sets of x8 lanes.
The motherboard obviously makes use of these lanes, but there's a catch. If you are only using one graphics card, the only way to connect it via all 16 lanes is via the top slot. If the card is placed in the bottom slot, only 8 of the 16 lanes will be used. If you're using two cards, this isn't an issue as you'll only get 8 lanes to each no matter what. So, why is this a problem?
Well, all modern graphics cards are dual-slot, taking up the slot the card is placed in as well as the slot below it. If you use the top slot to get all 16 lanes to one card, this then prevents you from using the PCIe slot directly below it.
It's worth noting that if you wanted to use another PCIe card, and put it in the other "x16" slot (really an x8 ), you'd rob those 8 lanes from the top slot, regardless of how many lanes the card requires. In short, there is no way to use a dual-slot graphics card in x16 and make use of the PCIe lanes on the north bridge.
2. Memory Compatibility
Back when I bought the board, I bought 8GB of RAM that I thought would work with it. It wasn't on the memory compatibility list, but it was identical (spec-wise, e.g. voltage, freq, timings, etc.) to other RAM that was. However, it didn't work in the board--and the RAM wasn't at fault. I tested it in a friend's machine to verify that it was good. I returned it for memory that was directly off the list, which did work, but the RMA cost me, both time and money. On previous machines I've built, I've never bought memory directly off the compat list, and have never had issues.
Asus has since released numerous BIOS updates to improve memory compatibility, so this may no longer be an issue, but better safe than sorry IMHO.
This board is all solid capacitors, and has good cooling for components. But alas, the problems overshadow the positive points. First off, I had problems with the IEEE 1394 (AKA Firewire) on the board. Simply put, it would work for some time, and then die. Windows would still (usually) detect devices being plugged in, but the connection itself would not work.
I even tried connecting it directly to my friends machine to transfer some files via TCP/IP over 1394. His machine has a Texas Instruments 1394 controller, whereas this board has a VIA (ick!) controller. I like VIA's NanoITX boards, but their 1394 controller certainly isn't a strong point. After a while of transferring files, the connection would simply die--the file transfer would time out, and nothing short of rebooting the system could restore it. However, it was just my machine with the VIA controller that needed to be rebooted
--his continued to function perfectly.
This could be a driver issue, but Asus never releases updated drivers.
As you'll see, this is a problem elsewhere.
I like to use integrated audio for several reasons. However, the drivers Asus provides for the integrated VT2020 (yes, that's VIA Technologies
) are terrible (V126.96.36.19930). I would get random BSoDs, usually when playing games, and each one would point to the VIA audio drivers. VIA doesn't let you download drivers for this device on their site, and eventually I ended up just downloading the drivers for the same device on a different motherboard on Asus's site.
Uhm, hello Asus, if you're listening, why should this be so hard? You have the drivers on your site, why not list them with all compatible motherboards? Oh, but wait, it gets better...meet my friend X-Fi.
Creative's X-Fi implementation is supposedly "on the motherboard." However, from what I've read it's basically completely done in software with little or no components on the motherboard itself. X-Fi would be great if it ever worked. Okay, so it works sometimes
, but every time I tried to load a saved game in The Elder Scrolls IV: Oblivion the system would BSoD
and list Creative's X-Fi driver as the culprit. Interestingly enough, it doesn't BSoD when you start a new game, so if you could somehow play though the entire game in one sitting (Hah!) or just leave it running the the background (and hope you don't need to reboot and the system doesn't crash) until you did, and never died, you would probably have no issues. Impractical? Extremely.
After I upgraded the audio drivers (to V188.8.131.5200 from the "Maximus III Extreme" motherboard), the X-Fi control panel could no longer find the "X-Fi device" (don't remember the exact error), and X-Fi seemed to be disabled. Further proof was provided when I could actually load my Oblivion save!
However, this had a darker turn I did not immediately realize--Mumble, my choice VoIP program, doesn't work w/o the X-Fi device installed and working!
There's literally no sound out from the program, save for the Microsoft Sam TTS engine (possibly due to some mixing it was doing on the Direct Sound Voice Output). I can't hear other people talk, but they can hear me. The X-Fi does not work w/ the newer sound card drivers, and I don't want to downgrade those for reasons already discussed, or even reinstall X-Fi at this point if I want any hope of playing Oblivion (and quite probably other games). I'm also not completely sure that even the upgraded audio drivers have fixed all the problems, they just haven't caused a BSoD...yet.
If I had known what I know now, back when I bought the motherboard, I would have returned it for a refund, eaten the restocking fee, and bought something else. Better yet, never bought it in the first place. However, there wasn't much else available at the time that was LGA1156 that I could confirm would recognize 8GB of RAM. Lesson learned, but hopefully this will keep you from having to go through the same terrible experience I did.
Don't waste your time or money on this motherboard.
Update Aug 17, 2010:
First off, I am now strictly running Win7 (to avoid BSoDs related to the audio drivers, and more importantly, for development reasons), there are no sound-related BSoDs thanks to its virtualized audio subsystem. As I don't see any updates to the audio drivers on Asus's site, this will still be a problem if you are using XP. But, if you are using Win7, read on...
Several BIOS revisions have come out since the original posting. These are 1801, and 2001. 1801 greatly reduced the occurrences of BSoDs, but there were still a few remaining every now and then. After updating to 2001, I have not had a BSoD yet, and I hope to remain that way. These BIOS updates have drastically changed the system's stability under Windows 7.
It has finally, months later, made this board a worthwhile purchase, for me, at least. I only wish I could have avoided the months of prior frustration.