With mainstream boards based on Intel's P67 Express chipset now priced well beyond $150, we reached into the $200-250 range to see what kinds of enhancements high-end buyers could expect just ahead of the anticipated Z68 Express launch.
The prices we expect to pay for PCs don't change much. An entry-level machine is around $500, and has hovered around there for a while. A potent gaming system is completely doable for $1000, and that's a largely static figure, too. But our expectations of what you should get out of PCs at either of those price points is higher than ever.
An ever-increasing number of next-gen interfaces and add-on controllers continue pushing so-called mainstream motherboards well into what most enthusiasts used to consider high-end territory, shoving the cost of a real high-end board beyond $200, even for models that employ what the industry collectively considers mainstream chipsets, like Intel's P67 Express.

Today’s $200 P67-based motherboards are anything but extreme, even lacking Nvidia's NF200 controller, which enhances SLI and CrossFire graphics support. For that, one must go farther into the $250+ market (check out GeForce And Radeon On Intel's P67: PCIe Scaling Explored for more on what NF200 can do for graphics performance).
Yet, these same boards do have a few so-called extreme features, such as improved voltage regulators for increased overclocking capability. All of them also support three graphics cards (though in the least-favorable x8/x8/x4 PCIe configuration we can think of), dual gigabit Ethernet ports, and additional USB 3.0 or SATA 6Gb/s ports compared to mainstream versions.
| High-End P67 Motherboard Features | |||||
|---|---|---|---|---|---|
| ASRock Fatal1ty P67 Professional | ASRock P67 Extreme6 | Asus P8P67 Deluxe | Asus P8P67 EVO | MSI P67A-GD80 | |
| PCB Revision | 1.02 | 1.04 | 1.03 | 1.03 | 2.1 |
| Chipset | Intel P67 Express | Intel P67 Express | Intel P67 Express | Intel P67 Express | Intel P67 Express |
| Voltage Regulator | Eighteen Phases | Eighteen Phases | Sixteen Phases | Twelve Phases | Twelve Phases |
| BIOS | P1.70 (03/16/2011) | P1.60 (03/16/2011) | 1502 (03/02/2011) | 1502 (03/02/2011) | A0 (03/21/2011) |
| 100.0 MHz BCLK | 99.8 (-0.2%) | 99.8 (-0.2%) | 100.0 (+0.0%) | 100.0 (+0.0%) | 100.0 (+0.0%) |
| Clock Generator | P67 Integrated | P67 Integrated | P67 Integrated | P67 Integrated | P67 Integrated |
| Internal Interfaces | |||||
| PCIe x16 | 3 (x16/x0/x4 or x8/x8/x4) | 3 (x16/x0/x4 or x8/x8/x4) | 3 (x16/x0/x4 or x8/x8/x4) | 3 (x16/x0/x4 or x8/x8/x4) | 3 (x16/x0/x4 or x8/x8/x4) |
| PCIe x1/x4 | 2/0 | 2/0 | 2/0 | 2/0 | 2/0 |
| Legacy PCI | 2 | 2 | 2 | 2 | 2 |
| USB 2.0 | 4 (8-ports) | 4 (8-ports) | 2 (4-ports) | 3 (6-ports) | 3 (6-ports) |
| USB 3.0 | 1 (2-ports) | 1 (2-ports) | 1 (2-ports) | 1 (2-ports) | 2 (4-ports) |
| IEEE-1394 | 1 | 1 | 1 | 1 | 1 |
| Serial Port | 1 | 1 | None | None | None |
| Parallel Port | None | None | None | None | None |
| Floppy | Yes | Yes | No | No | No |
| Ultra-ATA 133 | 1 (2-drives) | None | None | None | None |
| SATA 3Gb/s | 4 | 4 | 4 | 4 | 4 |
| SATA 6Gb/s | 6 | 6 | 4 | 4 | 2 |
| 4-Pin Fan | 2 | 2 | 2 | 2 | 1 |
| 3-Pin Fan | 4 | 4 | 3 | 2 | 4 |
| FP-Audio | Yes | Yes | Yes | Yes | Yes |
| CD-Audio | Yes | No | No | No | No |
| S/PDIF I/O | Output Only | Output Only | Output Only | Output Only | Output Only |
| Power Button | Yes | Yes | Yes | Yes | Yes |
| Reset Button | Yes | Yes | Yes | Yes | Yes |
| CLR_CMOS Button | Jumper-Only | Jumper-Only | Jumper-Only | Jumper-Only | Jumper-Only |
| Diagnostics Panel | Numeric | Numeric | Numeric | Pass/Fail LEDs | Pass/Fail LEDs |
| I/O Panel Connectors | |||||
| P/S 2 | 2 | 2 | 1 | 2 | 1 |
| USB 2.0 | 4 | 4 | 8 | 6 | 4 |
| USB 3.0 | 4 | 4 | 2 | 2 | 6 |
| IEEE-1394 | 1 | 1 | 1 | 1 | 1 |
| Network | Dual | Dual | Dual | Dual | Dual |
| eSATA | 1 | 1 | 2 total, 1 powered | 2 total, 1 powered | 2 |
| CLR_CMOS Button | Yes | Yes | Yes | Yes | Yes |
| Digital Audio Out | Optical + Coaxial | Optical + Coaxial | Optical + Coaxial | Optical + Coaxial | Optical + Coaxial |
| Digital Audio In | None | None | None | None | None |
| Analog Audio | 6 | 6 | 6 | 6 | 6 |
| Other Devices | None | None | Bluetooth Transceiver | Bluetooth Transceiver | None |
| Mass Storage Controllers | |||||
| Chipset SATA | 2 x SATA 6Gb/s 4 x SATA 3Gb/s | 2 x SATA 6Gb/s 4 x SATA 3Gb/s | 2 x SATA 6Gb/s 4 x SATA 3Gb/s | 2 x SATA 6Gb/s 4 x SATA 3Gb/s | 2 x SATA 6Gb/s 4 x SATA 3Gb/s |
| Chipset RAID Modes | 0, 1, 5, 10 | 0, 1, 5, 10 | 0, 1, 5, 10 | 0, 1, 5, 10 | 0, 1, 5, 10 |
| Add-In SATA | 2 x Marvell 9120 PCIe 4 x SATA 6Gb/s 1 x eSATA (shared) | 2 x Marvell 9120 PCIe 4 x SATA 6Gb/s 1 x eSATA (shared) | Marvell 9128 PCIe 2 x SATA 6Gb/s JMB 362 PCIe 2 x eSATA 3Gb/s | Marvell 9120 PCIe 2 x SATA 6Gb/s JMB 362 PCIe 2 x eSATA 3Gb/s | Marvell 9128 PCIe 2 x eSATA 6Gb/s Supports RAID 0, 1 |
| Add-In Ultra ATA | VIA VT6330 PCIe | None | None | None | None |
| USB 3.0 | 3 x Etron EJ168A PCIe | 3 x Etron EJ168A PCIe | 2 x NEC D720200F1 | 2 x NEC D720200F1 | 2 x NEC D720200F1 2 x VLI VL810 Hub |
| IEEE-1394 | VIA VT6330 PCIe 2 x 400 Mb/s | VT6315N PCIe 2 x 400 Mb/s | VT6315N PCIe 2 x 400 Mb/s | VT6308P PCI 2 x 400 Mb/s | VT6308P PCI 2 x 400 Mb/s |
| Gigabit Ethernet | |||||
| Primary LAN | RTL8111E PCIe | RTL8111E PCIe | WG82579V PHY | WG82579V PHY | RTL8111E PCIe |
| Secondary LAN | RTL8111E PCIe | RTL8111E PCIe | RTL8111E PCIe | RTL8110SC PCI | RTL8111E PCIe |
| Audio | |||||
| HD Audio Codec | ALC892 | ALC892 | ALC889 | ALC892 | ALC892 |
| DDL/DTS Connect | None | None | None | None | None |
Nice article would have been nice to have a Gigabyte board in there as well.
So glad I grabbed my P8P67 Deluxe!
It had all the features I was looking for at a low enough price to make it very appealing.
Nice article would have been nice to have a Gigabyte board in there as well.
Please tell Gigabyte to produce something for this market!
Tom's Hardware included the UD4 in its $150-200 motherboard roundup, and the UD5 costs more than $250.
How te hell is it possible that a website like this keeps ignoring my question as in WHY its been for so many months if not year or whatever, that they dont fix this *** and I cant click to submit my comment from IE? How can THIS be the only website with issues with IE? I find it rather sad. Its why i quit commenting instead of being forced to open a different browser for this site which I used to love and respect. Till they started ignoring this issue Ive been pointing out ( and not just me ).
Thank you for the charts tho id love to see one with Maximus IV included x.x ( I edited cause I asked something stupid as in why I dont see it in the chart. Sorry.. Been working all night and no time to read the article. Bits only.And no I obviously hadnt read the title x.x My bad. Happy Easter!
It would have been nice to see if there is a performance gain in these "high end" boards over a value P67 board.
For an even better article also throw in one of each value rated H67 and H61 boards. ($240 vs $130 vs $70 boards)
It would have been nice to see if there is a performance gain in these "high end" boards over a value P67 board. For an even better article also throw in one of each value rated H67 and H61 boards. ($240 vs $130 vs $70 boards)
this, i would love to see how the high end stacks up with the low end. the low may not have as much as the high end, but performance is really all that matters considering we can just get expansion cards for things we dont have.
Nice analysis glad to know ASUS is good to go.
Great article. Now it´s time for the high-end overclocking oriented boards, like the ASUS Maximus IV Extreme.
I'm gonna go all out on this chipset when it matures a bit. A cool 5 grand i'm thinking for my next gen build.
GA-P67A-UD7 doesn't count?
Thanks for the wake-up call on MSI; I had thought to maybe risk buying one of their boards again. Sounds like I should be happy to stick with ASRock.
Edit: Oh, and those shots of the ASRock boards show three pairs of two SATA cables, not just three individual cables.
Looks like the mid-range mobos do better in some of the gaming tests. That's surprising to me, since the test setups are identical.
Please tell Gigabyte to produce something for this market!
GA-P67A-UD5-B3 http://www.gigabyte.us/products/pr [...] id=3762#ov
hope if u can make a review of Asus Saertooth P67
How te hell is it possible that a website like this keeps ignoring my question as in WHY its been for so many months if not year or whatever, that they dont fix this *** and I cant click to submit my comment from IE? How can THIS be the only website with issues with IE? I find it rather sad. Its why i quit commenting instead of being forced to open a different browser for this site which I used to love and respect. Till they started ignoring this issue Ive been pointing out ( and not just me ).Thank you for the charts tho id love to see one with Maximus IV included x.x ( I edited cause I asked something stupid as in why I dont see it in the chart. Sorry.. Been working all night and no time to read the article. Bits only.And no I obviously hadnt read the title x.x My bad. Happy Easter!
I am using IE and Firefox at home and both allow me to comment on articles.
I would like to see the 'thumbs up' and 'thumbs down' buttons function appropriately, though.
I too am waiting for the return of the thumbs. It's been months; come on guys, it USED to work just fine, so please roll it back. In the forums, it says I've already voted, and here it is possible to vote, but not see the results.
Hi everyone
it's nice to be a part of Tom's Hardware forum community
I'm constant Tom's Hardware reader and I just love this site, especially for professional and reliable reviews.
My Q6600 based PC just died and I'm building Sandy Bridge machine. I'm having a tough time understanding new architecture.
I can't get the idea of second paragraph at "ASRock UEFI" page:
and it made me somewhat lost. I've read a lot about Sandy Bridge and thought that I know how things are, but now I'm not so sure, so please clarify some things for me.
Below statements are to be confirmed:
In general
1) Stock 2500K run at 3300MHz (2600K at 3400MHz) when all 4 cores are active.
1a) With SpeedStep is enabled when there is no load multiplier drops to x16.
1b) With Turbo Boost enabled, when 4/3/2/1 cores are stressed, they run at +1/+2/+3/+4 bins respectively (so multiplier increases by given value).
2) When we are overclocking Sandy Bridge we set maximal Turbo Boost multiplier.
2a) With both Turbo Boost and SpeedStep disabled CPU run constantly at set multiplier.
2b) With both Turbo Boost and SpeedStep enabled CPU run at x16 multiplier when idle and at set multiplier when at load (it doesn't matter how many cores are stressed, the multiplier is fixed at set value).
Are above statements correct? Do they also apply to ASRock?
I would also like to ask what quoted paragraph means, because I get an impression from it, that at stock with Turbo Boost enabled CPU run constantly at +4 bins and I don't get the point of "does just the opposite" part, because it's contrasting two things which are the same (maxiing multiplier all the time). This may be just a problem of my poor understanding, because I'm not native English speaker, so maybe just saying it in other words would help.
Thank you very much for all help
P.S. At first page we got information that CLR_CMOS Button in ASRock's is Jumper-Only. Why back pannel CLR CMOS doesn't qualify?
1) If you ignore Turbo Boost, this is correct.
1a) Correct.
1b) Correct.
2) Depending on the board. For ASRock, this is correct.
2a) You also have to disable C1E and all of the CStates settings.
2b) Depends on the board. For ASRock, this is correct.
After reading this review I will stay away from future MSI motherboards. What a sham.