ASRock X79 Extreme4-M and X79 Extreme4 Review – Sandy Bridge-E meets mATX
by Ian Cutress on December 9, 2011 12:00 PM EST- Posted in
- Motherboards
- ASRock
- X79
In The Box
As mentioned in the Extreme4-M section, if you remember the P67 Extreme4 by ASRock, for ~$160, we got a substantial haul in the box, including a front panel USB 3.0 panel and SSD holder. Though compared to the X79 Extreme4-M, we get more in the box for our extra $10:
4 x SATA Cables
IO Panel
Driver CD
3 Slot SLI Bridge
Tri-SLI Bridge (3 slot, 2 slot – to fit this board)
I am a little disappointed to be honest, given ASRock’s previous tenacity when it comes to box bundling.
Board Features
| ASRock X79 Extreme4 | |
| Size | ATX |
| CPU Interface | LGA2011 |
| CPU Support | Intel Second Generation Core i7 Sandy Bridge E |
| Chipset | Intel X79 |
| Base Clock Frequency | 100.0 MHz |
| Core Voltage | Default, 0.6 V to 1.7 V |
| CPU Clock Multiplier | Auto, 12x to 60x |
| DRAM Voltage | Auto, 1.207 V to 1.806 V |
| DRAM Command Rate | Auto, 1N to 3N |
| Memory Slots |
Four DDR3 DIMM slots supporting up to 32 GB Up to Quad Channel Support for DDR3, 800-2400 MHz |
| Expansion Slots |
2 x PCIe Gen 3 x16 1 x PCIe Gen 3 x8 2 x PCIe x1 2 x PCI |
| Onboard SATA/RAID |
2 x SATA 6 Gbps, Support for RAID 0, 1, 5, 10 4 x SATA 3 Gbps, Support for RAID 0, 1, 5, 10 3 x SATA 6 Gbps (Controller) |
| Onboard |
4 x SATA 3 Gbps (PCH) 5 x SATA 6 Gbps (2 PCH, 3 Controller) 6 x Fan Headers 1 x 4-pin Molex CFX/SLI Power Connector 1 x HDMI_SPDIF Header 1 x Front Panel Header 1 x Front Panel Audio Header 3 x USB 2.0 Headers 1 x USB 3.0 Header 1 x COM Header 1 x IEEE 1394a Header Power / Reset / Clear CMOS Buttons + Debug LED |
| Onboard LAN | Broadcom BCM57781 Gigabit LAN |
| Onboard Audio |
Realtek ALC898 7.1 Ch HD, Supports THX TruStudio |
| Power Connectors |
1 x 24-pin ATX connector 1 x 8-pin 12V connector 1 x 4-pin Molex CFX/SLI Power Connector |
| Fan Headers |
2 x CPU Fan Header 3 x Chassis Headers 1 x PWR Header 1 x SB Header (occupied) |
| IO Panel |
1 x PS/2 Mouse Port 1 x PS/2 Keyboard Port 1 x Optical S/PDIF Out Port 1 x Coaxial S/PDIF Out Port 6 x USB 2.0 2 x USB 3.0 1 x eSATA 6 Gbps 1 x Gigabit Ethernet 1 x Firewire 1 x Clear CMOS Audio Jacks |
| BIOS Version | 1.5 |
| Warranty Period | 2 Years |
ASRock are starting to use Broadcom NICs on their products. As mentioned in the intro, it is nice to see a high end Realtek Audio Codec in there (ALC898). On the flip side, with the 4-pin molex CFX/SLI power connector on board in an odd position, one has to wonder whether it is really needed when other boards do not require it.
Overclocking
At the time of testing, the latest BIOS available for the X79 Extreme4 is the 1.50 BIOS. This, according to ASRock, affords a better overclocking experience. The 1.50 was not available for the 4-M, so we may get different results here.
ASRock always like offering overclock presets, and the X79 Extreme4 is no different. We can select between 4.0 GHz and 5.2 GHz in 200 MHz increments, however your mileage may vary depending on the CPU itself, and cooling. I went straight in at the 4.8 GHz setting, not expecting any trouble. However, while the board did POST, it did not want to load any OS – the screen would hang with a blinking carat, then after 20 seconds the whole board would reset.
On the 4.6 GHz setting, it all went swimmingly – the board booted without issue. When running 3DPM (multi-threaded mode), we saw a constant 4.6 GHz in CPU-Z, although the CPU voltage was fairly alarming, running at 1.496 V at full load. This is about 0.1 volts too much!! From this, a temperature of 83 degrees Celsius was seen in 3DPM, on an open test bed with the Intel All-in-One Liquid cooler. When running a thorough CPU and memory test using Blender, the board would declock the CPU to 3.3 GHz when the CPU hit 84 degrees Celsius, and stay there until the end of any CPU load, wholly negating any overclock.
When adjusting the settings manually, I used my common X79 overclock presets – CPU at 1.4 volts, Power Limits to 500W and Core Limits to 500A. With this, I went straight in with a 46x multiplier (4.6 GHz) without issue. During 3DPM-MT, the highest temperature seen was 76 degrees Celsius. However, during the Blender test, due to the lower voltage compared to the auto settings in the previous paragraph, it took about 7 minutes to reach 84 degrees Celsius, and then the CPU backed off to 3.3 GHz. I turned off CPU Thermal Throttling in the BIOS, and reran the test. This time, at 85 degrees Celsius, the board decided to shut off completely, with no warning whatsoever. This must be an ultimate temperature failsafe for the board, however it does leave us with what to suggest with an appropriate overclock. With an ASRock it seems, it all depends on your cooling – our 4.6 GHz manual adjustment only hit the throttling after several minutes of 100% CPU, so could offer good speed in all but the most strenuous loads.
For memory overclock, we have several options (with a CPU frequency of 100 MHz), from DDR3-800 to DDR3-2400, going up in typical memory straps, as well as standard XMP. At 1866 MHz and 2133 MHz, the board gave automatic subtimings of 9-11-9-28 2T. At 2400 MHz, which did not work with the Extreme4-M, we had a completely stable system with the Extreme4, with automatic timings of 10-12-10-33 2T. This was completely Blender stable. Unfortunately, the system doesn’t offer further straps than this, so we had to bump the CPU bus frequency to see more. At the 1.25x gear ratio (125 MHz on CPU, multiplier was lowered accordingly for the same CPU overall speed), the straps offer different values, including 2000 MHz, 2333 MHz, 2666 MHz and 3000 MHz. At 2000 MHz, the memory defaulted to 11-11-11 which equates to JEDEC settings on the memory. Thus at 2333 MHz, when the board didn’t boot, I assume it was trying to implement 9-11-9-28 2T, which for this kit is a no go.
An overall overclock of 4.6 GHz and DDR3-2400 (for all 16GB) is a respectable result, limited only by the cooling and the thermal throttling of the board.
Facebook
Twitter
RSS
54 Comments
View All Comments
DanNeely - Friday, December 9, 2011 - link
A lot more of them do have reasons to be migrated than you'd think. Even 5 years ago I started seeing L shaped cards at the low end ( a tiny strip of PCB along the bracked, and a second for the PCI plug); chopping the 2nd half of the L cuts manufacturing costs by an amount that more than pays for the engineering over larger production runs.Devices that need larger PCBs but which have minimal bandwidth needs are an ever shrinking segment of the market. Even when total bandwidth isn't an issue the fact that PCIe bandwidth is dedicated instead of being shared means you no longer need to put as much hardware into buffering to avoid latency bursts when something else is using the bus more heavily.
Finally, once legacy PCI starts disappearing on a non-trivial fraction of boards total collapse from mainstream devices is inevitable. Once supporting PCIe becomes mandatory it's only a matter of time until redesigning the core chip on the card to be native PCIe instead of PCI and using a PCI-PCIe bridge chip becomes the cheaper option (probably with the next scheduled redesign). While they might initially maintain back compatability with a bridge chip going the other direction; however being doublely niche parts AGP gfx cards from 3 or 4 generations ago is probably a good comparison example. nVidia didn't make any at all, and the handful of ATI 3xxx/4xxx cards went at significant price premiums.
Eventually it'll end up like ISA; if you're willing to pay a large enough price premium (eg because the industrial/lab equiptment you're controlling costs thousands or millions of dollars to replace) there will be a handful of companies willing to sell you semi-custom boards at a large price premium and technology lag. The last time I looked I couldn't find ISA on anything newer than LGA775, until after intel finally pulls the plug on the last 775. That probably won't be for a while; if you look at their CPU database intel hasn't discontinued its embedded p4's yet, and probably won't file a while. IIRC they typically have contractual agreements to keep embedded parts in stock for a decade.
darckhart - Friday, December 9, 2011 - link
nope. still useful for pci graphics cards for troubleshooting video probs.Blibbax - Friday, December 9, 2011 - link
The differences in load power consumption might just demonstrate the margin of error on that test. Worth keeping in mind for other comparisons.The other possibilities are that the 5850s use a lot more power when they're a little bit hotter, and that the power circuitry on the M-ATX board is just awful under high load.
Concillian - Friday, December 9, 2011 - link
I was reading through the article and thought I had hit back instead of forward since I had read the page I was reading before.After finding all my marbles, I noticed that the article has two sets of Page 2 & 3. page order is 1, 2, 3, 2, 3, 4, 5...
hal74 - Friday, December 9, 2011 - link
I am always disappointed to find an article written by Ian. I know that I'll get an article written by someone who fails at plural vs singular when talking about a company and who doesn't come up with interesting comparison charts. Ian chose to throw in an E350 into the mix and didn't even add any comparisons with an x58, or any other core i7. Also, whats with fascination with older video cards in SLI?Is this seriously what people want in an article from Anandtech?
Spivonious - Friday, December 9, 2011 - link
Usually, I'll let grammar mistakes slide, but when I find multiple mistakes on a single page it really starts distracting me from the material.Can we get editors for the articles written by non-native English speakers?
Spivonious - Friday, December 9, 2011 - link
And "tenacity" is used incorrectly."...given ASRock’s previous tenacity when it comes to box bundling."
Does this make sense?
"...given ASRock's previous stubbornness when it comes to box bundling."
JonnyDough - Friday, December 9, 2011 - link
Yes. It does. Maybe your reading comprehension is lacking. The sentence means that ASRock will not budge when it comes to box bundling.Can we get some editors for the comments written by non-literate English readers?
Spivonious - Friday, December 9, 2011 - link
"3.2 GiB limit of 32-bit"*scratches head*
I thought 2^32-1 was 4GiB...
Aisalem - Friday, December 9, 2011 - link
using 32-bit you are able to address 4GB but unfortunately you will not be able to use whole 4GB in most of the 32-bit Windows installations, that also depend on the additional hardware you have.Now you shouldn't *scratches head*.