The ASRock X99 Extreme11 Review: Eighteen SATA Ports with Haswell-E
by Ian Cutress on March 11, 2015 8:00 AM EST- Posted in
- Motherboards
- Storage
- ASRock
- X99
- LGA2011-3
ASRock X99 Extreme11 In The Box
With money-is-no-object type motherboards, the package has to consider the market. Do prosumers want 18 SATA cables, or are they using a system with a backplane that comes with it all? Are buyers going to want to game, or are they Xeon Phi users for compute and don't need SLI bridges? One could argue that given the cost of the package, it should all be bundled anyway to encompass all users, but event at this price bracket if the manufacturer can save a few cents, they might do so.
In the ASRock X99 Extreme11 box we get the following:
Driver DVD
User Manuals
Rear IO Shield
HDD Saver Cable
Six SATA Cables
Two Rigid 2-way dual-slot SLI connectors
One Rigid 2-way quad-slot SLI connector
One Rigid 4-way SLI connector
Two M.2 Screws
A Carry Bag
This is pretty much what I would have expected from a motherboard like this. As it does not fall under the gaming or overclocking lines, there are no gaming or OC add-ons: just cables and SLI bridges. It might have been interesting to have included a drive bay for the two onboard USB 3.0 headers, especially one that might fit a boot drive also. Perhaps because in 2015 a lot of cases come with at least one USB 3.1 header now, ASRock sees little need.
Many thanks to...
We must thank the following companies for kindly providing hardware for our test bed:
Thank you to OCZ for providing us with PSUs and SSDs.
Thank you to G.Skill for providing us with memory.
Thank you to Corsair for providing us with an AX1200i PSU.
Thank you to MSI for providing us with the NVIDIA GTX 770 Lightning GPUs.
Thank you to Rosewill for providing us with PSUs and RK-9100 keyboards.
Thank you to ASRock for providing us with some IO testing kit.
Thank you to Cooler Master for providing us with Nepton 140XL CLCs.
Test Setup
| Test Setup | |
| Processor | Intel Core i7-5960X ES 8 Cores, 16 Threads, 3.0 GHz (3.5 GHz Turbo) |
| Motherboards | ASRock X99 Extreme11 |
| Cooling | Cooler Master Nepton 140XL |
| Power Supply | OCZ 1250W Gold ZX Series Corsair AX1200i Platinum PSU |
| Memory | Corsair DDR4-2133 C15 4x8 GB 1.2V G.Skill Ripjaws 4 DDR4-2133 C15 4x8 GB 1.2V |
| Memory Settings | JEDEC @ 2133 |
| Video Cards | MSI GTX 770 Lightning 2GB (1150/1202 Boost) |
| Video Drivers | NVIDIA Drivers 332.21 |
| Optical Drive | LG GH22NS50 |
| Case | Open Test Bed |
| Operating System | Windows 7 64-bit SP1 |
ASRock X99 Extreme11 Overclocking
Experience with ASRock X99 Extreme11
While the positioning of the Extreme11 indicates it is a board more so for functionality rather than overclocking, it does offer ASRock’s base array of overclocking options in both the BIOS and software. This includes the Optimized CPU OC Configuration drop downs in both the BIOS and software, although it should be noted that Xeons cannot overclock via the multiplier.
With our mediocre CPU, the automatic overclocks at 4.4 GHz and beyond caused BSODs when under AVX load, but manual overclocking did give 4.4 GHz at a rather high voltage. There is not much to conclude, due to our processor not being the best, but other motherboards have achieved around the same result with the CPU we have.
Methodology
Our standard overclocking methodology is as follows. We select the automatic overclock options and test for stability with PovRay and OCCT to simulate high-end workloads. These stability tests aim to catch any immediate causes for memory or CPU errors.
For manual overclocks, based on the information gathered from previous testing, starts off at a nominal voltage and CPU multiplier, and the multiplier is increased until the stability tests are failed. The CPU voltage is increased gradually until the stability tests are passed, and the process repeated until the motherboard reduces the multiplier automatically (due to safety protocol) or the CPU temperature reaches a stupidly high level (100ºC+). Our test bed is not in a case, which should push overclocks higher with fresher (cooler) air.
Overclock Results
Power delta between the stock and highest overclocked performance gives +152W, and when allowing for the 140W TDP gives an estimated total power consumption at 292W when overclocked to 4.4 GHz.
Facebook
Twitter
RSS
58 Comments
View All Comments
wyewye - Friday, March 13, 2015 - link
"this system is just a confused jumble of parts slapped together"This is the best conclusion for this mobo.
I think they hope marketing/sales guys will be able to bamboozle dummies to sell this as a 18 port raid server mobo. Anyone who spends 600$ on a high-end mobo without reading a review, deserves whats coming to them.
swaaye - Wednesday, March 11, 2015 - link
That chipset fan is cheesy. They most definitely could have come up with a better cooling solution.Hairs_ - Wednesday, March 11, 2015 - link
As pointed out above, this board doesn't answer a single question any user is asking, and it doesn't fulfill any logical useage case.I'm struggling to see why it was reviewed other than the possible reason "reviewer only wants to review weird expensive stuff". Getting in a board whose supposed only reason to exist is the number of storage ports, then not yet the storage, and say " in sure it's the same as the one I reviewed a few years ago " is... Troubling. What was the point of getting it in for review at all??
ClockHound - Wednesday, March 11, 2015 - link
What's the point?It's the new Anandtech, where the point is clicks! Catchy headlines with dubious content, it's how Purch is improving a once great review site. Thanks, Purch!
ap90033 - Friday, March 13, 2015 - link
I think you may be on to something! Sad to see..Stylex - Wednesday, March 11, 2015 - link
I don't understand how motherboards still have usb2 ports. Did it seriously take this long for it to transition from usb1.1 to usb2?DanNeely - Wednesday, March 11, 2015 - link
At the USB1.1-2.0 transition time, Intel chipsets had at most 6 ports; and since the new standard didn't need any more IO pins so they could cut over all at once. Not needing any more IO pins is important because it's been the the limiting factor for chipset cost for a number of years; with the die size being determined by the number of output pins added.The bottleneck for USB3 has been the chipsets. Pre-IVB they had no USB3. IVB added support for 4 ports, haswell bumped it to 6, the 9x series chipsets that were supported to launch with broadwell were essentially unchanged from the previous model. As a result, mobo makers who wanted to add more USB3 have had to spend extra money on 3rd party chips to do so. Initially it was on USB3 controllers which generally ate a PCIe lane for every pair of ports added. More recent designs are using 4 port USB hub chips; which give better bang for the buck but still drive prices up.
When skylake launches later this year, the situation should improve; its higher end versions will offer up to 10 ports. That might be enough ports to make all USB3 configurations possible in the mid range without either using a very small total number of ports or driving the board price up with extra controller chips. High end boards will probably still have some ports attached to a controller though, because Intel's expanding it's use of flexible IO ports and native USB3 will be competing with PCIe storage for IO pins. In both cases though, I suspect a number of boards will also expose the 4 remaining 2.0 ports; probably 1 internal header and 2 external ports. That won't be just a case of 'gotta use them all'; older OSes (eg win7) without native support for USB3 are easier to install if you've got a few 2.0 ports available; and there will be residual demand for 2.0 headers from people with older cases or internal card readers.
The situation is similar with AMD chipsets. But due to their being only able to compete in the value segment of the market, they're behind Intel; topping out at 4 native USB3 ports.
Stylex - Friday, March 13, 2015 - link
Ah, the pin count makes a lot of sense, thanks for that insight!But still, how much more could it possibly drive up the price to use a third party controller, $5-10? I'd pay that for all usb3, especially on a board like this one.
DanNeely - Sunday, March 15, 2015 - link
Estimates I've seen over the last few years put a 2 port USB3-PCIe controller as adding $10 to the retail price of a board; a 4 port USB3 hub chip added $5. The caveats are that hubs only add ports not total bandwidth; which is fine if you're only interested in being able to plug a USB3 device into any port but use sufficiently few of them that sticking multiple high speed devices on the same hub isn't a problem. Controllers don't have that problem; but do need PCIe lanes. Those tend to be in short supply on intels 8x/9x chipsets. Using a 4-8 lane PLX on the chipset relieves that pressure somewhat but is another $10 or $20 to the board price. The situation there will be better for skylake due to the 100 series chipets having 28 high speed IO lanes instead of only 18; but that's partially counter balanced by m2/SataExpress connections needing several lanes each.The lack of native 3.1 support means that the next generation of mobos will probably go the controller route; not hubs to bump up the port count. With Intel rarely doing any major updates on the Tock versions of the chipset, it will probably be at least 2017 before external USB3 controllers mostly go way.
darkfalz - Thursday, March 12, 2015 - link
You have a keyboard, mouse or gamepad that requires 100 MB/sec bandwidth, do you?