News ASRock Unveils the X299E-ITX/ac: Mini ITX + X299 + Quad-channel Memory

Mod edit:

Detailed overview of what we know about the X299E-ITX/ac thus far here: https://smallformfactor.net/news/asrock-x299e-itxac-little-monster-detailed

Original:

ASRock did it! Finally, there's an Intel HEDT platform motherboard with full quad-channel DDR4 memory. The new X299E-ITX/ac is for those who need up to 18 CPU cores and up to 64 GB of quad-channel DDR4 memory in their SFF machines for reasons. The board manages its limited PCB real-estate by going vertical. It features two riser cards, one with a few onboard controllers, and a pair of 32 Gb/s M.2 slots), and the other riser with SATA 6 Gb/s ports, a third M.2 slot, and the headers such as USB 3.1. The board draws power from 24-pin ATX and 8-pin EPS connectors, conditioning it for the LGA2066 CPU using a 7-phase VRM. The lone expansion slot is a PCI-Express 3.0 x16, memory is handled by four DDR4 SO-DIMM slots. Connectivity includes two Intel I219-V driven gigabit Ethernet interfaces, 802.11ac WLAN, and Bluetooth 4.1.

Source

 

[QuantumBraced]

James, thanks very much for all your work/interaction with ASRock on this, and keeping us updated.

I'll say again, I'd love to see a rear panel power button, I think it would be easier to implement than moving all the headers to the motherboard, maybe they can fit it above the clear CMOS button or even replace the clear CMOS button with it. Solves your problem elegantly and pushes the board into an even more SFF direction.

 

[jeshikat]

thanks very much for all your work/interaction with ASRock on this, and keeping us updated.

No problem! This is the most excited I've ever been for a motherboard, and I really want to see it succeed so ASRock continues to do crazy stuff like this.

Because while I have misgivings about Intel's handling of X299 (especially product segmentation), I'll begrudgingly switch to it because comparatively the AM4 ITX boards just don't hold a candle to the X299E-ITX/ac.

 

[Curiosity]

Same here.
The only thing really giving me serious pause for the X299 stuff is the TDP's, the i5 is 112W!
(Cost isn't encouraging me either, but isn't as big of an issue overall)

 

[jeshikat]

While Ryzen does well under more normal tasks, under Prime95 the 1800X is only drawing 6W less than the 6900K: http://www.tomshardware.co.uk/amd-ryzen-7-1700-cpu-review,review-33854-8.html

So you could say that up to moderate loads the 45W difference in rated TDP is somewhat true. But the 1800X is a 8-core CPU with clock speeds and process node comparable to Intel's HEDT and the power consumption at max load reflects that.

 

[Phryq]

I'll be giddy if you can run the board without either daughterboard with two M.2 on the back and the space for a really nice heatsink.

So with 18mm ram and no daughter boards, and external power brick, you could build a 0.5L case around this.

 

[Dyson Poindexter]

18 cores in an M350 ought to win some sort of prize.

 

[jeshikat]

So with 18mm ram and no daughter boards, and external power brick, you could build a 0.5L case around this.

Well, there's also the rear IO, standoffs, case sheet thickness, cooling, GPU, and so forth that take up space.

 

[QuantumBraced]

So with 18mm ram and no daughter boards, and external power brick, you could build a 0.5L case around this.

Right, if you can find a 6mm heatsink to cool a 165W CPU with

 

[jeshikat]

Just keep the case submerged in liquid nitrogen, problem solved.

 

[|||]

External water chillers don't count againt the volume, just like power bricks, right?

 

[QuantumBraced]

So I have a Noctua C14 (the original one, not a C14S), which is incidentally cooling my 6800K inside a X99E-ITX/ac inside an NCase M1. I guess it's too early for speculation, and I don't plan on upgrading, but I'm wondering if that cooler will fit on this board without its bottom fan. The only way it fits in the NCase M1 is with the heatpipes facing toward the rear, and I can't figure out if they will collide with the rear riser card... I need to know the height of the pipes at the point where the riser card is and I really can't eyeball it. It's the best air cooler for the NCase M1 and will probably be the air cooler of choice for people intent on cooling 140W or 165W CPUs on air in that case (given, most will probably use water). The card can always be removed but then goodbye USB 3.1 (and possibly other things). Maybe more appropriate to ask in the NCase M1 thread in [H]...

 

[jeshikat]

I think the L12 heatpipes will just barely clear the M.2 board and comparing pictures it looks as though the heatpipes on the C14 start bending closer to the block.

So more than likely the NH-C14 will be just fine even with the bottom fan installed.

And the M.2 board has a USB 3.0 controller, the USB 3.1 controller is on the backside of the motherboard.

 

[QuantumBraced]

Okay, thanks that's reassuring.

 

[jeshikat]

Was watching the KitGuru video on this board and learned that GSkill has 3466MHz SODIMM in the works: https://gskill.com/en/press/view/g-skill-demos-super-speed-ddr4-at-computex-2017-with-new-intel®-core™-x-series-processors

 

[Mackan]

There is still potentially time to change the front panel header situation but it's unlikely since there is just such limited space available for anything left on the motherboard. But they're checking into it.

Would be a mistake to keep it on the daughterboard, imo. If i were to buy it, I would remove the both daughterboards and enjoy a much cleaner board without drawback (in my case).

 

[jeshikat]

In their defense "normal" people aren't going to remove the board so I can understand why they did it.

But hopefully I've conveyed that our community isn't normal and removing all the things to reduce size is perfectly acceptable here

 

[QuantumBraced]

I think they should consider, in moving the headers to the board, to only include the power button and power LED headers and just dump the rest. It'll be a bold, but respectable move. The power button itself is mostly useless nowadays IMO, let alone the reset and HDD LED, I haven't used those since the 1990s. I think nearly all ITX cases should start coming with rear power buttons, to give users more versatility and also to enable ultra SFF cases with no front/separate power buttons. /mini rant

 

[Kmpkt]

My suggestion would be to put the power header (9 pins) where they currently have the USB 2.0 header (9 pins). They could either move the USB 2.0 up onto the daughterboard (which would make infinitely more sense) or alternatively do away with it entirely because let's be honest, who's going to miss 2x USB 2.0 since there's already front panel support for 2x USB 3.0.

 

[jeshikat]

Bad news for anyone wanting to do without the M.2 board, it runs all four USB 3.0 ports and both of the NICs on the rear IO. So it'll be pretty difficult to live without it

I think they should consider, in moving the headers to the board, to only include the power button and power LED headers and just dump the rest.

I agree, many of the indie SFF cases don't have a reset button or HDD LED anyway. Even mass production cases are moving away from the reset button too.

My suggestion would be to put the power header (9 pins) where they currently have the USB 2.0 header (9 pins). They could either move the USB 2.0 up onto the daughterboard (which would make infinitely more sense) or alternatively do away with it entirely because let's be honest, who's going to miss 2x USB 2.0 since there's already front panel support for 2x USB 3.0.

Nah, we gotta move the front panel header but keep the USB 2.0 on the board so we can remove the SATA board and still use an Iris 16 button

 

[Dyson Poindexter]

While wildly impractical, a collection of various "riser cards" (this term is getting overused) with different I/O combinations would be neat. Some of us have no need for any SATA ports and would rather have two more M.2, while others would like 12 SATA ports for a compact storage server. I think two SATA ports and some high-power PWM fan/pump headers would be neat.