The ClearFog ARM ITX Workstation Performance Is Looking Very Good

[linux4kix]

Now THAT is sexxy. Especially those 10GbE SFP+ ports. Request: Linux and FreeBSD network performance on 10GbE.

Wanna see

sustained bandwith, packets per second TCP, UDP, mixed, sustained. And then set up some complex routing/firewall/IDS rules, and run again. Both Linux, and FreeBSD.

this might be some serious networking hardware we have. It would have been the icing on the cake if it had 2.5/5G ethernet.

They are on the list. The 2.5 / 5G ethernet was left off due to chip constraints, but supporting multi-gig is on our radar. Perhaps the new board spin in 2020 when the next revision of the SOC is released.

[linux4kix]

That boards are around ~2500 $, ...you can't compare them.
They are for a different purpose..

I agree here. The NXP documentation is quite thorough and we are already starting to document the COM module, https://developer.solid-run.com/know...7-user-manual/ Where the products differ are that we are making a general Workstation / Server software development platform, where as the Xilinx platform is for hardware prototyping. This of course needs to have far greater documentation regarding the specifics of the hardware.

[linux4kix]

A very interesting board, indeed. I've been wondering for years why no company would bring an ARM-based workstation motherboard to the market. This finally seems like a suitable and affordable candidate to build a daily driver around.

linux4kix Can you perhaps tell us a bit about your experiences in plugging a discrete graphics card into this board? One with an AMD or NVIDIA GPU? Are there drivers available to make such mainstream cards work on an ARM64 architecture yet, with 3D hardware acceleration and all? Perhaps the open source AMD GPU drivers could be made to work with some minimal patches? Thanks for taking the time to answer our questions here, by the way.

We are currently using the work that Linaro did for the Macchiatobin board, which uses qemu to initialize x86 uefi bios's on modern graphics cards. We will have probably 4-5 cards that we "certify" compatible at launch. These cards will rely on a fully OSS graphics stack, which means we are heavily favouring ATI Radeon cards over NVidia. The Radeon OSS stack is just far better supported than Nouveau, although we will test a few GTX cards as well.

Because the SOC is already low such low power, and most developers I have spoken with value silence over crazy specs I am looking at. The RX550, RX560, and RX570 Saphire Pulse cards. I am choosing the Pulse line up because they have semi-passive fan operation. They all support OSS 4K encoding and decoding. I am hoping to also verify RADEONOpenCompute on them, https://github.com/RadeonOpenCompute . There are some slight issues here as this requires PCIe Atomics to work on these cards, and there are some issues with this functionality on the PCIe IP. We are working with NXP on the status of this. The RX570 seems like overkill considering the power consumption, but if want to run quad-monitors, etc who am I to judge? Then we will probably just verify a NVidia 1030 and 1050ti card, because if Nouveau hackers want to run ARM64 we don't want to leave them out.

[linux4kix]

This is for sure a very Nice Board, and a must have( if the price is competitive )

The closest thing we saw, was the AMD Opteron A1XX processor in a board, but no testings with it..
Its Like a Ghost..
Comparing to the AMD Opteron, I am following another Chip that makes me curious about is the:
Baikal-M

It has 10 MB Cache, 8 cores aCortexA57( it should be like the AMD Opteron A1xx or so..though the amd one had higher clock frequencies, but half the core count, I think.. )
Although it has a big advantage, in my point of view..
A 8 Core MaliT628( which is nice to attach to a display.. without consuming 50W of power for graphics.. ).
No words about the Power Consumption yet, has the processor are still to ship samples in 2nd half 2019...

Well the RX550 is 50Watts max TDP. So basically it hits that only when running at 100% utilization. Toms hardware has some nice comparisons and in general it looks like it is about 6W at idle, and <20W at typical multi-monitor use and when doing accelerated video playback. To me this is quite acceptable and fits well with a 32W max TDP SOC.

I think a Workstation class desktop that is using around 50Watts is very respectable and still far more efficient than x86 counterparts.

[tuxd3v]

Well the RX550 is 50Watts max TDP. So basically it hits that only when running at 100% utilization. Toms hardware has some nice comparisons and in general it looks like it is about 6W at idle, and <20W at typical multi-monitor use and when doing accelerated video playback. To me this is quite acceptable and fits well with a 32W max TDP SOC.

I think a Workstation class desktop that is using around 50Watts is very respectable and still far more efficient than x86 counterparts.

I Agree,
That card has a low power consumption..
Its a nice target..

This board you are releasing, its a very nice Desktop/Workstation!

[jo-erlend]

What about ECC?

[linux4kix]

What about ECC?

It is supported. https://www.nxp.com/docs/en/fact-sheet/LX2160AFS.pdf

[Sonadow]

So comparing the numbers here to some ryzen 1700 on the open benchmarking size, it seems this board+SoC has 50-75% of the performance.

Given that both perf is lower and price is higher (than combined ryzen + motherboard), what is the target market of this?

Main thing I'm seeing is that power might be lower and that requires less cooling (The board is shown without fan, which might point to that direction)

As a HEDT user this is a no-go for me. Going to have to pass, since it's never going to hold up to my dual-Xeon and 2990wx workstations.

The Ultimate goal is for SBSA compliance which means any distro with a new enough kernel should boot and run on the board. Of course getting everything into mainline will take some time. We are already working with many of the major distros making sure we have everything necessary to have the board supported as soon as possible.

Ultimate goal? This means that for now it's still doing to depend distributions shipping on board-specific bits?

[linux4kix]

As a HEDT user this is a no-go for me. Going to have to pass, since it's never going to hold up to my dual-Xeon and 2990wx workstations.

Really? You expect a 16-core mid-range workstation board + COM with a TDP of 32W that in total is being sold for < $1000 to stack up to computing setups that the CPUs alone are twice the cost? The additional cost of running those units in energy a year probably is most likely comparable to half the cost of our product.

Ultimate goal? This means that for now it's still doing to depend distributions shipping on board-specific bits?

Well this will be the case regardless. SBSA compliance does not cover all specifics regarding running Linux on the device. It covers the standards for booting and basic device compatibility over PCIe. All the other drivers and power management still needs to be integrated into the mainline kernel before the distributions will ship with support.

[edwaleni]

A very interesting board, indeed. I've been wondering for years why no company would bring an ARM-based workstation motherboard to the market. This finally seems like a suitable and affordable candidate to build a daily driver around.

linux4kix Can you perhaps tell us a bit about your experiences in plugging a discrete graphics card into this board? One with an AMD or NVIDIA GPU? Are there drivers available to make such mainstream cards work on an ARM64 architecture yet, with 3D hardware acceleration and all? Perhaps the open source AMD GPU drivers could be made to work with some minimal patches? Thanks for taking the time to answer our questions here, by the way.

Cavium and Gigabyte did cooperate to create a ATX ARM workstation motherboard using the ThunderX CPU. Cavium showed off the ThunderStation at various trade shows.

I tried to order this board from one of Gigabyte's distributors and I got the third degree on what I was planning to do with it. Seems Cavium is very protective of how the ThunderX (and subsequent ThunderX2) will be used to avoid comparisons to Xeon.

I did not order it after all. After several email questions that I thought were none of their business, I dropped the idea.