Hacker News new | past | comments | ask | show | jobs | submit login
64-bit ARM desktop hardware? (juszkiewicz.com.pl)
243 points by ashitlerferad on July 25, 2016 | hide | past | web | favorite | 169 comments

This is something I don't understand with any cpu manufacturer which isn't x86. Why don't the manufacturers offer standard pc motherboards tweaked for their corresponding cpu? Since the mid-90ies, they would have made nice Linux platforms. I think, many Linux enthusiasts would be thrilled to run an alternative CPU platform. By offering cheap desktop solutions, they would drive the software base and interest for the more expensive enterprise hardware. Back then it was Sparc - it was still competitive in performance with Intel, later PowerPc, now Arm64.

+1 to that. I feel his pain, having been there myself.

Back when DEC Alpha CPUs were still a going concern, I was one of the maintainers of the FreeBSD/alpha port. While DEC shipped their workstations, they were beyond the reach of OSS developers. The first "normal" Alpha board was the API UP1000. It was basically an Alpha 21264 on an AMD motherboard & it was totally awesome. API shipped me one before they were generally available so I could get FreeBSD working on it. I used it as my desktop (running FreeBSD/alpha) for years. Sadly all things Alpha were killed off in favor of Itanium.

Here is the best description I can find of it in ~30 seconds of googling: http://www.thefreelibrary.com/API+Announces+Program+to+Encou...

I fondly remember getting an AXP-PCI 33 board as a teenager, which were (at some point) really affordable [1]. It was my first non-x86 machine.

[1] http://manx.classiccmp.org/collections/mds-199909/cd1/alpha/...

"API shipped me one before they were generally available so I could get FreeBSD working on it. "

I didn't know about that. That's pretty neat. One of Alpha's main advantages was PALcode. It was close to microcode's advantages but with only assembly knowledge required. Did you or other FreeBSD developers ever do anything cool with it? And what?

Typing in the domain of that company today shows how sad the IT market works when incredible products are involved that aren't market leaders:


There were loads of DEC multia workstations selling cheap at some point. But even those where a pain to get booting into linux of FreeBSD.

I think the cold truth is that if they did, the non-x86 system would cost almost as much as an x86 system, and would perform worse. When you add more ports, connectors, RAM sockets, CPU socket etc., the cost of the CPU starts to becomes relatively unimportant. A cheap x86 Celeron, which is as fast, if not faster, than an ARM CPU, will only add a small percentage to your budget. And even a decent i3 won't cost that much more either, percentage-wise.

This why we only see Raspberry Pi-esque devices, I believe. ARM CPUs excel in the embedded space, which is not surprising since this is where they grew up. Their qualities are low cost and low absolute power usage (as opposed to performance per watt). This doesn't make sense for desktop systems.

There were times, when such a system might have outperformed x86 hardware (Sparc, Power). And even if not strictly competetive, people who care about writing low level software should be delighted working on something different than x86. I certainly would like to work on an Arm64 system, ideally with 8+ cores. That should be quite interesting.

May well be. I can't shake the feel that all performance we have gotten have been from smaller chip features allowing for more more heat headroom, and thus more "waste" via out-of-order, all branches at once, massive caches, etc etc etc. And once we hit the brick wall that is electron leakage, RISC or CISC will mean crap all for performance pr watt.

Never mind that in recent years both AMD and Intel has taken their offerings into the real of SoCs. AMD have largely lead the way, by moving more and more of what used to be the northbridge onto the CPU die.

The only thing keeping both out of the mobile world seems to be price pr unit. Intel bled money to get Asian OEMs to use their mobile ATOMs rather than Chinese ARM based SoCs.

The only thing really missing for many use cases is a better storage interface. The rest is gravy.

There are several single-board computers with various permutations of USB 3.0, GbE, SATA, mSATA, miniPCIe, etc.

Check out the ODroid series, Banana Pi Pro, ECS Liva, Jetson TK-1, etc.

While I have to acknowledge the RPi's pioneering contributions to the ultra-cheap SBC market, I find it immensely frustrating that it sucks up all of the oxygen on this topic. There are many, many other options at this point and many of them are better choices for the typical "mini Linux server" or "low-power desktop system" use-case.

> Check out the ODroid series, Banana Pi Pro, ECS Liva, Jetson TK-1, etc.

For which of these do regular updates (in particular kernel updates) from the mainline distribution exist?

Depends on your use case. I believe the Banana Pi Pro is supported by Debian Jessie with the stock kernel if you can do without video and audio out and a handful of other features you're unlikely to need in a headless server. See http://linux-sunxi.org/Linux_mainlining_effort - Banana Pi /Pro is A20, Jessie ships Linux 3.16. (Audio and unaccelerated graphics are supported by newer mainline kernels, in theory at least.)

The ODROID-C2 (their AArch64 offering) was added to the Linux kernel device tree in March: https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux....

But Hardkernel does seem to have a lot of additional patches in the kernel that they ship: https://github.com/hardkernel/linux/tree/odroidc2-3.14.y

There was a motherboard for the StrongARM SA110 CPU, I still have one. Fits in a standard PC case, takes PCI expansion cards, connectors for IDE etc. in the standard places.

Sun made a Sparc motherboard too but I don't know how easy it was to buy one.

Sun's "SME" division sold the Ultra AXi and AXMP boards, which were 100% standard ATX boards with PCI slots.

The Ultra 5 and Ultra 10 were basically PCs with SPARC chips.

31 Oct 2015 Titanium AM5728 Mini-DTX

Four 3Gbps SATA connections

Two PCI Express root complexes


Huh, that actually looks standard.

Approx $590.

I can't get a clear idea if this is 64-bit or not though.

Also, with this board, I don't see anywhere obvious to bolt a heatsink - and at 1.1GHz, I'd be be reluctant to do overnight compiles on the thing.

It's not necessarily cheap to provide and support a full desktop system, especially given the very small minority of people who might actually be interested in it.

> Why don't the manufacturers offer standard pc motherboards tweaked for their corresponding cpu?

Because there's no standard interface. Each SoC has its own set of pins, which changes every generation for every manufacturer. Each board would essentially have to be redesigned for each CPU, meaning that the idea of having a separate "motherboard" would be completely pointless, as it'd be totally coupled to the CPU. You could have motherboards with pre-soldered, rather than socketed, SoCs, but then again, manufacturers would essentially have to put as much work into each revision of those as they would into a true PC motherboard that has a much broader audience.

The economics just don't work out.

My old StrongARM motherboard has the CPU soldered in place.

Most current SoCs don't have a PCI bus so there isn't very much you would be able to add on. A manufacturer could provide one or two SODIMM sockets instead of soldering DRAM onto the board though.

> Why don't the manufacturers offer standard pc motherboards tweaked for their corresponding cpu?

No expertise and x86 is baked deeply into PC's.

1) No Expertise

ARM manufacturers tend to synthesize logic a lot. They don't tend to pull out circuit simulators and design things by hand. As such, the advanced techniques behind fast memory, CPU and graphics interfaces simply aren't there.

2) x86 is baked into PC's

People forget that there is a lot of "initialization"-type code on things that plug into buses. ALL of that code is x86. The Alpha PC's had to include emulators for x86 at boot time in order to init the old PCI cards.

Chicken and egg. It costs more than you think, for probably fewer geeks than you think.

ARM seems really well positioned, if there has ever been something to square off with Intel, it's ARM, you can get cheap SoCs and play with them. The software ecosystem is growing so maybe someone will throw down and be willing to put together a full ATX system, I suspect it will cost more and perform less than an Intel based product though.

RISC-V might have this edge if a manufacturer picks it up. Maybe the SiFive folks are listening.

I'd definitely be interested in running a system like that, if it had PCIe ports to run a supported GPU, perhaps a fancy NIC or a RAID card.

If they would make a PC-format RISC-V board with standard attachments, I would buy several in an instant. Though I know it's not free to license all of those bits, this being a major expense on x86 boards.

Quite simple.... there is no market for them.

There is a commodity PC market. There isn't a commodity ARM motherboard market as the type of devices ARMs go into aren't PC's, they're devices.

Yes... there is a hobbyist market. But that isn't worth investing in.

> Yes... there is a hobbyist market. But that isn't worth investing in.

The home computer market was originally a hobbyist market.

>I think, many Linux enthusiasts would be thrilled to run an alternative CPU platform.

forget Linux enthusiasts. Whole countries, like Russia or China, have national security interest in it - beside everything else, Intel chips has VNC server built-in on the silicon :). China has been making great progress in building its national computing platform, though not sure whether it has reached the desktops of their government employees. Wrt. Russia - it again lifted the ban on buying foreign hardware for security services, time to upgrade has come again i guess :) I mean, even such guaranteed markets and money available is still not enough to even come close to Intel.

Russia mostly uses SPARC: an ISA used from embedded to high-end NUMA's that was first to open up. Their mainframe CPU's are called Elbrus. Fujitsu has their own that they work with Oracle on. There's also a MIPS company. They also cloned and improved Itanium but sold that to Intel. Reason they avoid ARM, like I tell everyone else to, is that ARM charges huge licensing fees (up to $10mil I hear) with royalties. MIPS charges around $700-900k. SPARC requires a $99 registration fee. See why Japs and Russians use SPARC? ;)

The Chinese first, powerful CPU was an Alpha clone called Shinwei. Their next one was MIPS line called Loongson (or Godson). It supports x86 emulation at hardware level. So, that MIPS CPU is their main strategy for getting off Intel. I don't know if Russians have a strategy for that. But that's what the two are doing in terms of CPU's.

EDIT: China also built its own state-of-the-art fabs with a 28nm coming online recently. The Russians have fabs but not that many or that good that I know. That's two areas they should worry about.

Russia seems to be not very interested in ARM.


VNC on silicon? Do you have sources for that?

It has different names. AMT is one of the earliest implementations.

There is a separate chip with its own OS, connecting to the video chip and the nic. No driver needed. Can not be permanently removed.

http://www.meshcommander.com/meshcommander as a client to the Intel AMT and maybe ME, which if you do some more research, you'll see is indeliby baked into most newer Intel hardware.

I assume people aren't making non-x86 desktop or workstation systems because they cost a lot of money to develop, even just a motherboard. If the system cannot run Windows, you've just made the whole idea economically infeasible.

I think the issue is more demand than supply. I'm currently working on an embedded software project where the arm based SoC supports 2 x4 pcie buses, 3 usb ports, sata and ddr. The evaluation board looks like a normal motherboard with 2 pcie slots, a ddr slot and usb and eSata ports. And it was running ubuntu.

People know how to build non-x86 desktops. Building non-x86 desktops is not difficult to do or expensive. People just don't want it.

What's the price of that evaluation board? Can the CPU run faster than in embedded scenario thanks to better cooling?

A comparable example: https://boundarydevices.com/product/nitrogen6x-board-imx6-ar...

This kind of thing is stuck in a volume trap: you need to produce enough units to make it cheaper than a PC, but nobody will buy it until you do. In the meantime, people are buying ARM tablets and 'mini-PC's from various Chinese OEMs, but they're non-modular units.

The modularity of the PC is almost an accident of history.

So that's what all that Windows RT stuff was about. I'm a bit ashamed I didn't realize Windows (proper) doesn't run on ARM. I'm assuming MS really wants to be able to run Windows on different types of processors but can't practically make that happen and keep backwards compatibility. It sounds like they've shot themselves in the foot. Unfortunately, they were stepping on all our heads at the same time.

Windows NT was written to be cross-platform, and it was actually developed on RISC chips -- hence the HAL (Hardware Abstraction Layer). Unlike Linux, it wasn't written on and for an x86 processor....

At one time, commercial Windows NT PCs were sold with DEC Alpha processors.

However, the volume of x86 PC sales drives huge economies of scale that make it hard for other platforms. This actually benefited Linux in particular, but it also encouraged Apple switch from its own hardware to PC-based x86 hardware, which resulted in cheaper Macs.

Microsoft has never developed an OS specifically for Intel x86 chips (ie Windows NT and CE: the basis for DOS was bought in) to try to avoid being wholly dependent on Intel. Intel has also supported alternative operating systems to avoid being wholly dependent on Microsoft. But neither is immune from market forces.

I still have an old NT 4(?) install disc with x86, MIPS, Alpha, and PPC arches.

NT4 was the last to support MIPS & PPC, but only x86 and Alpha got SP3 and beyond.

Digital/Compaq did have a beta build of Win2k for Alpha, but the plug got pulled right before the RTM.

Did the code ever escape in any form?

Windows itself ran fine on ARM. Windows RT failed for a couple reasons: there was no backwards compatibility for x86 Windows software, and there were a couple artificial limitations in place that made ARM-native desktop software impossible. Essentially, Windows RT threw out the good part of what makes Windows Windows (decades worth of third-party software) in an attempt to make a "dedicated" tablet OS a la iOS or Android.

True. However, Microsoft thought users would be OK with Microsoft Office (bundled free with Surface RT) and sandboxed tablet apps.

Windows RT threw out the good parts of backwards compatibility, but it also threw out the bad parts: viruses and malware, software downloads from dodgy websites, being your own IT manager etc.

Some would consider this a reasonable trade-off, but the market didn't buy it. Odd, really, since Windows RT has all the benefits of a Chromebook, but also offers much more.

If MS had enabled Domain Join & Group Policy, the RT devices could have taken a lot of enterprise sales. At the time, there wasn't a great story on any tablet around user account management; RT could have been that platform.

Agreed. It would also have filled the hole for school testing, which is what generated sales of Chromebooks....

The Amiga people kept getting new PPC's built for their systems. Almost nobody bought them. Gaisler open-sourced his Leon3 SPARC CPU and I.P. under GPL. Sun did same for OpenSPARC T1 & T2. Nobody built anything or pooled together money. MIPS and ARM stayed in embedded space mostly with big vendors (esp SGI) switching to Intel/AMD since they performed better per dollar. That along with licensing cost meant nobody was doing a desktop CPU there since it be a financial sinkhold. I still think Cavium's Octeon III and ThunderX could become great desktops with a PC board. Yet, they show no interest in that outside cloud servers maybe. PA-RISC and Alpha similarly died since no demand to pay for increasing expense on CPU's. SuperH, which powered Dreamcast, stayed in embedded with higher-end one doing multimedia products.

So, not only are there no desktops, everyone that tried either sold jack and/or went out of business. Your theory hasn't been true for a long time. Linux and FOSS types talk a lot but buy very little if it's not x86. A desktop could easily be made in Shenzen or whatever based on existing components, esp Leon3 or Leon4 SOC. eASIC's mask-programmed tech already supports Leon to crank them out quick and cheap. Peripherals could be handled with cheap microcontrollers if necessary that have onboard HW accelerators for them. The tech is there just no suppliers right now and usually no actual buyers.

That's the $1000+ server market. In Germany it sells for 850 EUR : http://www.heise.de/preisvergleich/gigabyte-mp30-ar0-9mp30ar...

That's perfect in every way except price.

The CPU is 2.4GHz and 8 cores.

It can handle 64GB ECC DDR3.

It has onboard VGA since GPU support (which typically involves x86-specific initialization instructions at boot time) will be spotty.

Remote management (dedicated NIC) w/ iKVM and VGA.


2x 10G SFP

What's not so good: only USB 2.0 - two external ports and one header. Not sure how many root hubs and what goes where. That said, you can use eg a USB3 HDD (or do almost anything else involving USB3) via the 20Gbit Ethernet to another box, so...

I'd settle for a standard power/backplate, maybe about 0.6x the size of the PC backplate that's used from ITX through ATX boards... then have the MB size be that, squared... which is bigger than most enthusiast devices (rpi and similar). Something with a fairly standard mounting interface, and power adapter would be a good start.

From there, a standard interface for booting and mounting the OS and additional devices. There'd be a lot of work in that, and a lot of competition, but would be nice to see some work in the open for a more standard platform for ARM devices.

They were available in Germany mit 90ies. Nobody bought them.

Vobis Highscreen Alpha 5000, based on alpha processor 21164PC

Hardware companies are really really traditional in their business thinking and do not understand virality or 'priming the pump' at all. They do a straight-up market size analysis, conclude that the (current) market is small, and move on.

Their margins are low, and it's super expensive to tool up a factory, so they are (intelligently) going to be somewhat risk adverse.

Nvidia Shield console? 3GB of RAM, a fancy video card (with support for CUDA), and gigabit ethernet. The guys at Hypriot have even demonstrated a working Docker install [0]

The custom partition layout is kind of wonky, so I'd likely just by the non-PRO version and use USB 3 and an SSD for additional storage. I'd be happier still if Nvidia supported running other distros on it -- I wouldn't even be mad if I could no longer use it for gaming or watching movies.

[0] http://blog.hypriot.com/post/getting-docker-running-on-a-hig...

Or the Jetson TX1 dev kit, $599

- 4GB RAM - 1GB Ethernet - 802.11ac wifi - 16GB eMMC onboard - m2 slot - PCIe Gen2 x4 slot - SATA interface - USB 2 and USB 3 ports


NVIDIA should really sell a case for it to make it into a NUC-style desktop.

> <http://www.nvidia.com/object/jetson-tx1-dev-kit.html>

Really wish hackernews would fix this bug already.

Interesting. Honest questions: If I would buy one, can I count on driver updates for at least 3-4 years? And after 3-4 years can I count on having Open Source drivers not much worse than proprietary?

Linux kernel 4.5 has quite a lot of changes to support the SoC on the Tegra X1 in the mainline kernel.


There is basic GPU support via Nouveau, though for any serious OpenGL/Vulkan/CUDA or video stuff you'd want to use the NVIDIA proprietary drivers, which are very good on Linux.

> There is basic GPU support via Nouveau, though for any serious OpenGL/Vulkan/CUDA or video stuff you'd want to use the NVIDIA proprietary drivers, which are very good on Linux.

And for how long will NVidia be providing support for the driver?

Is it not the thrust of the article that they don't want to 'make do' by repurposing fruity consumer products, but want actual workstation hardware, or at least hardware that is 'serious' by virtue of having utilitarian form-factor?

I've got the Shield TV, and it's really nice. I've seen instructions on how to install Ubuntu on it but haven't tried yet. http://forum.xda-developers.com/shield-tv/general/ubuntu-uto...

I wonder if dual boot android/ubuntu is doable?

I've felt this same frustration with POWER for what feels like forever. I'd love to build and run a POWER desktop system, but traditionally there hasn't been a readily available, standard form-factor, somewhat inexpensive motherboard that you could just order up from Tyan or whoever. On and off over the years there's been a lot of talk about how this company or that was going to build a standard ATX motherboard for POWER cpu's, but they seem to never materialize, or if they do, they're insanely expensive, or something which keeps the whole idea just out of reach.

This most recent push for openness in the POWER world might actually change that though. Tyan does have a kinda-sorta not-too-expensive POWER8 based server available, and there was a document floating around a while back listing all the companies who were committing to build machines, motherboards, etc. based on POWER8. As I recall, there were a lot of names on the list, which is encouraging.

But still... why can't we hit up NewEgg and order an ATX motherboard, a CPU and some RAM for, say, $1000.00 or less? sigh

Guess you are waiting for TALOS to be released? Bit too expensive though.


Yeah, I like the idea behind TALOS, but nearly $4k for just the motherboard and CPU??? That's a bit much for me to swallow. :-(

I agree that 4K is a tough pill to swallow, hopefully something like this eventually goes down to the ~2K price range, I could probably jump on it then.

If people don't help support it at $4k, it will never reach $2k.

The issue is at 4K I might as well buy a "proper" POWER server from IBM (like the S812LC) for the same price since I don't need a "workstation" form-factor (I've already got some rackmount equipment at home, and my current development system is just a Fedora VM running on my home server).

The S812LC is designed, and made, by Tyan.


If by 'proper' power system you mean IBM Pseries, this is not it.

That's not that much actually. Look at Thinkpad P70 or other Xeon-based high-end workstation.

It is possible to know how far they are into having enough pre-order ?

While not an ATX Desktop board, there is an initiative to make a POWER-based notebook: http://www.powerpc-notebook.org/en/

I think nintendo wii has a power cpu and you can run linux on it !

The big limitation is that you only get 96MB of RAM. But still, it's a pretty reasonable choice if you want a big-endian test system on the cheap.

Pretty much all ARM boards will run big endian, although not sure Linux has armeb distros much (NetBSD works).

Nice... I had no idea you could run Linux on the Wii.

There's only ~100MB of RAM so it's a bit ascetic. You're better off running an old PowerPC Mac with Linux as that'll give you better but nearly equivalent hardware.

What about the Gamecube? Might even be a cheaper option.

It has been done before (http://www.gc-linux.org/wiki/Main_Page), but the simplicity of jailbreaking the Wii, plus its ability to connect to the Internet, make it much more viable for anything more than a simple project.

I got interested in ARM desktop & laptop hardware while I was looking for more open solutions. Remarkably, the only modern hardware that is truly open top-down, including firmware, are cheap Rockchip Chromebooks [1].

I'm also looking forward to POWER offerings, which might be more performant and thus suitable as workstations [2].

[1] https://libreboot.org/docs/hcl/c201.html

[2] https://www.raptorengineering.com/TALOS/prerelease.php

This is still 32-bit only, but you might be interested in the EOMA68 modular computing hardware campaign on Crowd Supply. [1] Since it's modular, at least we can potentially have a 64-bit computer card later, while re-using the micro-desktop and laptop housing.

[1]: https://www.crowdsupply.com/eoma68/micro-desktop

Or the Novena, also on CrowdSupply [1]. It's the closest to a "real" ARM computer as I've found. It has 4GB of RAM, SATA, and all the peripherals you'd expect for a laptop or low profile desktop. I've used it to design PCBs in KiCAD - second generation open hardware :)

Though again, it's only 32bit, and while that works great for most purposes, it still leaves the OP's point valid. (I have one of the early 64-bit 8-core hikey 96boards mentioned, and I'd take the Novena over it any day)

[1] https://www.crowdsupply.com/sutajio-kosagi/novena

I use an RK3288 as my main dev machine these days. It is faster than the 64 bit x86 thinkpad it replaced. Overall, pretty nice. Some of my experiences are summarized here: http://kmkeen.com/c100p-tweaks/

In terms of this article, the RK3288 is a 32bit ARM7.

I've also been looking at this. I think the Google Pixel C might be a good option, for a 64bit ARM laptop. I've seen some talk of running Ubuntu on it. But it uses the Nvidia TX1 which has good Ubuntu support on the TX1 Jetson.

I thought the Pixel C bootloader was looked. Do you have any pointers to instructions on how to boot Ubuntu on this machine?

The problem with ARM architectures is the lack of a standard boot system.

It is locked. There is a switch like with the Chromebooks, but to reach it you have to disassemble it. OTOH rooting and running custom kernels works. Not ideal but it works for some use-cases.

edit: The Pixel C runs on Coreboot of course.

Yep, but has anyone successfully installed Linux (aside from ChromeOS, obviously)?

UEFI is the standard firmware these days, at least for ARM servers.

Those who have tried the Odroid C2 will not fail to be impressed by how delightfully close it is to offering a significant desktop experience for a $40 device. It's nearly unbelievable that this tiny, portable device sips powers and can run accelerated videos upto 4k, runs 64 bit Ubuntu Mate and even 720p Youtube on Chromium. And it could be more but for ARM.

The potential for disruption is massive and yet here is ARM dragging its feet with it's increasingly inexplicable and extremely closed nature. Things only work on Android, the GPU especially is a big problem, and given Android runs the Linux kernel I am sure Arm and Google are having a laugh at the point of open source.

Most current mid to high end ARM SOCs and those coming make Intel's pricing and value look extremely questionable. These SOCs are a fraction of the price Intel demands. Enough to raise doubts about the future of the platform beyond high end gaming and GPU use cases. And with Vulcan even that is up in the air.

The iPad Pro benches close the the Macbook. The A72 SOCs are at 1500 single core and 4000-5000 multicore on Geekbench.

This is real. Yet ARM for reasons best known to them either do not see the potential or do not want to pursue it, giving Intel a breather. Now with Softbank acquiring ARM at this critical juncture its anyone's guess where this goes.

There's this little desktop box, sold as a development system [1]:

    AMD Opteron A1100 series processor
    4 x 64-bit ARM Cortex A57 Cores
    2 x RDIMM with 8GB DDR4 DRAM
    1 x 1GBase-T Ethernet
    2 x USB 3.0 ports
    2 x SATA 3.0 ports
    1 x 1TB HDD
(Pre-Order Now – Delivery July) - not good seeing that on July 25th.

No graphics, though. It's really a server board in a desktop package.

There are lots of small boards which are really tablet-type systems-on-a-chip where the connections from the chip were brought out to connectors. (That's what a Raspberry Pi is.) But they don't usually have PCI or SATA interfaces, so disk I/O capability is limited. There are good embedded 32-bit ARM boards, but the embedded world isn't going 64-bit yet.

Everybody seems to assume you're going to cross-compile and remote debug for embedded and server work. QNX doesn't even self-host any more. (I rather liked having a QNX desktop.)

[1] https://shop.softiron.co.uk/product/overdrive-1000/

Site now reads "Pre-Order Now – Delivery August". Even less good.

What's wrong with the Gigabyte MP30-AR0? http://b2b.gigabyte.com/products/product-page.aspx?pid=5422#...

It's sold as a server board in the $1000 range, but seems to fit the bill as far as the points in the blog article are concerned.

We bought a Gigabyte 1U server based on one of these at my old workplace.


It definitely seemed a half-hearted effort from Gigabyte. Both our usual suppliers were able to source one, but the delivery date kept slipping. It did eventually arrive, months later, and by then we'd lost enthusiasm and/or found alternative ways of doing the work we'd planned for it.

Given my experience with another Gigabyte board, I can't recommend using their server platforms until they show a bit more care.

I know they've been doing custom integration in this space for a long time -- I remember an article about Google Nehalem servers, for example, that were OEMd by Gigabyte. And many of their efforts look like contract designs pushed out to market -- nothing wrong in that.

But they need to take a lesson from Supermicro's book in terms of attention-to-detail in and end-user support before I'll consider them worthy again.

Which is a shame, because their new CaviumX ARM platform looks very interesting indeed: http://www.servethehome.com/cavium-thunderx-micro-benchmarks...

A bunch - it's using the relatively ancient X-gene 1 chip, it doesn't come with SBSA/SBBR out of the box (although it is installable), it's expensive.

There are some good things though: The IPMI support is excellent, it takes 128GB of RAM, and it's a solid workhorse. Here is mine: https://rwmj.wordpress.com/2016/03/04/gigabyte-mp30-ar0/

I'm waiting for the new Cavium-based Gigabyte boards, which have 48 up to 384 (sic) cores.


> I'm waiting for the new Cavium-based Gigabyte boards, which have 48 up to 384 (sic) cores.

Are you sure that these core counts are per board and not per server/enclosure (i.e. 384 = 8 nodes x 48 cores per node in one case)?

Correct, the "384 core machine" is 4 blades in a 2U enclosure. Each blade has two sockets making a total of 96 cores (per motherboard). Picture: http://b2b.gigabyte.com/products/product-page.aspx?pid=5460#...

I've been shopping for a low-power server board, and I absolutely want that one. The problem is, nobody seems to sell it. I can't find any notice that it's been discontinued, but every reseller that google's given me has said that the board is out of stock (I think I found one in Spain, actually, but I'd rather order from somewhere in NA). Do you know where a person can actually buy that board?

They are definitely available in the EU. I got one from xcase.co.uk. Note there is a slightly newer variant now (MP30-AR1), although it still uses the X-gene 1.

If I were you I'd wait a bit for the Cavium-based Gigabyte boards. I have pricing for these which unfortunately I'm not permitted to reveal, but it will be very competitive.

The Caviums have lots of cores (48 / socket), but each core is pretty slow, like an Intel Atom. Caviums have tons of fast I/O. The challenge is to make software which can deal with this, since it's quite unlike Intel hardware (few, fast cores, slow I/O).

I feel like this might be going off topic, and maybe you can't discuss the specifics anyhow, but do you mean very competitive with the existing AR0/AR1 board pricing? The only Cavium-based systems I'm seeing are the full server machines, and those are listing 3 40Gbps and 4 10Gbps ethernet ports, plus the rest. That board can't possibly come in anywhere near $1000, can it? I'll be keeping my eyes open for sure, but I'm also eager to learn more :)

Not in NA, but looks like this company has one in stock: https://www.xcase.co.uk/products/gigabyte-mp30-ar0-with-appl...

The author seemed to imply that $1000+ was too much to pay.

Regardless, I did a quick search and couldn't find anyone actually selling that board or listing a price.

> I did a quick search and couldn't find anyone actually selling that board or listing a price.

Large german online store Mindfactory claims availability with 3-5 days for delivery time: http://www.mindfactory.de/product_info.php/Gigabyte-MP30-AR0...

I'd love a system that fits the form factor of my desktop motherboard but supports an arm chipset.

My main system mother board is a little... strange from the normal. I've build my system around being 100% silent and low power while still having enough juice to run some IDEs and be productive.

My current mobo is from ASRock and has an interesting feature [0]. It's got support to use a laptop charger as a PSU built into the board.

If there was a low power, passive, ARM system for a desktop like this I'd switch. This is just the lowest power and quietest build I could think of currently. (Only problem is no MSATA :()

[0] - http://www.asrock.com/mb/AMD/AM1H-ITX/

Whatever happened to the Acorn RISC PC's from the mid/late 90's?

I remember guys in the UK on USENET with the 'StrongARM...' in their signatures.

Acorn seemed to fail to advertise their products outside the education market, which rapidly dried up as education shifted to teaching Microsoft Word on cheap PCs.

The final workstation system was cancelled in 1998 [1], ARM spun off as a separate company, and the remaining part of Acorn made set-top boxes for digital TV. It was then bought by Broadcom. [2]

[1] https://en.wikipedia.org/wiki/Phoebe_(computer)

[2] https://en.wikipedia.org/wiki/Acorn_Computers#1998.E2.80.932...

When you consider how shit the RM Nimbus was, I refuse to believe Acorn lost on technical merit, or even on price/performance.

RM made some truly shockingly bad computers. But they had (have?) very good contacts at high level in government.

RM supported PCs, and PCs were what schools wanted. Also, the hardware was only a part of the whole RM infrastructure.

Considering the UK government subsidized the cost of Acorn computers in schools, and considering it got massive amounts of free BBC advertising for the Acorn platform, I think you need more than a conspiracy theory to account for RM's success and Acorn's failure.

Not true. In fact Acorn paid the BBC a commission on every machine sold for use of the logo.

And the RM infrastructure was just as shitty as their hardware. If people were desperate for PC compatible, why was everything on the Nimbus non-standard? They even used the 80186 that no-one else did and you had to emulate an 8086 on it to run any third party software!

Which bit is not true? It's certainly true that the UK government subsidized the cost of computers in schools, and Acorn got the largest benefit from that scheme.

> In fact Acorn paid the BBC a commission on every machine sold for use of the logo.

Citation? How did it compare with the UK government paying half the cost of a BBC B?

The 80186-based Nimbus wasn't PC standard and wasn't particularly good (yes, I did use one), but RM did move to standard PCs.

> And the RM infrastructure was just as shitty as their hardware.

In your opinion. But at least it had one.

Your timescales are well out. The BBC Micro was promoted in the early 1980s. By the end of the 1990s, schools were buying awful PCs from RM, when they could have gone to (eg) Dell and got much better hardware for much lower cost. Why were they still buying stuff from RM - it makes no sense, right? Because RM had the right contacts in LEAs and government.

Your timescales are well out: this was the 1980s.

> By the end of the 1990s, schools were buying awful PCs from RM, when they could have gone to (eg) Dell and got much better hardware for much lower cost. Why were they still buying stuff from RM - it makes no sense, right? Because RM had the right contacts in LEAs and government.

I'm sure some did buy from Dell, but RM was an approved, specialist supplier to the education market. This involves a lot more than knocking out cheap PCs.

RM Education is "The division that deals with technology infrastructure, software and services - including learning platforms, interactive classroom equipment, connectivity, networking software, school management software and support services." https://en.wikipedia.org/wiki/RM_Education

Business wise, about the same thing that happened to the other groups with their machines after the 8-bit transition: Intel & Microsoft. People just wouldn't buy machines that had inched up into the same price range as PCs without the compatibility. Combine that with some of the worst marketing and channel relations gets them all to disappear in the 90s. At the time I had hoped that UK / European manufactures would survive, but it seemed more like a short delay before the tsunami.

I dearly wish someone had been able to keep alive the alternate segment (other than Apple who was premium even with 8-bits), but it looked like video game machines took the money from that market.

Commodore Amiga had by far the best chance. Commodore could have been Apple with better management.

Also somewhat related but MIPS, not ARM. The Cobalt Qube [1] and Raq [2] (though later ones replaced the MIPS CPU with AMD K6/Pentium)

[1] https://en.wikipedia.org/wiki/Cobalt_Qube

[2] https://en.wikipedia.org/wiki/Cobalt_RaQ

I remember those weird little web hosting appliances. Found more than one customer who was locked into something developed on one (they were easy to migrate to a generic RedHat setup) and we even co-located a handful of them for people.

Yea. ARM was originally a desktop processor and then was spun out from Acorn and moved toward mobile/low power. Bits of Acorn eventually got bought by various people and it no longer exists.

And x86 was original embedded...

Just out of curiosity, what are the advantages of using ARM on a desktop over a x86-64 architecture for an end-user? Or it is just to enable developers targeting server softwares to buy a less expensive hardware? I can see the advantages of using it on servers and IoT though.

There the NVIDIA Jetson TK1 and TX1. An excellent desktop platform IMHO. When the TK1 came out, you could get one free by telling them what you planned to do with it.

Unfortunately they tripled the price from the TK1 ($200) to the TX1 ($600). You can get a TK1 on ebay for ~$150.

This thing is much faster than the million dollar supercomputer I used in college ;)

This thing looks pretty nice, but the catch of course, is that you need all the expansion boards as well, so the costs go way up. Plus, there's the whole availability thing .. I'd love to hear from someone who has one and can tell us what its like as a machine.

I'm in the camp of those who would like to have a smart Linux workstation with modern peripherals and not be x86, just for the kicks of it .. so actually, the TX1 seems like it might be a pretty viable investment.

I have a TK1; it's perfectly usable as a desktop machine. I don't use it because I don't like Ubuntu and the project I was using it for has finished, but plug a SSD into it and it works a treat.

Would you consider selling it, then?

Oh, sorry, missed this. Sure. Send me an email to the listed address if you're still interested.

Sorry for the delay. Yes, I'll sell it.

I also have the TK1 and find that it works fine. I added an SSD drive to mine. The Puget Systems acrylic enclosure has an SSD mount.

Maybe Apple might switch from Intel to ARM for the Mac at some point. They've switched architecture twice before (68k to PowerPC, PowerPC to Intel), they could do it again. I'm sure they'd relish the chance to design their own desktop CPUs, not just ones for mobile devices.

NextStep ran on 68k, Sparc, PA RISC, and I think a few others.

Btw - laptop CPUs are A LOT like mobile CPUs. Single threaded performance on Apple ARM CPUs are very competitive with recent Intel laptop processors.

I keep hoping that's why there's been no new Mac hardware recently ...

more likely that they will retire the whole MacOS line at that point. Not sure if they have alienated enough of the media production market to do so though.

Why would ARM be the death of macOS? If it can run on Intel, PowerPC and 68k, what's another processor transition?

And why would that “alienate the media production market”? Apple could build a beefy ARM machine if they needed.

SoftIron have a 'SoftIron 1000' for $599


Did ARM ever get a formalized platform? I remember reading that one of the major issues with adoption of ARM was that there wasn't a common platform for the CPU to interface with external devices. It was all custom and proprietary which is fine for embedded systems but not so good for desktops or consumer grade PCs.

Is that no longer the case?

So is that perhaps the reason why there has been little adoption or investment from OEMs into producing standardized boards?

Yes, sort of. The other problem, shared with 32 bit ARM, is that there are billions of shitty Android chips made each year, and it's very easy to take one of these, slap it into a development board, add a proprietary kernel with random blobs, and sell it for $50.

These crowd out properly designed server hardware, or anything that cares about freedom, upstream Linux, maintainability, virtualization, etc.

Well, plugging things directly on the processor bus is something that is "out of fashion" since the ISA days (in the PC world at least)

Right but isn't it also advantageous from a low power perspective because it reduces the number of chips and over all power profile of the device?

How about the Gigabyte MP30-AR0, apparently actually available to consumers through this one retailer?: https://www.xcase.co.uk/products/gigabyte-mp30-ar0-with-appl... It seems Richard WM Jones has got RHEL running on it: https://rwmj.wordpress.com/2016/03/08/gigabyte-mp30-ar0-flas...

Aside from the Overdrive 1000 as already mentioned by others: http://softiron.co.uk/products/, there is the LeMaker Cello forthcoming (which was supposed to have already shipped but...) available for pre-order here: http://www.lenovator.com/product/103.html

Not loading for me, here's the google cache: https://webcache.googleusercontent.com/search?q=cache:FbgwL1...

Loads fine for me, just a moment ago. Try:


I have a nvidia shield just sitting around (because to be honest it's pretty useless as a console / with android on it), I've been considering looking up to see if it can run Fedora / CentOS's ARM port for some time. This might just be the push over the edge I need...

Why not make a 32 core arm chip with an amd pin out so it can fit into existing low cost motherboards ?


We detached this subthread from https://news.ycombinator.com/item?id=12158651 and marked it off-topic.

I think the parent mentions both and I agree with their comment. A desktop motherboard, to me, includes things like audio and video support which can be a pain to support.

If all you want is a board with a USB port and memory then you can have that today.

>A desktop motherboard, to me, includes things like audio and video support

Ideally, what most of us want, I think, is a ARM ATX motherboard, with a few PCI Express slots. I believe that it would limit the selection of processor though.

Yes, exactly that. The typical PC motherboards start at $100, and have so since the 90ies. Just one of those, with a socket for their cpu and a translated bios just good enough to bootstrap Linux should not be so much more expensive. Even if they had to sell it at cost it would be worth spreading the technology.

They are, however, mass-manufactured, and how to build an Intel motherboard is well-known, there's very little research cost involved. An ARM motherboard really couldn't be mass-manufactured - there wouldn't be enough demand - and you're starting almost from scratch in terms of knowledge of how it integrates with the processor.

Also, if you wanted a socket, you'd have to actually design a socket for your ARM chip, and then sell ARM chips that are socket-able, with few people who actually want them.

Just buy Raspberry Pi 3 and you will have 64-bit ARM desktop hardware...

The raspberry pi 3 only runs in 32-bit mode right now [1]. There is no 64-bit kernel for it currently.

[1]: https://www.linux.com/news/raspberry-pi-3-still-essentially-...

The raspberry pi 3 runs in 32 and 64 mode. There is no >>LINUX<< 64-bit kernel for it currently...

Exactly which OS runs on rpi3 in 64-bit mode? Because Windows 10 and RiscOS ain't it.

Have not tried this myself, but it seems like FreeBSD does: https://news.ycombinator.com/item?id=11226386

Author is asking for 64-bit ARM hardware, not software...

You, however, were claiming, that rpi3 does run in 64-bit mode, just not with linux:

> The raspberry pi 3 runs in 32 and 64 mode. There is no >>LINUX<< 64-bit kernel for it currently...

To my knowledge, rpi3 has 64-bit CPU, but all operating systems and bootloaders available run only in 32-bit mode, no exceptions. That means that rpi3 in future theoretically can run in 64-bit mode, but certainly does not today.

Pee bootloader runs on the gpu, so that doesnt matter much. Current bootloader can bootstrap ARM in 64bit mode, as is evidenced by freebsd port and https://www.raspberrypi.org/forums/viewtopic.php?f=72&t=1379... (all but the gpu works)

You are claiming that you know >>all<< operating systems and bootloaders and that >>all<< run only in 32-bit mode - with no exceptions... a big claim about your knowledge...

So you know even all non-public bootloaders and OS-es...

Whoa - really - this is interesting. Why did they do this? Memory too constrained?

My understanding was that the limited memory meant that it wouldn't be very useful, and they wanted to make sure that software for the earlier Pis would still be 100% compatible, and vice versa.

The author is looking for something that can replace a conventional x86 based desktop. You can't plug SATA disks directly into a Pi (or boot from them), you can't directly mount it in an ATX case, you can't plug more RAM into it, and you certainly can't attach an Nvidia GPU either.

The Pi 3 is strictly inferior to the ODROID-C2 and Pine64 in the areas he cares about - less RAM, slower I/O, proprietary bootloader that can't even boot 64-bit kernels, etc. Plus it still doesn't use a standard PC form factor. It's not even cheaper either.

It's not developer class desktop hardware, at least for native code. Good for all kinds of things, but doesn't satisfy what the author is looking for.

I finally saw a retail package in an actual store right next to the drones. Quite a selection of packages (e.g. bare, complete, etc.) too. It is one of those, "I know what those are" purchases since there was no support material (signs, books, software) around them.

Does any software actually enable the 64-bit mode yet?

Not end-user usable software, as far as I can tell. It appears that some people have built 64 bit kernels but with many features broken or missing.

Ultibo may soon be working in 64-bit mode: https://ultibo.org/forum/viewtopic.php?t=60

Registration is open for Startup School 2019. Classes start July 22nd.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact