$10 Orange Pi 2G-IoT Released to Compete With Pi Zero W

A new single-board computer by Orange Pi has popped up for sale on AliExpress. The Orange Pi 2G-IoT is designed to compete with the Raspberry Pi Zero, and if specs are anything to go by they have done a nice job.

There are a lot of options for extra small single board computers these days and there’s a growing list at the lowest price points. Let’s call it the sub-$20 cost range (to quell the argument of shipping fees). We have seen C.H.I.P., the Raspberry Pi Foundation released the Pi Zero W (an update to the Zero line that included WiFi and Bluetooth), the already available Orange Pi Zero (which was featured in a project on Monday), and now add to that list the unfortunately named Orange Pi 2G-IoT.

The 2g-IoT is sporting an ARM Cortex-A5 32bit clocked at 1GHz with 256MB DDR2 RAM. It’s nice to see 500 MB of on-board NAND to go along with an SD card slot for larger storage. It also has a CSI camera connector, WiFi, Bluetooth, an FM Radio and GSM/GPRS with a sim card slot on the bottom. It is pin compatible with Raspberry Pi’s almost standardized GPIO layout.

All this for $10 is quite impressive to say the least, especially the addition of GSM/GPRS. Will it kill Raspberry Pi Zero W sales? We think not. While the Orange Pi’s are great little computers, they don’t have the community support that is afforded to Raspberry Pi products making for less support online when you run into a problem. That’s if you can even get the thing running in the first place. The Orange Pi’s website has not yet been updated to reflect the new release. However if you are interested in getting one for yourself right now, head over to your favorite Chinese electronics supplier.

[via Geeky Gadgets and CNX]

72 thoughts on “$10 Orange Pi 2G-IoT Released to Compete With Pi Zero W

    1. There is no Armbian port for this board yet and it is questionable if there will be one. Read the CNX comments. There is a vendor supplied 3.10 kernel with an ubuntu user space.

      Given that 2G is in the process of being shut down in several countries, you should make sure that you can get a new 2G activation before ordering one. Just because a carrier’s 2G net is still operating does not necessarily mean that they are going to allow new devices to attach to it.

        1. The Orange Pi PC2 (which you linked to) is a completely different board. As @jonsmirl pointed out, the 2G-IoT is a completely different design, and right now it’s unknown if Armbian will bother to support it.

          1. I have no actually-useful-for-software-support data on this ‘RDA’ thing; but ‘2g cellular + crazy cheap’ brought this report on a $7 phone to mind.

            Likely the same vendor, who knows how many models/variants they have. The optimist would assume ‘some kind of cheapo ARM, please?’ rather than it being something really weird with limited toolchain options(or even OS options; it’s generally sort of assumed that ‘cheap single board computer’ will be Linux compatible, even if it’s an awful, blobified, Android-focused BSP; but that isn’t a given).

            Pretty unlikely to be an Allwinner; or even look much like one.

      1. It isn’t just a Zero build with an extra driver? And yeah I know about the 2G thing, that is why they are cheap, they are only relevant to less wealthy parts of the world, where 2/3 of humans live.

        1. I wouldn’t call all of Europe the less wealthy part of the world, as long as you have a valid SIM card you can connect to a valid network pretty much anywhere and roam around without problems. Sometimes you do have roaming fees, but you even have GPRS M2M packages that cost something like 1 euro per month per megabyte, that sounds low but you can send a lot of samples from a remote sensor with 1MB.

        2. Not sure where you live but for most of the ‘wealthy world’ the 2G network is still active alongside the usual 3g and 4g, it doesnt make much sense to disable 2G all-together just because theres something faster available, theres more then likely tech thats happy to use this slower and now mostly unused network (like IoT devices, emergency services communication, etc)

          Its kind of silly that 2G is already deprecated in the USA, apparently people with slightly less up to date devices or not living in/near a city (read; 3G users) are not important/IoT in the USA will go over 3G now, pinching the connection.

          You dont wanna have all the IoT stuff run over 4G, besides the coverage problems, lets face it, nobody will accept stuttering in the Netflix stream on their iPhone because all refrigerators in the neighborhood decided now is the time to pull in a firmware update.

          1. Here in Oz the 2G has been turned off so unfortunately that part of the board is useless :(
            I wish it was a 3G for that price but 3G modules look to cost much more. For example, a 2G shield for an Arduino can be got for about $20 but a 3G is around $100.

          2. @[dendad]

            I (in Oz to) was given a M2M project about 2 months ago and 3G or 4G/LTE modems were around the $80 – $100 mark at the time because there was a vacuum because 2G was on the out so I delayed the project.

            Now I see them going for as low as $20 on ebay for a new product (3G). These are PCI-e modems and not full modules but you just need some comms and a sim socket / PSU / UL.f to antenna and your done. Admittedly the $20 modems are useless due to unavailable or poor documentation / libraries but they are getting cheaper and around the $30 – $50 mark you can find a useful solution now.

            For OZ the trick is to find a product that has libraries for all the major network providers and enough FLASH to fit all these libraries in or you have to re-FLASH to change carrier.

            Eventually we will have *narrow band* LTE (on 4G) but historically modem providers don’t provide firmware upgrades so your better off using the lower power and cost 3G for now and replace hardware when narrow band LTE becomes available. Narrow band LTE is an upgrade to normal LTE that allows narrow band low power, low data rates just to suite M2M or IoT.

        3. The Zero-W is just a zero with a wifi/BT device added; this ‘Orange Pi’ device appears to be based on something entirely different than other ‘Orange Pi’ hardware; hence the concern about support.

      1. In many regional areas too… one of our customers was relying on the 2G network for hand-held computers.

        They had replacement WiFi devices already in stock, ready to go, they just needed to roll out the network, which was moving along at glacial pace.

        One day, Telstra turned off the 2G in Townsville, and they rang us complaining their computers had stopped working. Mass hilarity ensued.

          1. The 2G computers were supplied about 8 years before… they would have been chosen based on the needs at the time. Unfortunately I can’t ask the relevant people now, one left the company a few years ago, the other died last year (and is the reason for my helmet project on hackaday.io).

    1. Here in Europe I don’t see it taking less than 4 years for that to happen anywhere, when it happens you upgrade it for something with a LTE modem instead. I wouldn’t be surprised if 3G went out first, at the moment there is an insane amount of automation and telemetry that relies on 2G/GPRS.

      1. ^ This.
        A lot of people dont seem to realize that its more logical to re-purpose a network like 2G (like has been happening everywhere but in America) then to just shut it down.

        1. The 2g shutdown in the US has been happening over the objections of a wide variety of M2M/distributed sensor/etc. customers; it’s not as though the telcos just can’t think of anything to do with a slow network, or like throwing away working hardware just for fun; the issue is RF spectrum:

          2G is less efficient, in terms of spectrum required for a given amount of data transmission, than later standards; so the telcos want it gone so they can use the spectrum for more efficient and/or more profitable services.

          I’m not sure if the competitive landscape is just different in Europe, or the early embrace of GSM means that the embedded users are more numerous and vocal; but so long as spectrum remains a valuable; and limited-by-the-pesky-laws-of-the-universe asset, you can’t keep around a legacy network without occupying space that could be used for a new and shiny one.

          If 2G could silently overlap, they’d probably keep it lit for years to come; so long as someone would rather pay jacked up prices for GPRS and SMS than upgrade their own systems; but it can’t; so the knives are out, because there are other uses for the spectrum

  1. OrangePi’s major problem is they don’t bother to get their software ducks in a row before each new product launch. All the operating systems, manuals, dimensioned drawings etc should be in place, on the website BEFORE releasing the hardware.

    1. I see this as a rush to market to capture and create demand for their product kind of move.Also, it could be that they are doing so thinking that the community will be the ones to help fill in the gaps of support and then officially bring out support to augment the communities effort. Backwards? Yes, but it does help build sales after the fact. It’s a gutsy move on OrangePi’s part.

  2. While this board tempting from budget-wise perspective, the lack of software support really BIG thing to consider, at least from noob like me.
    I remember when i got free intel galileo board, that thing is really a hard PITA and enlighten me somehow
    never again i said to myself
    never again

  3. I will buy one of these because unlike the Raspberry Pi Zero, I *can* actually buy one at a reasonable cost and the Orange Pi (including shipping) is less than a fraction of what I would have to pay for shipping alone for a Pi Zero-stock.

    Looks like a good media server. For me it will be a Wi-Fi VPN gateway / print server.

    It’s a pity they chose 2G support as that is being phased out in most places but it was obviously a cost consideration. We (in my country) have only one network provider that still supports 2G and they will be shutting that down in July. So 2G modules are dirt cheap right now.

    For places where there is still 2G support this is the best IoT solution. 2G is much less power hungry than the higher data bandwidth modes.

    The NAND FLASH is a definite bonus – it’s wont suffer the corrupted boot SD card like the Raspberry Pi series.

    It’s low on RAM but hay! it’s $20. Just don’t try to do too much on one card. Buy an extra and split functionality between cards.

    I am looking forward to a version with 3G or LTE (4G) support and then it will truly be an IoT card. Ultimately we will have Narrow Band LTE (on 4G) which is designed to be low data rates and low power especially for IoT but don’t hold your breath for that as the manufactures are still trying to agree on a standard. It could be a year or more in leading countries.

    If you want to do something fancy or out of the ordinary then check driver availability first.

    And remember,, the “S” in IoT is for security.

    1. I completely agree. I can’t find Pi Zeros of either variety here. The reasonably priced units are all sold out within minutes of receiving a shipment. The ones available are all parallel imported units going for 4x the normal price or more. The places from where you can ship overseas all restrict to 1 unit per customer. That doesn’t help me when I’m trying to buy for meetup members so we can all have the same thing at the same time. I contacted RPi about this and they refuse to sell to anyone that they aren’t already selling to. I contacted them long before they began selling here and they didn’t reply at that time. Why they don’t ramp up the production to meet the demand, I’ll never know. They’d make a higher profit since they’d be dealing with higher quantities, but they’re a non-profit, so maybe they don’t care that much. I’ll start avoiding RPi for anything but personal projects. This might be a good alternative. RPi may see their market share decrease if they keep up their current business model.

  4. Orange Pi’s rock — I will definitely be rushing out to get a few of these. I have no intent of doing anything with the 2G.
    But an A5 chip and 1/4 G of ram allows a person to do a lot. The NAND is a very nice plus.

  5. Pros:
    – you can buy as many as you want, compared to the Zero W
    – lower profile than previous boards
    – price is right
    – external antenna (wifi and BT or just wifi?)
    – NAND boot – if you are thrifty you don’t need SD card.
    – battery power option?
    Cons:
    – dismal software support
    – virtually no community or company support
    – potential HW bugs (some old boards had issues with power regulator)
    – 2G – why oh why?
    – no reset pin
    – no HDMI

      1. Normally, I’d agree. As a matter of fact, even in this case I agree. If battery operation is the intent, though, the extra current draw may be undesirable. I can understand why it might be considered a con to some people.

    1. It most likely will, its just a matter of finding the documentation.

      Most of these chips (‘radio’ (gsm) wifi bluetooth etc) are just 1 general chip for all purposes, your average wifi chip in any phone will be more then happy to act as a accespoint, support all sorts of flavors of protection, bridge, etc, its just that most manufacturers dont implement all these features because they are of no use too the average joe.

      And in all the security agency leaks from the last year or 2 there’s a bunch of clear signs of using hacked cellphones and diagnostics hardware to put up fake ‘cell towers’, so 95% sure your average gsm chip will be happy to act as a cell tower and/or allow a direct connection.

  6. After bad experiences with an Odroid-W and -C0 I’m not touching any of those wanna-be clones anymore. Spec-wise they look better and have some nice ideas the RPi foundation should consider (e.g. battery management or audio out on the RPi Zero), but software does not work or stops working, community support sucks, updates never come.
    You might make the initial vendor-supplied image work, but beyond that it’s a minefield. I’m not going there anymore…

    1. “might make the initial vendor-supplied image work” – but there is no guarantee for this. It somehow worked, but i could not turn of HDCP, so my OrangePi-PC did not work with a DVI-to-HDMI cable. But luckily Armbian solved this problem.

  7. got bothered by all those chinese boards that are aggressive about price but totally crap about software support.
    I got crazy for weeks to make them working as i wish.

    HEY, CHINESE MANUFACTURERS: price your stuffs +3 USD and give us a fu*king updated distro image with all drivers (i2c, spi, gpio, etc) up and running!

    1. Yeah too many great budget hardware choices, too little good coding.
      It’s like looking at the software of Samsung devices – all great TVs, but their firmware and general coding sucks.
      Chinese suppliers are relying on the community to do it RaspStyle, but there is a limited supply of coders willing to put in time on yet another bizarre hardware…

    2. > HEY, CHINESE MANUFACTURERS: price your stuffs +3 USD and give us a fu*king updated distro image with all drivers (i2c, spi, gpio, etc) up and running!

      It’ll never happen. Because every time one of these is released with new features and a new lower price, all the tech blogs go ape over it. If you want an example, google “VoCore2 Lite” and all the articles about the “$4 Linux computer” and the people behind it won’t even mass produce it. It’s only available in sample quantities from IndieGoGo.

      My point is tech “journalism” copy+pasting the manufacturer’s press release when these boards are put for sale are a big part of the problem. They write the article to fill their quota, but never go in depth enough to realize the software support is terribad. Just look at the title of this article. The price is the *first* part of the title.

      So a manufacturer that adds $3USD to their price is just going to get a bunch of mentions about being “just like the XY board, but costs $13 instead of $10.” People being people will buy the cheaper but shit one, and then complain about the terrible software support.

    3. Clearly there is some reason for it; because it’s the case across the board; but it’s honestly a bit surprising: software is expensive to develop; but costs nothing to copy; and it has been clearly demonstrated that people will pay nontrivial premiums for antique Broadcom parts, ATmegas, ‘Nexus’ or ‘developer edition’ handsets just because they know that the software support will be better.

      If I were a generic, fabless, producer of mostly ARM reference designs; I’d think that the best way to improve my(likely razor-thin) margins would be to be as cooperative as possible with the mainline kernel; and do whatever combination of internal software development and actually providing datasheets is necessary to be thought of as something other than a disaster on the software side.

      It’s not as though rPis sell because Broadcom rocks at SoCs; Allwinner, Mediatek, AMLogic, etc. crush them on specs; but that doesn’t matter too much if it’s a nightmare hassle to get the thing working; and you have to assume that you’ll be stuck on some antique kernel version forever.

  8. My problem with them is the bulky ports they put on these boards. Sure they are small, but they are not thin. I would love to have just one Pi competitor bring out an arm board as thin as the Pi zero, and a user programable m0+ coprocessor and/or an fpga would be a much nicer feature than 2G is. A Zynq with a Samd21 and wifi/bluetooth would be perfect. It doesn’t have to be stupid cheap either, your paying $20+ for a zero after shipping them one at a time anyways. The closest thing is the parella board, but it has it’s own problems.

    1. Availability isn’t even their biggest problem. The boneheaded CHROME PLUGIN they require as a flasher that only works 10% of the time, and their abysmal forum support are much bigger problems.

      I also find it strange that they are constantly having supply issues, when their FAQ says the reason why they are able to offer it for such low cost is because they manufactured boatloads of quantity. Yet they keep having supply issues…. some shady stuff right there. I’m certain the entire thing is a scam that will implode in a couple years time if not sooner.

    2. CHIP bucked the xPi orphan board trend by hiring the very professional Free Electrons to do a great mainlining and software support job…even 3D is now supported. And no SD card issues, great power management too, plus LiPo support.

      Don’t know about the GR8 upgrade but given the excellent track record thus far, I am betting CHIP will remain the price- performance winner. I just wish they could sell a quadcore version for +$2.

      The Orange-pi-zero, much hyped ( again, here) still doesn’t have working WiFi and Armbian now craftily rebrands it’s WiFi as “low end, new driver coming soon.’ Coming soon ;)

        1. @Greg Miller

          My OPi0 remark is recently an Armbian Dev’s comment on cnxsoft on the same topic a few days ago. The WiFi driver for the obscure XR819 chip has been in and out of working mode. You got lucky.

  9. Even in places where 2G networks still operate, they’re not viable for new deployments, unless you really think so poorly of your work that you already plan to replace it within a few months. When I automate something, I want it fire-and-forget for the better part of a decade, and I can’t imagine investing the effort in a new platform if it’s obsolete before it’s even released. That’s just insane.

    The other specs are decent, so maybe if we ignore the 2G radio entirely (can it be powered off?), it’s still a worthy board, just bulkier than necessary.

  10. Hmm interesting. I currently use a pi zero with a 2G to serial module soldered to it at work as an IRC bridge. As they block IRC there ( i suppose they expect us to work lol). But seriously I like to keep an eye on IRC.

    This might be a nice solution. Will order a couple and see if i get the software to work. Too bad the software support sounds to be pretty bad.

  11. Am I the only one to notice that this Orange Pi has no video output? It looks like there’s no GPU in the “RDA”-branded chip.
    Since I don’t need 2G, I’ll stick with plain Allwinner-based SBCs. I have one as a Kodi media center and it works (with external WiFi dongle and microcontroller for IR remote, GPIO and CEC not supported, but at least it works).

    1. I have read, that there are issues with Kodi and HW video acceleration support on Allwinner devices, partly because of the GPL things and partly because of too little support/documentation from the manufacturer. This was when I looked for Kodi support of my OrangePi PC. For me “Media center” includes 1080p video playback, so this is important. Are these troubles already sorted out?

      1. @Martin

        FWIW, the $9 CHIP seems to run Kodi and has had 3D since November. Search getchip.com forums.

        Reason is their foresight to engage Free Electrons professional team to mainline and support the CHIP. Currently it is being upgraded from R8 Soc to their own GR8 chip.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s