A Router-Based Dev Board That Isn't a Router

An anonymous reader writes with a link to an intriguing device highlighted at Hackaday (it's an Indiegogo project, too, if it excites you $90 worth, and seems well on its way to meeting its modest goal): The DPT Board is something that may be of interest to anyone looking to hack up a router for their own connected project or IoT implementation: hardware based on a fairly standard router, loaded up with OpenWRT, with a ton of I/O to connect to anything.

It's called the DPT Board, and it's basically an hugely improved version of the off-the-shelf routers you can pick up through the usual channels. On board are 20 GPIOs, USB host, 16MB Flash, 64MB RAM, two Ethernet ports, on-board 802.11n and a USB host port. This small system on board is pre-installed with OpenWRT, making it relatively easy to connect this small router-like device to LED strips, sensors, or whatever other project you have in mind.

Comments

Scamaday

[Anonymous Coward]

Chinese rip off of http://8devices.com/carambola-2 but of course more expensive because kickscammer or indiescam

Re:Scamaday

[BronsCon]

Actually... less than half the price and includes a few bits the Carambola lacks.

Re:Scamaday

[Anonymous Coward]

Not half the price, but actually 10 bucks less, and the only added features are a Darlington array and some optos, but much bigger and with suboptimal non-factory calibrated RF.

Re:Scamaday

[BronsCon]

Ahh yes, you are correct, it's only about 10 bucks cheaper; I was looking at the module price, not the price of the whole package. And much larger, but also supporting a wider range of logic levels and providing better overvoltage protection on its inputs. Also, what gives you the idea that the DPT's RF isn't calibrated?

Face it, your argument boils down to "it costs less and has more features". The only potential downside is for projects relying on "perfect" RF, assuming you are correct about the DPT-board's RF not being factory calibrated; for prototyping, or projects that will primarily make use of the ethernet ports (or not rely on networking at all), it's still a less expensive and better-featured option than the Carambola. If you're far-enough along in your design process that size has become a consideration, you're probably past the point of needing physical ports on your development board, which makes the DPT module (the blue board) about $24 cheaper and a better option; if your project outgrows that, you're far enough along that you're having custom boards fabricated and neither product is an option.

This has its place, especially given the software suite it will ship with. That place is early-stage prototyping and proof-of-concept development, where silly mistakes like overvolting an input or connecting to the wrong pin, which would kill the Carambola, are most likely.

Re:Scamaday

[TheRealMindChild]

If you require "perfect RF", you are doing something wrong

Re:Scamaday

[BronsCon]

I think my point was made for anyone who wasn't insistent on being pedantic...

At least one order of magnitude too expensive

[Anonymous Coward]

You used the IoT buzzword. That limits your price range to where you'll never order just one because shipping costs more than the device. Besides, almost $100 for 64MB RAM and 16MB flash is ridiculous and not at all an improvement, much less "hugely improved", compared to off-the-shelf routers sold for less.

Re:At least one order of magnitude too expensive

[Anonymous Coward]

Yeah, would be interesting at $20

Re:At least one order of magnitude too expensive

[Anonymous Coward]

The blurb says $90. Anyway, $35 is still too expensive, given the specs.

At least one order of magnitude too expensive

[sensationull]

Its $35 RTFA

Re:A ton of BS

[Anonymous Coward]

I believe you're missing the fact it contains 20 GPIO pins; See if you're COTS router gives you that level of flexibility.

Re:A ton of BS

[Mister Liberty]

Yes, I've got the impression too that someone succeeded in getting their pet project overly hyped.

Re:A ton of BS

[Anonymous Coward]

On board are 20 GPIOs, USB host, 16MB Flash, 64MB RAM, two Ethernet ports, on-board 802.11n and a USB host port.

I think they are referring more to the GPIOs than ethernet or USB ports when saying "with a ton of I/O to connect to anything".

Raspberry Pi

[Anonymous Coward]

Without RTFS-ing even - I barely skimmed TFS and some comments - is there anything this can do that my Raspberry Pi cannot? I mean, judging by the comments, even if I have to buy extra gear for my Rpi to match features I'd still be saving money.

Use it for a computer

[inode_buddha]

First thing I would do is hook up a screen, keyboard, and pointing device and use it for a computer. Imagine.. a Beowulf cluster of these!

It's a "flexible scamming" project.

[Animats]

It's one of those "flexible funding" Indiegogo projects, where they get to keep the money even if they don't get enough money to make anything. Great scam; just come up with some popular idea, overprice the project, and keep the money.

Re:A ton of BS

[hawguy]

On board are 20 GPIOs, USB host, 16MB Flash, 64MB RAM, two Ethernet ports, on-board 802.11n and a USB host port.

I think they are referring more to the GPIOs than ethernet or USB ports when saying "with a ton of I/O to connect to anything".

I'm curious what people would want to use these GPIOs for on a router... does anyone have any real-world projects where they use them? Not just "It would be cool if it it did X", but actual real-world projects.

I'd rather have more ethernet ports on a router so I don't have to VLAN my network.

Re:A ton of BS

[BronsCon]

Well, you wouldn't necessarily use 20 GPIOs on a router... but, then, this isn't a router, just a dev board based on a SoC commonly used in routers, running on a software stack also commonly used in routers.

WRTnode

[complete loony]

A similar board, http://wrtnode.com/.

Wow you know slashdot is fucked

[Anonymous Coward]

When it has 1/3 the comments as a website as HackaDay and 99% of the comments on slashdot are factually incorrect or just plain stupid. What the fuck happened guys.

Re:A ton of BS

[Anonymous Coward]

ah yes, GPIO pins are crucial for routers.

USB Host?

[Anonymous Coward]

"You said USB Host twice."

"I like USB Host."

Thank you, Mel Brooks, for warping my mind.

Re:A ton of BS

[Anonymous Coward]

See if you're COTS router gives you that level of flexibility.

Obligatory response to emphasized portion: I am not a COTS router! I am a free man.

meh

[Osgeld]

A) I didnt know router hacking was still a thing
B) you can get a wrt compatible linksys any day of the week at goodwill for 4.99$ if you really want to dick with it
C) there's a thousand ways to make a custom router with already available and cheaper hardware from old pc's to the invasion of the crappy ARM sbc's (pi, beagle bone black etc)

Re:A ton of BS

[Anonymous Coward]

After getting tepidly optimistic about your comment:

I can buy a 16-port gigabit router on eBay for $50

I went and looked for one. Sure you can get an unmanaged, wired, switch that is not WRT (Wireless Receiver Transmitter) compatible for fifty bucks. If you want what you say, a managed, WRT compatible router with 16 gigabit ports it will cost a thousand. So in other words, not only what you said was not true, you don't even know enough to understand why it is not true. A ton of BS indeed. Hopefully no one pays you to do anything networking related.

IoT which chipsets?

[labnet]

We are about to embark on designing a similar product. Essentially we want a gateway that can select from either a Cell Phone embeded data modem such as the Telit HE910, or local WiFi, then provide an internal data link via USB or Ethernet to our device and some remote connectivity for setup via Bluetooth.
We need good power management, and the ability to add local peripherals (such as a keypad, status LEDs, etc)
OpenWRT looks like the right foundation, but which chipset to select is more difficult.
Any suggestions from hardware devs out there?

Re:A ton of BS

[sexconker]

Well, you wouldn't necessarily use 20 GPIOs on a router... but, then, this isn't a router, just a dev board based on a SoC commonly used in routers, running on a software stack also commonly used in routers.

This isn't a duck. It just walks like a duck and quacks like a duck.

Re:A ton of BS

[BronsCon]

Right. It doesn't swim in the lake like a duck.

Re:A ton of BS

[Sarten-X]

It's a $35 drop-in networking robot controller.

Re:A ton of BS

[BronsCon]

Bingo. Or a small and low-powered testing environment... or any other number of uses.

Why flash and not microSD?

[no_such_user]

Instead of flash memory soldered to the board, microSD is ubiquitous and cheap -- and makes the device effectively unbrickable. Sure, there are bootloaders with recovery features, but it's not as simple as writing a new image to SD. Raspberry Pi got it right in this department. It's a shame there's no PCIe bus on the raspi...

Re:Why flash and not microSD?

[petes_PoV]

MicroSD required mechanical connectors to the device and is a great deal more expensive that flash - in a very price-sensitive market. Given that an IoT thing could find its way into any environment, the last thing you want is for its operation to be dependent on the correct operation of nasty, cheap (and they *would* have to be cheap for comparable production costs) connectors and uSD cards of variable quality - that are outside your control.

Far better to have everything firmly and permanently attached to the board. Why solder in a connector whan it's just as easy (and takes the same amount of board space) to solder in flash instead. That way you don't get the blame when an idiot user "recycles" an old uSD card and blabs all over the internet how crap and unreliable your product is, as their card keeps corrupting.

RPi got it completely wrong in this respect. You don't hear of corrupted software & kernels on all the cards that use flash. If it's more "difficult" for noobs to use, then that's no bad thing either as it discourages those who are lacking in the clue department. This is not meant to be a plaything for children.

Re:Why flash and not microSD?

[Agripa]

Do SD cards implement background scrubbing? Using NAND Flash as a replacement for NOR Flash always strikes me as unreliable given the short storage time of high density NAND memory. I have already seen older consumer devices that use high density NAND Flash for firmware storage "self brick" after a period of time and I suspect this is what caused it.

Re:A ton of BS

[Anonymous Coward]

It would be nice if it could run Broadband-Hamnet...

2 ports?

[Anonymous Coward]

generally a 'router' has more than two ports...a two port router is a 'bump on the wire'

$90 was for two. One's $40-60

[billstewart]

It's $20 for the module, $35 for a development board with the module soldered on it (which you'd almost certainly want), $45 for that AC power, cables, $50 for the board plus a spare module and pre-installed software, about $5-8 for shipping.

GPIO's useful for Internet of Things

[billstewart]

Sure, if you're just routing, and don't want to connect to various hardware I/O things, you can get a simpler board. But if you want to talk to sensors or build yourself a toaster controller or weather station or add lots of blinky lights or whatever, they're useful.

Re:$90 was for two. One's $40-60

[Anonymous Coward]

So ... more expensive and less capable than a Raspberry Pi.

Got it.

Only 64MB? WTF Maybe if this was 2004...

Price is reasonable - $35, not $90

[billstewart]

It's $35 plus shipping for the development board with the module soldered on it, so it's about the same as an Arduino; the $89 price was for two of them plus accessories like cables and power supplies. They're asking for not very much money to finish their software development, and the real question is whether their software is any good.

Re:Price is reasonable - $35, not $90

[Imbrondir]

Yes thats a good price if he delivers the project. He is simply pointing out the great difference between Indiegogo and kickstarter, where in the former if you miss project target funding, you may still keep all funding without delivering anything to funders.

Re:Price is reasonable - $35, not $90

[cdrudge]

He is simply pointing out the great difference between Indiegogo and kickstarter, where in the former if you miss project target funding, you may still keep all funding without delivering anything to funders.

Or in the latter where if you make your project target funding, you may still keep all funding without delivery anything to funders.

Re:Price is reasonable - $35, not $90

[melstav]

That's why it's important to actually read what they wrote instead of just stopping at the first "red flag" you come to.

Why flexible funding? We choose flexible funding because we want to give people a chance to contribute to the software as early as possible. The hardware part is already done and we have sold units to existing customers who were very happy about it. Specially for this campaign we made a new revision ready for mass production so we can sell it at an even better price than we already had in our shop: https://dptechnics.com/shop/?q...

They already have finalized hardware in production. They're not trying to fund hardware development and production. They've already done that. They're using indiegogo as an advertising channel and as a secondary storefront.

Re: Yeah, 2 ports + WiFi - so?

[billstewart]

No, generally a router has an inside and an outside, and sometimes a third port as a DMZ; you're thinking of a router with an ethernet hub attached, like many home routers. There are routers with more routed ports, and there are one-armed routers also, though that's less likely to be useful.

Re:A ton of BS

[BronsCon]

The DPT board runs on a completely open stack and everything used by BH is open... so... it can. It doesn't yet, which is no surprise since the board is still under development and you're probably the only person that's voiced that request so far.

Others available from $10

[petes_PoV]

There are other people out there making similar devices.

One I have heard of is Olimex who reckon their product (still in design, with an RT5350F) will be 10 USD in 1,000 off quantities. Over time and with better integration of future devices we can safely assume that will halve.

WRTnode

[Anonymous Coward]

It was in preorder in september last year, with a promise of delivery in one and a half months.
Looks nice, but it'll probably cost far more.

Re:$90 was for two. One's $40-60

[drinkypoo]

You're still looking at a higher price tag than just buying a Mikrotik Routerboard used on eBay. If your goal is to get a router it's a goofy thing to buy.

me

[abdel hadi]

yes, you are correct,

Re:A ton of BS

[khellendros1984]

It sounds more like an IoT prototyping board. The RJ-45 and wireless are for the "Internet" part of that, and the USB and GPIO are the interfaces to the "Things".

Re:A ton of BS

[rpstrong]

And wooden you say: "I am not a lumber! I am a tree man."

Re: Yeah, 2 ports + WiFi - so?

[Anonymous Coward]

no, i was thinking about a router..you know, like a router, with tens or hundreds of ports