WRTnode Opened for $25(wrtnode.com) |
WRTnode Opened for $25(wrtnode.com) |
But a few niggles:
- Quoting RAM and flash in Mbits? Whilst that may be the norm for buying components, I'd argue most people familiar with OpenWrt and the Raspberry-Pi generation are going to miss this detail and think "Oh wow, 512MB RAM and 128MB flash!", when in fact it's 1/8th of that.
- There's some mention of a standard shield, but no information on what it includes or when it will be available.
- Like others have said, that scrolling on the website is extremely annoying.
Regardless, I will definitely be buying one anyway.
- OpenWrt's table of hardware page: http://wiki.openwrt.org/toh/start
- Raspberry Pi specs: http://www.raspberrypi.org/product/model-b/
- Minnowboard Max specs: http://www.minnowboard.org/meet-minnowboard-max/
- Beaglebone Black specs: http://beagleboard.org/black
Oh and I love the reference to TL-WR703N, that little thing was pretty special! I was always disappointed by the MR3020, which is basically identical but a lot larger for no obvious reason.
Once it exists, its almost like computorium, becoming the center of nearly every maker project and then staying there because its so cheap you don't have to choose which projects get your precious dev boards.
This is a fantastic step towards that goal.
To the "designers" :
Next time do a simple web page without stupid scrolling hacks,for god sake.I'm sick of it ,really. Dont mess up with basic usability just because it looks good on your Macintosh.I guess you didnt even bother testing on mobile.
Sorry for that rant,but we are in 2014,and I cant take more of that stuff.
A lot of really bad ideas are being used on web sites now to make them look sophisticated. Ironic considering all of the simple to use tools developers now have to measure conversion rates and user engagement.
This is what allows adruino to be open as they share layout files and software, but the Atmel AVR micro-controller they use does not have an open design.
It does get confusing since there are open hardware projects that go down to silicon, but for the most part that level of design is just beneath the concern of the people who want to internet enable their fridge or whatever. Just like most computers contain closed source code in the form of controllers and all but the most extreme open source types just don't care that they have a closed source USB controller.
Even if you can get original kernel source, it might not be trivial to get it to compile. Porting the required changes to a newer kernel once a security vulnerability was discovered can be very time consuming.
http://www.linuxx.eu/2014/08/banana-pi-raspberry-pi-upgraded...
according to one of WRTnode from http://openrouter.info/forum/viewtopic.php?p=36058
and WRTnode is about 20g in weight.
Same amount of RAM and flash, but different SOC.
SPI flash generally will do on the order of 0.5 MB/s erases, slightly faster writes, and somewhere in the realm of 4 to 12 MB/s reads, give or take. It's not high performance but it's simple, cheap, and low pin-count. The host controller is responsible for wear leveling, erasing each needed sector/block, and all writes, as SPI flash is unmanaged (compared to SD cards which are managed memory).
The chip is actually made for routers: it has 802.11n "MAC/BBP", 5-port 10/100 switch, etc. http://www.anz.ru/files/mediatek/MT7620_Datasheet.pdf I wonder how "Hardware NAT with IPv6 and 2 Gbps wired speed" works if the switch is only 100 Mbits...
I guess if what you want out of this is really a router, you are supposed to either build the rest of the hardware yourself or wait for yourself, or wait for the "motherboard" module for this.
From what I've seen, open source development that involves hardware typically involves reverse engineering specifications and/or emulation. Neither is practical for state-of-the-art.
Not to mention, MIMO is just a "wee" bit tougher than simpler schemes.
802.11ac WiFi, 5GHz, preferrably supporting a 4x4 MIMO
In the 2.4GHz band there is a limited amount of space available, and there are now many wireless routers in typical residential neighbourhoods all competing for the same space. Many other devices such as cordless phones, baby monitors, and wireless speakers also use 2.4GHz and interfere with wifi signals. Most microwaves generate significant amounts of noise in the 2.4 GHz band when in use.
In the 5GHz band there is much more space available, and there are many fewer competing devices and sources of interference. The penalty of 5GHz is slightly less range, and more expensive wifi equipment.
When I was taking Chinese classes in university, people would sometimes poke fun at the eccentric or badly accented English. I would as them how their Chinese sounds to their native-speaking instructors.
It is strange how the reaction to that was always one of confusion or dismissal.
Some manager who probably doesn't speak much English has tasked a random person on their staff who went to college in the US or UK with writing their English copy, without realizing the difference between "this person can communicate effectively 1:1 in English" and "this person can write formal English like a native speaker".
In other words, a task is being supervised by someone who doesn't understand the task. Just like every other company we deal with. As native (or at least highly fluent) speakers, we just happen to be competent to critique it.
However, I would not immediately characterise the copy as written by a US/UK college grad though - it feels like it is written by someone who learnt English on the mainland.
I don't really know why companies in various parts of East Asia don't hire/outsource to someone who actually is highly fluent in English to write/edit their copy. Sometimes I suspect it is simply due to cost, other times I suspect they just don't realise that their fluency is not quite.. 'international English' level
(No offence to anyone who runs/writes for East Asian company websites.)
When the peasants are revolting its very different from when the food is revolting!
You realize English should be capitalized, right?
Here i've cleaned up some of the copy from your sales page:
"WRTnode is a high performance, low power, small, inexpensive development board intended to run OpenWRT OS. It also easily ports open source software based on Linux.
Today, more and more devices are connecting to The Internet of Things. The WRTnode is a small node which can connect to the internet all by itself and perform complex tasks, like detecting a cat running by, or identifying your speech and twittering it, or checking your email and speaking it to you, or learning how your room is laid out and finding garbage to sweep while streaming video to you over the internet. [..] Open Source hardware for OpenWrt [..] [remove the line "smart machines' heart"] complete set of high performance I/O [..] Local dns uses i.wrtno.de, wrtnode.lan, and openwrt.lan [..] default WRTnode ip 192.168.8.1 [..] WRTnode additional features (source at github.com/WRTnode):"
In the future it shouldn't be too hard to find someone willing to support and a good open source project with some proof-reading. I may not be the best candidate, but it would be a good way for someone to get started helping the project, and could make a big difference if you are targeting a larger audience.
The ISA is just a small part of what's required. All the other interfaces are just as important such as pinouts, memory mappings, memory interface, bus support and so on.
There's very few SoCs from different vendors that are interchangeable even among the ones that uses an open ISA. This requires you to either build everyhing from scratch including the CPU, or to use more generic and perhaps interchangeable CPU or microcontrollers and rather push the special components to other chips instead of having those components on a SoC - which gains you nothing.
Meaning the value of using an open ISA is miniscule as long as the code is available and easily translated to another ISA by a compiler.
Their limitations are effectively hidden from them. Odd things still come out, but in informal contexts it's not worth correcting, and there's little opportunity for formal critique. (There's also the issue of "face" that complicates matters and drives me insane. It's easier with people who have spent a few years in the US, though.)
If there's no (near-)native speaker around to check, nobody's going to notice until it goes out to the world, and then there are a half billion people ready to jump on it.
If they recognized it, I don't think cost would be a factor. I could review/correct a site like this in an hour (or less). I might spend a day on the mini-manuals that come with some gadgets.
..."Born to be proud" [...] doesn't sound natural to a native English speaker...
It also has echoes of classic 1970s rock, which doesn't help (born to be wild, born to run).
I know it's made for routers, but most likely not for doing a lot in software. As you quote, it does hardware NAT. There's a reason they put that in.
There isn't a fundamental reason why the mips couldn't do the job, but running stock linux it's going to be thrashing its L1 cache, and it doesn't have an L2. Also, next time you play with one of those,check the hardware counters for how many instructions/sec it actually manages to process running flat out :-)
Bear in mind that it's probably a packet rate rather than a bit rate limitation. That really bit us in our application because we were trying to run many lines of voip, which means lots of small packets. - YMMV.
Doesn't destroy it, but it's certainly a black mark. Are 1gig chipsets that much more expensive than 100mbps these days?
Examples: network card; disk controller card; video card; SD card.
"Shield" is such an entrenched term in electronics, from long before computing, that it's completely stupid to try give it another meaning. It's also not from some specialized sub-domain whose rare terminology can be reused; shielding is employed in all kinds of devices.