This forum uses cookies
This forum makes use of cookies to store your login information if you are registered, and your last visit if you are not. Cookies are small text documents stored on your computer; the cookies set by this forum can only be used on this website and pose no security risk. Cookies on this forum also track the specific topics you have read and when you last read them. Please confirm whether you accept or reject these cookies being set.

A cookie will be stored in your browser regardless of choice to prevent you being asked this question again. You will be able to change your cookie settings at any time using the link in the footer.

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
A cool product with a seeming lack of commitment.
#11
Have you heard the saying horses for courses, I really got into the raspberry pi, thought it would suit me until I met foundation members on the forum, I found their staff so obnoxious that I looked for a raspberry pi replacement, I discovered that the foundation employees had upset so many people that companies thought it worth while to make raspberry pi clones, but you have found out like I have they are not real clones, things would go better for raspberry pi if they removed the owners and employees from the forum and didn't allow them to have contact with the public, where as big companies like ASUS have professional public faces unlike the obnoxious and unprofessional face the pi foundation have, so would have been great if the ASUS tinker was a drop in clone for the raspberry pi but it is not, it is however a fairly good server, I have one running on http://81.142.92.201/slclubs/ it has a 500gb hard drive connected and is powered by a raspberry pi 2amp power supply, been running for a week now with no issues, http://81.142.92.202/slclubs/ is a raspberry pi 3 with a 1 tb hard drive and a 2 amp power supply and has been running for a year, http://81.142.92.203/slclubs/ is a Odroid C2 with a 500gb hard drive been running for a week, I run all 3 headless using xrdp and remote desktop, they work fine, as for running the GPIO well raspberry pi is still the leader in that, both the tinker and Odroid have problems running the i2c stuff and the GPIO compared to the raspberry pi, so if you want a SBC to run the GPIO then use a raspberry pi and not the Odroid or tinker, but if you want a fast server then use the tinker or Odroid, it is horses for courses
Reply
#12
I will say, I went through some frustration with getting Up Boards working, but it never seemed like I was heading to a dead-end like I was with the Tinkerboard. I've been very pleased with them, but they are very different things than the ARM base boards. They are RPi formfactor boards that function like, and have features like industrial SBCs. They'll run pretty much any intel OS, even Android (though I never really tried). I'm sure you could even install MacOS on them with bootcamp if you really wanted to, just don't expect I2C, SPI or GPIO to work in any OS where this is not a common thing. But they run hot, really need good circulation, are power pigs (I run mine off of 6A/5V TDK/Lambda industrial din rail power supplies). The UP2 (Up squared) is an extended form factor version (Basically the RPis largest dimension squared, hence the name) with 2 GigEs, an M.2, a mPCI slot, a SATA port, multiple USB3 ports, 64GB eMMC and an i5 processor with up to 8GB of RAM, but in it's maxed out configuration it's $200. Cheap for what it is, but not in the RPi playground at all, even the original Up boards are $150. They are a complete opposite deal, in that OSs just run on them because they use all the same stuff any motherboard does and it's the typical Raspberry Pi-isms that are the foreign things to them, like I2C, SPI and GPIO, although these are also things native to Intel industrial boards since way before there was Pis.

The issue with the Tinkerboard is really that they deviated, just enough from the Rapberry Pi recipe, which is really in this space the reference for how this should work, so that things from that reference don't work as expected. So what you end up with is a board that looks like a Pi, has all the connections of a Pi, but is really no more like a Pi than any other board that is just a similar idea. I also didn't really understand why the Tinkerboard WiFi needed it's own distinct Rockchip driver hooks, when it was a separate RealTek chip that already has a driver in the main ARM distro. A driver BTW that has to be modularized into the kernel and connects to some Rockchip hook that can't be modularized. Pretty much if you changed any aspect of the Rockchip drivers in the kernel the kernel build blows up. Believe me I tried just about every logical possibility. The only Rockchip thing I was able to get to compile as a module that wasnt' when I started was the SPI. In order to make these deviations work as expected, especially considering they added piggybacked features onto the 40 pin bus that the reference (the Pi) doesn't have, is to release some method of doing so. Short of that it's just a 40 pin generic header that has very little to do with the Pi and it doesn't matter that it followed a similar layout.

What you end up with then is a kinda cool board, and the beginnings of an OS with chip integrations that not many people use in a way nobody else uses, oh and no documentation on how they really used them. Even the GPIO mappings are only known, if you happen to look in the home directory of the default linaro user and figure out that there is a port of the wiringpi package there that once installed can output the mapping, so long as you have an actual Pi with the same package installed to compare it against. I scoured the web, and found no document anywhere that showed those mappings, just a mention somewhere of the existence of that package in the linaro home directory, and it's the only reason I got as far as I did with the board, which for my project was still nowhere. I know of no other board that is as poorly documented as the Tinkerboard, there simply is just no real starting point. It just ends up being an ARM based SBC with a bunch of features you can't really use unless you're willing to spend the time to reverse engineer and/or re-engineer it. I probably could have gotten the board to do what I wanted it to do, but probably not without hacking kernel code, application code, and/or physically butchering peripherals so that they could get the power they needed and be re-addressed to less conflicted I/O pins. At that point it becomes something I'd have to rehack every time I wanted change something. I don't care about it being a custom project, but it gets to the point where it's an unserviceable project because you've had to change so many things you have to then create your own personal wiki just to be able to reproduce what you did.

Yes, I could have probably figured out how to fix the WiFi issue, yes I probably could have powered the package through the GPIO header with a power supply that had lots of excess capacity and could dial in the voltage, yes I could have modified the AdaFruit HAT to remap the GPIO pins it used to less conflicted ones by changing the socket connector on it to a pin connector and using a custom ribbon cable, I probably could have hacked some kernel code to get some stuff to work that doesn't, but at that point haven't I just turned a toaster into a coffee maker? At that point it's become something very much *not* a more capable plugin Pi replacement, and at that point it offers very few advantages over the way more capable and way better documented Up Board which I am only not using for this because I can't get all the capabilities I want into the size package I want with a low enough heat signature to be passively cooled and *not* require a specialized power source. Easier, cheaper, faster, pick 2 as the old saying goes, except the Tinkerboard only solves one, and it's the one that for me is the least important, and the two it isn't are pretty much unknowns to their extent, it could cost me double in do dads by the time I'm done, it could take me a month or more in time by the time it's done, and then I have to replicate that effort 3 more times and that will require more than just copying the SD card a few times. I'd rather be wishing for a little more speed by next weekend than still trying to get the first one working by Labor Day.
Reply
#13
Kinda agree.  


What you need is guides about peripherals i think.
Undecided They didn't provide enough documents about peripherals officially for tinker board user.  
The infomations about SOC is easy to get, you can get them form rockchip opensource website and experienced community guys, but most of users are concerned about peripherals.


If they don't plan to provide much guides and documents about peripherals, at least they should provide a working mainline kernel using the right driver(It's not diffcult for Rockchip SOC), so people could get help from the other community to make their peripherals work.


Big Grin Anyway, it's just a new board.I think it will be improved in the future.
Reply
#14
(06-07-2017, 03:06 AM)wzyy2 Wrote: Kinda agree.  


What you need is guides about peripherals i think.
Undecided They didn't provide enough documents about peripherals officially for tinker board user.  
The infomations about SOC is easy to get, you can get them form rockchip opensource website and experienced community guys, but most of users are concerned about peripherals.


If they don't plan to provide much guides and documents about peripherals, at least they should provide a working mainline kernel using the right driver(It's not diffcult for Rockchip SOC), so people could get help from the other community to make their peripherals work.


Big Grin Anyway, it's just a new board.I think it will be improved in the future.

Well, hate to keep using the Up Board as an example but there was more information about the Up Board when it was still on kickstarter, before the first board ever shipped, then there is for the Tinkerboard now. I don't really see this as an excuse. Also the Up Board has had a stable distro from the start, a supported Forum and Wiki with questions regularly answered from Product and Project Managers and Engineers, and remember the Up Board comes from a group within AAEON (ASUSs industrial division). It even has a direct store with tested option accessories available, and 3D printable case designs. I am using Pi products that were not on the tested list and just worked on the Up (and don't on the Tinker).

I spent less time getting WiFi working on the Up Boards than on the Tinkerboard, which still never worked, and it isn't a built in feature of the Up Board. I have AdaFruit PiTFT hats running on them, and the modules were already compiled into the kernel. I have relay boards working and SSDs running off of USB ports, and these aren't different systems, all of my Up systems have -all- these things running on them.

The Up organization is working on it's third board platform in just about a year and every one of these boards has orderable optional configurations, down to different eMMC, Memory and even CPU configurations. The current kernel sources were made available within 24 hours after the first person to ask about them on the official forum, the differences in SPI, I2S and DMA stuctures are documented as are the minimum power requirements and voltage thresholds. Even the manipulatable structures within sysfs filesystem are documented, there is a wiki section on Raspberry Pi HATs and products that work with it, and this was there way earlier in the product lifecycle than we are now with the Tinkerboard, and remember this is an Intel based board with few architectural similarities to the Pi, yet it plays better in the Pi playground than a board that is, in what little information there is on the official site, directly compared to the Pi as *if* it is a drop in replacement.

...And the Up Board is a board, sorry now a series of boards, that most people aren't even aware exists, oh and made by the same maker as the Tinkerboard (the kicker). I don't expect the Tinkerboard to be the same as a crowd funded one off, which by the way most of the crowd funded one offs are no longer one offs but multi-model product lines now, I expect the --ASUS-- Tinkerboard to be better than that, better than most, if not all, and what it is instead is like a crowd funded project put out by an organization with a bad case of ADD, it is a gleaming example of how not to do a product like this.

Seems it was designed to be able to do all sorts of things with very little effort to make it actually do any of those things, and no documentation to allow the end user to make it do any of those things either. Is it really realistic that in order to get this board to do the things it was designed to do, the end user needs to start by figuring out that you have to use an unreferenced procedure from a Raspberry Pi to flash and boot the board? ...and then dig through unreferenced data sheets for the SOC it used and try and guess how they implemented this to create this board? Really? Let me ask that again, really?? From what I've been able to find there is actually very little difference between the Tinkerboard version of Linaro Debian and the MiQi version which it seems to be based on, a board not backed by a major maker, with a target price half the Tinkerboards, didn't make it's goals in crowd funding and somehow serves as the basis of TinkerOS? What did I say before? Oh right...really?
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)