Hi,
Does the Rock 64 support Bluetooth BLE on Android? In some places I can see a reference to a USB with Bluetooth 4 but I can't find it in the store.
Thanks

   


The new Rock64 is running sweet !  

... just doing burn-in now;  got some python procs running;  ssh terminal stuff;  network and serial console right now, very pleased -- second Rock64 and many more to come.


Note:   very interesting;  unlike my pre release board, this production board (running the same image) is running much cooler;  the cpu|gpu is cool to the touch, and the temps idling are in the high 30s ℃  ;  loading the system with my PI routines drove the temps up to high 40s ℃ , and this without even a passive heatsink !

Yes!! I could use the RDP Desktop with Xfreerdp !!

I f you need installing Xfreerdp, please feel free to post here.

OK, my question comes from being very ignorant of Linux and the limits of the Pinebook. My expectation is to use the machine mostly to do writing via Google's Sheets and Docs. Given the availability of largish SD micros and USB sticks, why would I want to upgrade from the 16 to 64 GB eMMC? Would it be like buying a Jeep to drive to work on the freeway and dream of driving over the cars in front of me when traffic stops, but never try?

first of all: i think there should be no throttling on any cpu under full load when testing it to the edge. (never had that my any laptop to be honest).
the pinebook is simply no exception in my opinion, but it throttles under certain loads due to heat.
hmm...
..
..how to tackle this?
EXTRA COOLING!

note: i won't say the pinebook is bad because of the throttling, i just want to mention this could be improved for a tiny bit of money (china is way cheaper than the eu anyway regarding hardware)

since i haven't received my pinebook yet.
but already know some details on how tight extra cooling can be to limit throttling, i thought of a bare bones cooling solution.

i looked at a small heatpipe, a small blower style fan, and copper film, all either too expensive or complicated for my use.
so i had to rethink my plan basically, and came across something very clever.

a material that is basically in almost every home, transfers heat as good as a heatsink (when applied correctly)
i'm talking about aluminum foil (tin foil called in america) which is practically a good heat spreader (because it's aluminum) and is thin, ideal for putting it between the case and the existing pinebook heatspreader

it won't be hard to make it work, just hard to apply (need to keep wrinkles down to a minimum on the SOC area)

a few layers over the soc extending over the battery area and towards the trackpad area should work nicely.
since most hardware is not within the area to cause damage, it should give almost zero risk of a short circuit (when applies correctly).

i'm planning to put in on the main board/keyboard/battery part of the laptop using some artic silver thermal paste i have, with a 1.6mm fully copper clad pcb i have laying around on top.
purely to let it fress nicely against the existing heatspreader when the bottom plate is screwed back on.
all will be secured with the existing case screws.

in total i plan on using 5 layers of aluminum foil of about 20cm x 10cm
costs: less than 2€ in the NL for a whole roll of aluminum (tin) foil of 3000cm x 30cm (30m x 0.3m), some thermal paste, plus a spare piece of (un)etched pcb, shim, spacer or similar (1.6mm x 30mm x 30mm).
imagine if this were done in china, it would cost around 1/5th of the €2 (0,40€), maybe even less

small lessons for this include: extra cooling isn't hard to accomplish, and is VERY cheap if you think clever.
this can be done in the factory as well (a bit more professional by sticking it all to the bottom plate), to prevent thermal throttling at those SOC intensive tasks and extend the lifespan of the device.
(we all know heat degrades devices over time)

i will let you guys know how it went (once i get my pinebook) and post the following thermals: SOC + PMIC.
all tested under these conditions while fully charged, at 50% backlight:
idle (cold start, booted into OS, idle for at least 5 minutes)
10min 4 thread stress test (same requirements as the idle test before this starts)
20min 4 thread stress test (same requirements as the idle test before this starts)
10 min watching youtube at 360p on firefox (same requirements as the idle test before this starts)

all under armbian 9 at room remp of 20 degrees C.

maybe in the future i will repeat the tests at 25 degrees C sometime, not sure

for now, anyone knows a good package for stressing the whole CPU? (where the slight throttling occurs at?)

(this post is merely a global info and placeholder for the final results till i've done this mod, which will described in this post)

maybe pine64 should set a public development goal for new boards.
like $xxxx R&D for a new addon board or seperate dev board.
with a public counter on the website.
with each sold iten, let's say 2% of the paid amount goes to that R&D budget

and let users choose on the forum for which modules needs to be developed first (a thread with poll for example)
this way users get to have more interaction with what they want towards the dev teams and such
it's a win-win: users simply can vote for what they want the most.
and that'll be made first, where pine64 listens to the ideas and makes what the users really want, giving (possible) higher sale numbers.

note that this would be nice to have, and this could pine64 really competetive with other solutions where the ideas are only based on devs.

i am not trying to rework the pine64 business model, but just thinking out loud about a ideal way for customers and devs for pine64.

I've just quickly made the following script for python, only tested with python3, on the stock Mate pinebook install.
Not sure if other installs move around the xinput device for the touchpad (else replace the 8 with the right number)

This script disables the touchpad of the pinebook for 0.5 second whenever a key is pressed, any key, including ctrl/shift/alt/mod.  So it's not perfect and cannot be use for shift/ctrl+click very effectively yet. But the script could be modified for that.

Note that if you google for the touchpad problem, you will find a lot of mentions for "syndaemon", which doesn't work (and exits without a message) The "synclient" command however will tell you why, because our touchpad isn't showing up as a touchpad but as an emulated mouse.
Hench this hacky script.

And I've just typed this post without a single problem while the script is running. Figured I share this ASAP, as the very sensitive touch pad is at the top of the list of things that people find annoying.

I just received a new 14 inch Pinebook and find on many counts it is as good or better than I had expected. But one area that doesn't pass muster is the power supply. First, the 5V connector on the wall-wart cable is fragile. Mine was broken on arrival and failed entirely the second day - the design of this connector is an invitation to failure. Pine Support consists of telling me to order another supply.

I tried find a replacement. The only bricks that supply 5V at enough current are almost as expensive as the Pinebook, with no guarantee they would work with the computer. As for a replacement for the connector is hard to find.

As an engineer who has worked on the design of electronic products I can readily understand needing to keep Pinbook costs contained and internal power dissipation down, but a 19V SMP regulator wouldn't have dumped any more heat inside the box than a 5V and there are a lot of external power supplies made for various laptop and notebook hardware that would have been just as effective and just as cheap. It is a long established truism that power should be transmitted at high voltages and stepped down at the application -- 5V in was simply a bad mistake born of tunnel vision from the SBC Pine64.

Returning to the support part of this post, replacing a DOA supply should be standard policy.

this is mostly aimed for development of the pinebook, since the pinebook is upgradable by swapping mainboard/emmc/speakers and such.

are there any intentions of designing a faster mainboard? (as in: faster cpu, more ram?)

i think as a company you'lll need to invest in Research & Development as well, and not fully rely on the userbase (counts both for hardware and software).

i personally would love something like a higher clocked cpu (like a RK3288) with 4gb of ram (3gb will be a nice upgrade even).

yes i prefer more speed over more cores here, as i want to use my pinebook for compiling and emulation in the future.

i know this comes with extra heat and power draw, the battery is capable of that and i want to look into an effective way of cooling my current pinebook (as stated in another thread by me)

now my question to the community: what are your upgrade ideas?

my own wishlist:
higher clock cpu (1.6GHZ (or higher) quad core)
more ram (3 or 4 GB)
more storage options (for 128GB)
better charge options (12V, 2A would be nice and more compatible with universal charger blocks for on your desk)

nice to have extra's:
higher speed ram (LPDDR4?)
bigger capacity battery (perhaps multi cell, like 4x3500mah @ 3.7V?)
extra internal usb or 5VDC + 3.3V UART port (example: a MCU with some sensors to adjust a fan (as heat exhaust?) or backlight automatically)
extra 5V internal solderable header (5V class D amplifier)
i know the following will require quite some money to realize, but a standard SATA port (bootable) would be really nice to fir in a half slim sata drive, else i'd like to suggest EMMC 5.1 or UFS 2

this is more a global question.

for me personally it's the upgradability, swapability of EMMC and price that attracted me.