mangeurdenuage (email@example.com)'s status on Monday, 20-Mar-2017 20:57:16 UTC
mangeurdenuage (firstname.lastname@example.org)'s status on Monday, 20-Mar-2017 20:57:16 UTC mangeurdenuage #eoma68 update:
You Can't Bring Electronics into China
Special update, jumping the queue slightly, which can be summarised as “You Can’t Bring Electronics Into China”.
So, in order to complete the board bring-up of the RK3288 PCB, get the DDR3 memory up and running, as well as ask the factory to help put the alternative (Toshiba 128mb) NAND onto the board, and so on and so forth, I packed every single EOMA68 board, including the prototype Micro-desktops, very carefully, along with my Firefly RK3288 board, in the middle of a huge suitcase so that they would be properly protected, along with three other Babushka-doll suitcases so that we can pack up here in Zhuhai and return to Taiwan.
Except that this time, unlike all previous times, probably because we were on a ferry that had far fewer passengers, going as we were direct from Taiwan to HK and HK to Zhuhai without actually leaving immigration, Hong Kong customs had plenty of time on their hands and this time noticed that there were PCBs in our luggage… and took them out. There is now something like $6,000 USD worth of prototype PCBs - all of them including the only two Ingenic jz4775 Cards, all of the EOMA68 RK3288 prototypes, the working EOMA68-A20 Revision 2.4.1 prototypes that I am supposed to be getting ready to send out to engineers, reviewers and a prospective large customer, sitting outside of immigration / customs in a special holding area in Hong Kong.
Now, luckily, my partner has a US passport along with a multi-entry visa. This is unlike myself and my daughter who only have dual-entry visas, one of which has been used up already (children are not issued with multi-entry visas in China even if they have a US passport). So what my partner is going to have to do is travel to Hong Kong, meet a friend whom we will pay to hand-carry the $6,000 USD worth of Cards, representing the entire investment of time and effort in the EOMA68 project to date, on a flight back to Taiwan next week, then she will return here to Zhuhai so we can continue packing.
The moral of the story is: for god’s sake don’t put all your prototype PCBs into one single suitcase coming into China. It’s now going to cost somewhere close to $600 USD just to get them back. Normally it would be okay to use SF Express, for example, but I can’t possibly trust a standard courier with such a strategically valuable package.
I have to say I’m getting really fed up with the deliberate barriers that China erects to innovation. I’m running OfflineIMAP again so that if there’s yet another global DDOS attack using millions of IoT devices with passwords “admin, admin,” I can at least get email in and out of this country (just excruciatingly slowly). The visa situation is ridiculous. Internet speed is at levels we saw in Europe back around 1998 (ISDN, if anyone remembers that). They’re making it really rather difficult for me to give one of their citizens over a MILLION Yuan for helping to get you your pledged items.
So, anyway, whilst I am here it looks like what I will be doing is quite a lot of reverse-engineering of Rockchip’s proprietary 32_LPDDR2_200MHz_LPDDR3_200MHz_DDR3_200MHz_20150318.bin file - the one that I know can properly initialise the EOMA68-RK3288’s DDR3 RAM. There’s a key difference between this proprietary file and what’s in u-boot-rockchip, which is stopping 4 GB RAM ICs from being initialised. I will have to think over the next few days if it’s worth still traveling to Shenzhen to visit the factory, or whether to just ask them to ship whatever prototypes they are preparing (EOMA68-A20 revision 2.7.3 and Micro-desktop revision 1.7) directly to the apartment, here in Zhuhai.
Ah, what fun it is to have to adapt to circumstances…
Update: Our luggage arrived this evening. Only one of the boxes had been removed. The “tin box” containing the RK3288 Firefly and two Ingenic jz4775 EOMA68 Cards had not been removed, presumably as it had been missed by the X-Ray. Whilst I am here I can at least therefore do interactive exploration (binary assembly-level patching of the rockchip DDR3 initialiser to get it to print out debug messages). I’m really looking forward to returning to Taiwan.