Seriously Disappointed

[peterhowells]

After now having received my Oak board 2 weeks ago, I realise that I amy as well relegate them to join the DigiSparks in the bottom of my drawer. The DigiSparks were a complete failure not being able to be programmed with any certainty. Now of the 2 Oak boards that I ordered only one is working - the other sits on my desk with the power light on not being able to do anything else. Not even a factory reset has helped.
I think the road of supporting Digistump has come to a very disappointing end for me.

[defragster]

@Peter - indeed it is way more effort to start than it should be - I had to setup the LOCAL server to get my 6 to easily program.  The hardware I got works - there is just a big HUMP to get started unfortunately - i.e. the first UPDATE and Particle registration.

ONCE done however they program easily and well for me much more powerful than the DigiSpark - I just wrote a sketch to use SPIFFS 'disk area' that worked first time (and a few updates to get code worth posting).  After my first blink sketch I set mine aside for other things.  But wanted to show SPIFFS to work for others and the 1MB of File System that gives can really add to the utility of the OAK. I powered it up and hooked a serial adapter on to watch the debug spew and it was working!

[PeterF]

@peterhowells Sorry to hear you think that, but if its not working for you, its not working for you.  The Oak has been plagued with some issues with the reliability of the initial update code and the plethora of wifi configurations and chipsets, but once you get past that initial update, it is pretty bulletproof. You may have the occasional firmware upload fail, but it isn't disastrous - it usually reverts to the previously running code, so you just do the upload again. If you upload code that manages to b0rk your  Oak, you just power cycle it with P1 GNDed to enter safe mode and upload some less crash-inducing code. Other than that, it's just a matter of waiting for the code base to stabilise and more example stuff to come out showing what you can do with this board. There is plenty of stuff you can pull in from the ESP8266 Arduino example codebase to get you started also. If you don't have one, grab yourself a cheap 3V3 CH340 USB to Serial adapter also, as it is much easier to debug what is going on using local serial output.

I have to say, after coming from some of the other ESP8266 boards, that the OTA particle support has been the real plus (although I see the Arduino ESP8266 core now support OTA updates), as well as modules that have better antenna and seem more stable power and run-wise. I had no end of issues with the 8-pin ESP01s crashing and burning them selves out (deep sleep reset b0ork or plain 'ol lock -> SPLAT!), and have just one that has been running 24x7 for about 4 months now without a glitch. The ESP12 board I have is nice for development and is stable, but is over twice the size of the oak. The oak fits in perfectly as the middle ground, and is well suited to lower power applications for the ESP8266 CPU.

As far as the digispark, what problems were you having with them? I have both the original sparks and the successor, the pro, and haven't had any issues with them other than the odd upload fail, which was usually due to dodgy drivers or cables. Actually pulled a digispark out of it's box the other day and finally built one of those 'USB Volume Knob' thingys... and it programmed without a hitch, several times as I was refining the code and debugging.

[peterhowells]

@pfeerick Thanks for the very long reply. Unfortunately most of it does not apply in my case. As I stated in the first post one of the Oaks is working fine (after jumping through the hoops of getting it programmed). The other seems to be an OOB failure - only the power light comes on - not flashing led, no ability to update, not even able to update using a USB to serial converter. This points to a lack of quality control!

I have been programming embedded systems for almost 20 years so have experience in this - I just find these devices to be half baked- they should have not been released in this state. Compared to other devices which I use regularly (Particle Photon, Redbear Duo), these are not worth the trouble.

[defragster]

@peterhowells: No doubt the initial 'Update' software could have been (much) improved before shipping given another iteration and wider beta perhaps, small scale testing apparently didn't match the cruelty of a mass ship real world event - but the hardware is based on an off the shelf ESP-8266-12F. There could be a problem there or in the OAK connectivity support hardware - if a hardware failure is indicated - as it seems - email Erik and he'll work with you to make it right. There were at least 15-20,000 units built - I've not seen many reported as dead - indeed had 3 or even 2 of mine failed my opinion would no doubt be different.

For OAKS - I ended up with 7 of 7 complete and fully functional boards - I got my BETA hardware to update okay during the Beta period - but my final order 6 of 6 failed internet update.  I tried LOCAL update and they all came to life and worked reliably a time or two - save for one that I re-updated and now all seems well with it. But I had no lack of blinking 'life' indication as you are seeing.

I worked with the PRO and somehow (after many edit/upload cycles) bricked 3 units and got fresh ones - and they had already been soldered and used.  I could have reflashed them myself but hoped they would show Erik something so I even returned them. And the first replacement set went into the USPS and never came out - Erik didn't hesitate to re-ship another set.

As far as I can see (DigiSpark, Digix, Pro - and now OAK) Digistump makes honest efforts to make quality products - and stands behind them - if you haven't contacted Erik - you should.

[digistump]

I'm sorry to hear you are disappointed,I won't rehash what others have said, or what I have said elsewhere - but I do want to add: Please email me - support @ digistump . com - so I can replace the Oak that was DOA, I can't make up for the first firmware update issues, but I can say that the hardware failure rate is pretty low and I always replace DOA devices.

[PeterF]

Same here defragster - no failures bar initial update gremlins, all fixed by the local server update first time every time. I have five of the 'lil fellas all updated and tested, and have two of them running 24x7 (for about the last 3-4 weeks now?) with some periodic logging temperature on one, and a display on another. And a third comes out when I need a victim for some test code. And touching on the Digisparks and Digispark Pros again for a moment, I would have thought they are just as flakey as any other Attiny vUSB self-programming setup, due to the way they are cheating at self-programming. They don't have a protected bootloader, nor do they have a separate USB support chip to program them. Bet a Digispark Pro with a CH340 for programming would be virtually unkillable (programming-wise anyway)!

Sounds like you may indeed have a faulty unit if you been through the factory reset / OakRestore process. If that's the case, as Erik said earlier, contact him! He'll make sure it is rectified.

I can't speak for the Photon or other boards, but I can understand your frustration. However, I've tempered mine with the understanding that this is a new product by a small family business, and they don't have a big research and development division to test all this stuff out. And since I got it as part of the Kickstarter campaign, I basically said I was ready to be a guinea pig, rather than wait a few months for the more stable release to come out.