I’ll will gladly share my experience and to my best ability and availability help Particle users with their project-to-product questions.
I agree also, the more successful product examples the better for the rest of us attempting to do the same thing.
I’m in the middle stages of launching a product (beta test complete, about to launch to the public next month). I’m open to any questions or talking to you. US based.
Thanks for all the replies, I’ll drop you guys a PM and see if we can arrange a chat, that would be really useful.
I have already contacted Particle sales, who did not respond and Support who did respond but said they don’t have anyone I can speak to over the phone.
I’ve got to be honest, I really want this project to work, I love Particle and have been using them since the first batch of Cores shipped. I’ve tried several times in the past to get a project off the ground with Particle but I always seem to run into endless intermittent issues and inconsistencies. The community is great although I’ve never managed to get to the bottom of any of the issues. (BTW all these issues have remained even when using Tinker and other “stock firmware” for testing). I can see there are some people who seem to have the same experience as myself and are at their whit’s end trying to get a product released. However, I can also see a lot of people who have done this successfully. So the platform can’t be inherently broken otherwise no-one would have succeeded.
In the past support have been great as diagnosing specific issues, although 100% of the time they have ended up saying the Particle/Electron in question is faulty, which I suspect is not the case. What I don’t understand though is that the whole premise of Particle is it’s a platform that allows you to go from tinkering to producing a fully fledged consumer product. However from my experience when you try to make the move to something more serious it all seems to fall down and the enterprise level support doesn’t seem to be there.
I’m reluctant to start a new business venture where the core products will be based on a platform that I can’t pick up a phone and talk to someone about. On the face of it Particle seems the perfect choice, I get on great with it for one-off projects and tinkering and I really want to use it to do something amazing, but I just feel like when you dig a bit deeper cracks start to appear.
In my experience, Particle uses Zendesk and has resolved every ticket I’ve opened with them via their contact form:
I think the most important thing you can do as a founder is build your product by hand, set it up, use it, and resolve issues. I’ve been doing that for a while now, and its given me much more confidence in the Particle ecosystem.
As a subcontractor, I can say that one of the companies I did a project for is also getting ready to release a P1 based product in a month. Customers have been using the prototypes for 3 weeks now, and everything is running great.
@Nemiah I’m sorry you haven’t received a reply to your sales inquiry yet. When did you send the email? If a reply has been long delinquent, I’m happy to help to ping internally to see what the issue was or to ensure a conversation is scheduled.
You mentioned issues with your product and I don’t want to turn this thread into a troubleshooting conversation, but if there is feedback that you’d like or are willing to share over PM, I’d love to understand the good’s and bad’s of your experience so far so that I can work with our product and engineering teams to improve on that experience in the future.
Thanks – will reply on that issue in that thread to try and get an internal engineer to take a look.
Beyond that, a couple different things to note:
It is entirely possible that there is a bug in the code that Particle provides to support SoftAP setup mode. It is also entirely possible that the root cause is not located within Particle’s implementation but results from interactions between our library and custom code. Our Zendesk customer support team is really not designed to determine either – troubleshooting system features with Particle is best done via documented bug reports in GitHub that provide detailed info for recreating the bug and are actionable for our engineering team.
We absolutely have the ability to assist product creators with code reviews, advisory services, and phone conversations to ensure that our end customers are successful on the platform. That is not the role of our customer support team, but it is the responsibility of our sales and customer success teams. If you have a need for those kinds of services, I’d be more than happy to connect you with the right team members at Particle.
Thanks for taking the time to get involved @will I contacted sales via the web form, I’m not sure when exactly but it was the other day. If you could just get someone to drop me an email on firstname.lastname@example.org and arrange a phone/web call that would be brilliant. As mentioned I’d just like speak to someone about turning a prototype into a consumer product over the next 6 months. Even if there are issues that we can’t solve in-house I’m more than happy to pay for your enterprise services or elevated support, these are exactly the kind of things we are looking for to get this product launched in such a short space of time.
I’ll drop you a PM and let you know about my experience when I get a bit more time too if it’s useful (I’ll try and not bore you to death!)
Sure thing – will do!
Are you able to share any info about what you have built?
Can you give us any info on what you guys created?
I think it’s safe to say we would all love to hear about it if your able to talk about it
I built Smartfire BBQ (http://smartfirebbq.com/products/smartfire-bbq-controller) on top of Particle. First on Photons and then moved to P1 modules about a year, and have over a thousand in the field.
- Particle are well setup to service commercial customers, with an online wholesale portal with transparent pricing and lead times
- The team are genuinely friendly and trying to do well for customers while keeping an eye on longterm company sustainability. Seems a decent balance is going on.
- They have a crack team of hardcore firmware folk, years of effort has gone into the firmware and most the gremlins have been long buried.
- The customer forums are active, kicking and very well monitored and participated in by the Particle folk. There are some genuine heroes who seem to work 24/7 helping folk with libraries like @peekay123
- Fleet management is decent, you easily can roll your own like we did by watching particle subscribe mine
The Not Ideal
- No BT integration, we had to go to P1 and roll our own
- Particle keys occasionally get corrupted and requires you to RMA the customer’s device so you can replace it at your cost because the Particle cloud doesn’t have a neat way of redirecting the device to a non-secure portal. Our priority is customer devices working, communicating BBQ temperature status updates isn’t top secret so not particularly worried if a MITM attack occurred while updating the device keys with a fresh set
- Particle webhooks aren’t standards based and if they see an error code returned they’ll queue another attempt, and after N failures they’ll mark the webhook as bad and it’ll be taken offline. So even if the request was invalid make sure to return a 200 OK. It doesn’t make sense, its flawed, but just roll with it and use AWS HTTP gateway to massage everything into happy as larry 200’s on the response.
- Particle is a fairly limited platform in terms of cpu and memory, I wish it had a bit more grunt so we could connect to MQTT over TLS etc as I’d prefer to connect it to Google IoT cloud etc.
- Pricing is on the expensive side, it’s going to be commercially tempting to move to the likes of ESP32 over the next 2 years unless a new Particle generation comes out that is more competitive against a rapidly maturing competitor. Particle P1 is ~$11 and Microchip RN4870 for BT is another $9, whereas the new kid on the block ESP32 is $4. There’s whitespace there.
The Damn Ugly
- The wifi stack only supports B/G/N operating in 20Mhz bandwidth setting. When I raised this topic the surprising claim was that this hadn’t been seen as an issue. It’s a massive issue as modern wifi routers default to 40Mhz so after around 15 mins the Particle will be dropped from the wifi network as the ‘conversation’ drifts into the other 20Mhz of spectrum that the Particle isn’t aware of. This causes angry customers to jump on forums and the phone to complain that the product doesn’t work. I’ve personally spent hundreds of hours explaining to customers how to change their wifi router setup, for the technology challenged folk this has involved me setting up a calendar time to remotely connect to their computer via teamviewer to change their wifi router settings myself. Something to be careful of, and potentially proactively put into the setup professional services of your product if you have the opportunity to do so. On our consumer electronics product we don’t have that opportunity.
Overall, quite happy and would recommend for current commercial projects.
40MHz vs. 20MHz WiFi bands - possible issue
And to expand on @mterrill’s “Not Ideal” and “Damn Ugly” statements, Particle is working on a next-gen fleet of devices that will address almost all of the HW related points of (honest and fair IMO) criticism
That’s something to mark as “The Good”: Particle has an open ear for customer’s wishes/demands. They may need some time but they are not deaf to the needs of customers and the Elite members try to bring burning needs to Particle’s attention in case they just got burried in the noise of less-than-necessary “claims of demand”.
So things are bound to get better and the list of “Not Ideal” and “Damn Ugly” should get shorter soon.
P.S.: The keys issue should alredy be gone with system v0.7.0 (currently RC state) as it should feature “self-healing” keys.
Core issues fixed in 0.7.0?
Product is a POE enabled VoIP intercom.
Firstly, appreciate your involvement in the community forums as well!
Secondly, new hw generation? How do I get involved? Currently making commercial ordering decisions for the next 6 months of supply and new products. @will, happy to sign NDA’s or whatever is necessary.
Thirdly, keys are still broken on v0.70, tried and failed, feedback given on the thread…
I’m also nearing the launch of a commercial product based on the Photon. Would love to know if I need to make any future-proofing decisions in my PCB design. Are there any details about new hardware that can be released?
We are also close to launching of a commercial product based on the Photon. Let’s hope this announcement will be soon…
Can you give us a general idea of what the product is?
Can a P1 be setup to connect to the internet without claiming it?