The Core is EOL?

I see in this post from Will that the Core is now considered EOL by Particle and they are devoting most of their efforts to Photon. I can’t speak for other Particle customers, but such a short life-span raises lots of concerns for supporting my projects. For those who have and are developing products around the Core, it must be alarming. I would like to hear an official comment from Particle on the EOL status of Core, as well as their intentions going forward for the Photon, Electron and whatever comes next. I know we live in the pre-Singularity age, but 1-2 year life-spans are problematic at best.

2 Likes

Hey @Muskie – sorry I was not specific enough. EOL refers specifically to the hardware. We currently have no plans to manufacture more Cores, given that the Photon is faster/cheaper/more capable. That does not mean that firmware development for the Core will end–in fact, a new firmware version for the Core is due to be shipped in the next couple of weeks.

Furthermore, we’ve gone through significant efforts to make sure the Photon is backwards compatible with the Core in both hardware (exposed peripherals and GPIOs) and firmware. That means that the code you write in the IDE won’t change for the Photon from the Core, except that you’ll have more hardware features to take advantage of (switching between onboard and external antennas, for example). We’re not dropping support for the Core–although Photons are being shipped, your Core projects will continue to function as they have been, and we’ll continue to respond to support tickets and community posts as always.

I hope that helps to dissuade your worries. Really, the only thing that’s being deprecated is the older, less effective technology and $39 price point :slight_smile:

5 Likes

I've edited my post from the original to the following to be more specific:

Thanks for the feedback and the kind words! The Core is now out of active production so we only have a limited number of units remaining, but let us know if you have continued issues with the Cores on hand. We may still have some u.FL Cores in stock if you think it warrants replacement.

Hello @Will. I understand that I can still support my Cores for code updates, support, etc. My main concern is the hardware and obtaining replacement Cores as needed for some future time frame. Since you are no longer manufacturing Cores and as your stock depletes, my future for hardware replacements in my existing projects appears to be: start over with a Photon or whatever you are supplying at that time.

And can you provide an official lifespan for the Photon? Will it go EOL in about a year from now?

For hobbyists and hackers this may be only a minor concern, but for those who are building commercial products or whole companies around the parts, it is critical. EOL in 1-2 years can be a deal breaker.

Hey @Muskie,

Because the Photon is hardware and code compatible with the Core in most scenarios, switching from the Core to the Photon does not require you to “start over” per say. For the majority of use cases, it’s as simple as swapping one board in for the other (same pinout/footprint), and flashing the same firmware through our online IDE. In fact, you’ll likely see significantly improved connection speeds, faster over-the-air updates, and more reliable connectivity due to the improved radio firmware from Broadcom over TI. This all comes at a price point that is half that of the Core.

With that in mind, we’d actively support defective Cores in our customers’ projects by issuing replacement Photons, so any projects that are currently Core-powered have active hardware support in the event of malfunction or hardware error.

We don’t have an officially determined production lifespan for the Photon yet, since we’re still working through fulfillment of our first orders. We’ve invested significant development effort to make it a flexible development kit that can be used in small-scale production, but recommend using our P0 and P1 modules for large volume product deployments in order to reduce costs. In large part, Broadcom and USI, the manufacturer of the P0 and P1, will determine the lifespan of this part, since they control the supply chain that manufactures it. We can promise, however, that we’ll as transparent with our community as possible with respect to product life cycles as we learn about them from Broadcom/USI. There are a number of mass manufactured consumer products, including LIFX and Nest Protect, that use the same connectivity module, so that’s unlikely to happen anytime in the near future.

I hope that helps to answer some of your questions. Product life cycles can be difficult to manage, especially when the supply chain is spread out across a number of different manufacturers, but we appreciate the questions. This conversation has prompted some internal discussions about opportunities for improving documentation about switching from Cores to Photons, so thanks for raising the topic.

7 Likes

Thanks Will.

2 Likes

It is far too premature to state that the Photon and the Core are ‘code compatible’. I have tried 3 different displays in the last couple of days only to find that the existing ports of Arduino libraries are 'Core Only". So they are NOT interchangeable. It’s official hyperbole like this that makes me lose trust in the Particle team.
When the Photon software is capable of running Core software without issues, then it can be stated that the devices are interchangeable. Stopping manufacturing and sales of Cores before the Photon is completed, without providing a couple of years of of full hardware sales and support for the outgoing device, will make any ‘real’ customers look elsewhere.

@Awake, I understand the frustration. The issues with any existing hardware-based libraries like display drivers is that they were ported with very specific code for the Core, before the design of the new HAL (Hardware Abstraction Layer) which will come to the Core very soon. The need to update the existing libraries so they become future proof is acutely known to the Elites and the Particle team. With the upcoming release of the new Photon firmware version, the Elites and other library contributors will be able to update their code to work with all Particle devices (including the Electron!).

With this in mind, I had a discussion yesterday with Particle regarding the need to put out a Core to Photon hardware and software transition guide to make the change less frustrating.

In the meantime, the community and I can assist in your immediate library challenges :smiley:

4 Likes

@Muskie and @Awake, thanks for your comments.

There are a number of factors that are forcing us to move towards the Photon from the Core as quickly as possible. TI is no longer supporting the CC3000 (they’ve moved on to the CC3200, which is a totally new architecture), and their chips never got the widespread distribution in the market that the Photon’s Wi-Fi chip (the BCM43362) has gotten. The Broadcom software stack is much stronger than TI’s, so it’s an important move for us, and we believe for you too.

The embedded microcontroller world is very fragmented, and compatibility across hardware architectures is a real problem. We’re attempting to solve this with our Hardware Abstraction Layer (HAL).

That said, the HAL wasn’t yet implemented in the Core, so the Core sort of stands on its own little island, whereas all of our future hardware will be a lot more compatible with one another through the HAL. The Core will be upgraded to the HAL soon, but there’s still a bit of work we have to do behind the scenes to support this, since it’s a pretty dramatic change.

Once the Core is on the HAL, compatibility will move forward quite a bit. The HAL and the Photon are still both very young, so things are currently changing pretty rapidly, but as with the Core, a lot of this will settle out over the next couple of months; libraries will be ported by their authors in the community, and we’ll be able to maintain a lot more hardware compatibility going forward because of all this work.

I hear your frustrations, and I’m sorry about the disruption. We plan on supporting the Photon for much longer, and we plan to stabilize our firmware stack around the HAL so that future hardware transitions are more seamless. Thanks for your patience in the meantime.

8 Likes

@zach
Thank you for the reply. It was excellent and helps me further understand the evolution of your products.
My issue is with your staff calling the Photon a “Drop-in-replacement” for the Core, in the sense that if you are using the Core now, you can just buy a Photon and use it in its place, when that is absolutely not the case. Maybe someday, but not now.
Your team is obviously very smart, but maybe they are too smart for their own good. Lots of ‘advanced’ features are being developed, yet it feels like the basics are being ignored. It is getting to the point where unless you already know very well what you are doing, you can’t use the Photon. Like examples and documentation. Heck, I can’t find a single display that is compatible with existing libraries, and checking to see if a common one would work makes me jump through dozens of messages in a disorganized jumble of forum threads. We get things like “Azure data analytics”, but we don’t have a clear article on basic use of interrupts on the Photon. How about starting with an official list of ‘approved’ devices (temperature sensors, displays, GPS’s, etc) with usage example pages, so that the Arduino level user can get the Photon to work without just giving up?

I hear all of your concerns @Awake. Here are some of the things that we’re doing to address them:

Documentation: We’re going through a major refactor of the documentation that will improve the quality of the content and make it easier to navigate between Photon-specific docs and Core-specific docs. If everything goes as planned, this will be made available next week, or the week after at the latest.

Libraries: You’re right that we’re behind on this one, but unfortunately it’s the nature of how the libraries get developed. All of our libraries were developed by the community, and since we’re shipping out Photons to everybody at (roughly) the same time, everyone’s getting their Photons before the library authors have had time to port over their libraries. It would be great if the libraries didn’t require porting at all, and this is what the HAL will provide. But as I mentioned in my previous post, the HAL wasn’t ready yet for the Core, so in the meantime, libraries need to be tweaked for the Photon. We’re reaching out to library authors now to ask for their help, and some of the Elite have graciously volunteered to port over libraries, so if you need something specific, just ask!

I agree that there are probably some very widely used sensors and components that we should support as “first class citizens” and take the libraries under our ownership so that we can ensure that they always work with all of our hardware (and also write up great documentation and examples for those libraries). There is, of course, a negative effect there — we would prefer to be working with the community on this — but it’s something for our engineering team to consider, of course. I’ll be sure to raise the idea with the team.

3 Likes