I have been part of the PandaBoard project since the beginning. When we launched the PandaBoard about 14 months back we did not put a specific usecase around it. We had a single goal “create a open development platform available with latest cutting edge OMAP SoC and make it available at the lowest cost possible to enable broadest adoption”.
Developers around the world have embraced the PandaBoard and we have shipped a ton of platforms and strong demand continues.
The PandaBoard platform continues to evolve and lead the development platform pack with the launch of PandaBoard ES leveraging 1.2Ghz Dual Core OMAP4460 SoC.
Developers did amazing things with PandaBoard which we could not have possibly imagined at the time of conception. For example: Mind controlled robots, autonomous UAV & cars, Clusters & Builders, Software Defined Radio, Devices for assisting visually impaired, media centers to name a few.
A lot of PandaBoard developers have asked for making their own OMAP4 powered products. Given the OMAP44xx 0.4mm BGA package is not easy to work with and complex to build a board with for most production houses and challenges in support. We have worked with several 3rd Party SoM vendors to provide access to low volume SoM. This enables developers to design a board & product around these SoM and procure them in low volumes. We continue explore options to a enable broad range of products .
All in all this reinforces the belief of not putting a firm use-cases around the platform but to enable open innovation with the PandaBoard. I do want to attempt and pen down a few key things that set’s PandaBoard apart:
- Open: PandaBoard platform is open and will always be.
- Cutting Edge Application Processor: PandaBoard platform will have the high performance and low-power leading edge OMAP SoC
- Lowest cost: Low cost principle is to enable access to broadest range of developers without compromising the capabilities and application potential the OMAP SoC can enable. A couple of examples:
- we opted to have 1Gb of LPDD2 (RAM) instead of cutting a corner and saving a few dollars on the final price. This enables developers to leverage the full power of Dual Cortex-A9 powered OMAP4 SoC for their applications without crippling their creativity due to lack of RAM.
- We could have choosen to drop the simultaneous dual display support and save a few dollars but we believe it’s important to enable such differentiated features of OMAP SoC and enable developers to leverage it.
- We could have chosen not to have world-wide distributors sell the PandaBoard (i.e., build/ship directly) and save a several dollars typically in distributor mark up. But, we find that having the convenience to buy PandaBoard with familiar & local distributors makes lot of sense to most developers and we choose to make it broadly available with distributors. That said we will continue innovate on cost.
- Robust Platform: We learned from looking at other open-source hardware project experiments that it very critical to have a robust and fully functional hardware platform at launch. i.e., we cant expect to sell a <$200 platform and expect the developer to upgrade as something is not yet working in the hardware they have at hand. So, the core driving principle is that when we ship – we ensure all the hardware is fully working and all peripheral are as robust as possible
- Strong Upstream support: OMAP platform has been leading the pack interms of upstream kernel support and PandaBoard platform will continue to have solid support in upstream. We believe this is a key to enable developers and projects around the platform. In addition PandaBoard happens to be the ONLY development platform support in upstream Android Open source project.
The PandaBoard project will continue to evolve innovate and listen to the developer community. The future is bright and awesome!
Disclaimer: The content of this posting is my own opinion & view and dont necessarily represent my employer’s (Texas Instruments Inc.,) positions, strategies or opinions.