On Tue, Jan 11, 2022 at 05:23:02PM +0100, Patrick Wildt wrote: > Am Tue, Jan 11, 2022 at 01:04:52PM -0300 schrieb Crystal Kolipe: > > On Tue, Jan 11, 2022 at 04:54:08PM +0100, Patrick Wildt wrote: > > > Preorders for the Pro Explorer Edition will come up later today: > > > > > > https://www.pine64.org/2022/01/11/pinephone-pro-explorer-edition-pre-orders-open-january-11/ > > > > > > I'll order one. > > > > You might want to order the I2C keyboard as well. > > https://pine64.com/product/pinephone-pinephone-pro-keyboard-case/ > > This one?
Yes, that works with both the original and pro pinephones. I'm assuming that it will work with our ihidev driver, although that isn't currently in GENERIC on arm64. I should be receiving one of these keyboards sometime in the next few weeks. > > Does this mean that interest in running OpenBSD on the original model is > > going > > to fade? > > > > I'm not really inclined to purchase the pro with my own funds, and as far > > as I > > know Exotic Silicon are not planning to buy one for me to play with. > > Well, the Pinephone has an Allwinner A64, which is indeed quite old. So > I didn't have much interest in it anyway. :/ We have better support for > the Pro, as it's the same chip as on the Pinebook Pro, but there's still > plenty missing for a proper phone. Well, I think there are two separate aims here, firstly getting the hardware usable as a PDA, I.E. keyboard and framebuffer console support, working power management, USB, and use of the modem for data. Beyond that, to make it into a phone, rather than a PDA, you'd need a whole set of userland programs running on top of that, for a custom phone GUI with touchscreen support, etc. I'm more interested in the first part, because I think that writing the phone userland would be quite easy. The existing Linux-based projects seem to be suffering from the common problems in these sorts of projects, feature-creep, a lack of really skilled developers, lack of interest in writing code for the important core fundamentals, massively over-complicated multi-layered software stacks that developers do not really understand, leading to instability that is difficult to pinpoint and diagnose, etc, etc. So, the current choice of hardware comes down to either the original model, which is cheap, almost disposable, but powerful enough, and has had the known hardware bugs from the early revisions fixed. The issues that I have with the pro at the moment are: * It's the first version of the hardware in commercial production. The original pinephone went through a few revisions as hardware bugs were found. * The battery life is reportedly much less, which is to be expected from the new Rockchip chipset. * As much as I prefer Rockchip over Allwinner for SBCs, where the eMMC is removable for flashing, the pinephone pro has fixed eMMC. Rockchip chipsets prioritise booting from eMMC over a memory card. So unlike the original pinephone, if you write a bad image to the eMMC, you've got to short jumpers on the PCB to clear it. As far as I know there is no SPI flash to write a known good bootloader to. * Although we have framebuffer support for the new chipset, which is mostly what is missing at the moment on the original, at least from my point of view, the display is connected via MIPI, which we have no code to interface with. So it won't work out of the box anyway, and I don't know how realistic it is to write the necessary supporting code. So basically, if bringing framebuffer support to the original pinephone is not a major hurdle, even unaccelerated support, I don't see much point in spending $400 for the pro. But then, I've already got several original Pinephones. If I was buying my first, then maybe I'd go with the pro.
