Randy Bush wrote: > > Unless we are strongly time-constrained for the alpha > > not at all.
Sarcasm isn't helpful. Releasing is important but let's consider what the suggested change costs: I posted an ASCII schematic indicating the simplicity of the suggested circuit. The little Cortex-M3 might not have a symbol in EAGLE, creating a symbol with only the 10 or so neccessary pins named takes what, two hours? The PCB layout needs an LQFP-48 footprint, maybe another two hours. *Very* safe estimates. I spent about an hour creating the full footprint in the software I use. I'm happy to contribute hand-drawn schematic, .png routing and .txt BOM which I have used successfully. Add another two safe hours for deleting the old circuit, drawing the new one and routing it on board. Add 33% margin. That's one day for a change that brings significant technical benefits, few if any drawbacks, and seems desirable to a couple of team members. The toolchain used for the STM32 should work fine for the little CM3, if not there's a build script and binaries on my workshop site. I offered to contribute firmware. I will be visiting Stockholm in February anyway and will gladly join Fredrik for a day if he needs any onsite hardware assistance and/or to do firmware work. In case it wasn't already clear, a USB CDC (USB serial port) firmware is of course already available, so making the hardware change doesn't mandate designing a USB protocol immediately and doesn't otherwise close any doors for the software - it only opens new ones. //Peter _______________________________________________ Tech mailing list [email protected] https://lists.cryptech.is/listinfo/tech
