UPDATE: What happened to Mikronauts other boards, announced at UPEW?

Bill Henning

Hi,

I case any of you have been wondering what happened to the other boards I announced and showed at UPEW...

My previous PCB manufacturer let me down in two ways:

1) Some of the boards I ordered did not show up until the end of July

2) A lot of the boards had manufacturing issues

I had to find a new PCB manufacturer, and I had to send some boards to them to test out their capabilities.

I am happy to say that the new manufacturer does VERY nice work, and the two boards I have built and tested from the initial run worked right off the bat - thus I will be offering both PropCade and 485Plug for sale by Monday.

I will be testing Morpheus (pcb rev 2) and Mem+ (pcb rev 2) on PCB's from my new manufacturer, and should have them available for sale (assuming no PCB issues) by Oct.11

The other boards from UPEW will follow, as I have time to send them in for production to the new PCB manufacturer.

Sorry for the delays.

Bill

p.s.

I received the new boards one week ago.

p.p.s.

The manufacturing issues were due to the previous PCB manufacturer not following Sapieha's excellent gerbers correctly, and some via's were not plated correctly.

David Betz

Is FlexMem one of the ones that got delayed? PropCade looks pretty cool! I think it should run the software I'm currently working on for the C3 since it has the SPI SRAM capability. In fact, I guess I could use VMCOG or maybe my direct-mapped cache code with two of the SPI SRAMs and use two more for scratch file storage and get the best of both worlds! How are the SPI control lines and the chip selects handled for all of those SPI sockets?

Bill Henning

Hi David,

FlexMem is actually OK PCB wise - or so I think - however the need to find another PCB manufacturer, and work on the other boards, is causing delays with getting the driver running. It is pretty close, the latest test version gets 28 bits of 32 correct; there is a weird thing happening with the last nibble... I'll work on it some more this weekend.

I'd suggest using VMCOG, and I'd appreciate it if you emailed me the changes needed to get VMCOG running on the C3.

Why do I suggest this?

Because then your software will work on every other platform that VMCOG supports!

And I am working on FlexMem support for VMCOG...

FYI, the PropCade VMCOG driver is well tested, it was my VMCOG development platform.


PropCade SPI docs:

movd outa,#x ' select SPI device, 0-5 are SPI RAM/FLASH sockets, 6 is MCP23S17, 7 is uSD

/CS = P0
CLK = P1
MOSI = P2
MISO = P3

i.e. P9=A0, P10=A1, P11=A2 for the 74HC138 on PropCade, and P0 is connected to its /E1

As long as the address is set before accessing an SPI device, this method is compatible with existing Obex SD etc objects

David Betz wrote: »

Is FlexMem one of the ones that got delayed? PropCade looks pretty cool! I think it should run the software I'm currently working on for the C3 since it has the SPI SRAM capability. In fact, I guess I could use VMCOG or maybe my direct-mapped cache code with two of the SPI SRAMs and use two more for scratch file storage and get the best of both worlds! How are the SPI control lines and the chip selects handled for all of those SPI sockets?

Phil Pilgrim (PhiPi)

Bill,

'Care to divulge who the pervious and current fab houses are?

-Phil

Bill Henning

Phil - you have e-mail