[NTLK] MP2x00 Internal Interconnect Breakout

[Jake Bordens]

One note, I have been able to successfully call the 
SerialPortXXXLineDriverConfig functions and control the pins manually.  
Both Select0 and Select3 work as expected. I whipped up a little app to 
manually toggle the signals:

http://2.bp.blogspot.com/-2ZMlwBeOJNQ/VLcVzjiQKGI/AAAAAAAAAaE/1GlD45N1hLs/s
1600/serial_enabler_screenshot.jpg

This is based on Eckhart's code, which provides a good blueprint for 
calling arbitrary functions in ROM.  I don't call the WriteDIO/WriteGPIO 
functions directly, but call the LineDriverConfig functions to do it-- you 
just need to have the right parameters.

Also, just so that I have a good place to post images and content, I setup 
a little blogger blog roll.  I'm not really a blogger, but I wanted 
someplace that will have some sort of staying power.  I figure since 
blogger.com is google-owned, it'll likely be around for a good long time.

http://retronewton.blogspot.com

Most of the content is copied from my posts here... But there are a few 
new images, such as a oscilloscope trace of the 5ms pulse that may be 
Matthias' interrogation/handshake.

Jake




On 1/14/15, 12:37 PM, "Jake Bordens" <jake at allaboutjake.com> wrote:

>
>>If I add a little pause between blocks, I hope to slow down the 
>>connection just enough, without slowing down the PC - it just makes the 
>>USB buffer available a little bit later.
>
>Ahh. Okay.  I understand now. Thanks. Sounds like a Microcom networking 
>protocol issue.  Unfortunate that this couldn't be solved with a 
>host-side 
>virtual serial driver or something though.  Then it would be available to 
>everyone without a hardware patch.
>
>>Since the original tools no longer work on OS X, this problem has not 
>>crept up. I assum the current OS X tools are reliable. In either case, 
>>the delay would not be a disadvantage, since it only synchronizes 
>>actions.
>
>I can tell you that in my little time spent working with MNP (rewriting 
>the PackageBuddy receiver) that the tools descendent from UnixNPI have 
>indefinite waits for the acknowledgement before sending the next packet.  
>My guess is that all OS X tools do the same.
>
>----------------------------------------------------------------------
>
>The NewtonTalk Mailing List   http://newtontalk.net/
>The Official Newton FAQ       http://splorp.com/newton/faq/
>The Newton Glossary           http://splorp.com/newton/glossary/