TheBackShed.com

Forum Index : Microcontroller and PC projects : Micromite MK2++, beta testers wanted

Page 3 of 4

Author

Message

JohnS
Guru

Joined: 18/11/2011
Location: United Kingdom
Posts: 4044

Posted: 07:06am 06 Feb 2015

Copy link to clipboard

Print this post

It boots from SD card. You put on the card what you wish. Newbies are usually wise to start with something that does lots since give them a minimal system and a data sheet etc and they're lost. (Even if they tried to read it, which almost all won't.)

If you're genuinely interested do the reading, load something much simpler/smaller/faster/etc.

It all depends on your aims. RPi wasn't originally created for experienced uC guys to run bare metal code. It's sold 4 or is it 6 million so they got something right.

But it can be used differently if you want.

It's akin to someone wanting to run C on a 'mite. It was created for MMBASIC. Doesn't mean C can't be used.

There are Linux systems that boot in under a second. But not to a fully-featured OS with web browser, email, video player, audio in/out, GUI, drag-n-drop, USB host/OTG, and all that stuff. Gee, why does that stuff slow anything down? Why doesn't a maximite have it all?

JohnEdited by JohnS 2015-02-07

matherp
Guru

Joined: 11/12/2012
Location: United Kingdom
Posts: 10315

Posted: 07:09am 06 Feb 2015

Copy link to clipboard

Print this post

Here I am trying to encourage additional beta testers and the thread ends up as a discussion of some sort of computerised fruit pudding with my post relegated to a previous page

I'll try again.

Please PM me if you would like to help testing the new double speed, double memory, double pin count Micromite

Thanks

Peter

WhiteWizzard
Guru

Joined: 05/04/2013
Location: United Kingdom
Posts: 2944

Posted: 07:16am 06 Feb 2015

Copy link to clipboard

Print this post

You are so right Peter - I am sorry (even though I did start my post with 'Perhaps I should start a new thread . . . ')

To hopefully go some way to make up for the deviation, can I strongly recommend people out there with the time/skills/hardware to help out with Beta testing Peter's excellent MX470 code. Think of this as a MicroMite MX170 on steroids.

And lets not forget all the excellent code that Peter has written allowing many of us to now start using TFTs and OLED graphical displays - if we want this to continue then let as many of us as possible get down to some serious MX470 beta testing

100pinner now has 79 Digital I/Os by the way but this all needs testing . . . .

WW

Edited by WhiteWizzard 2015-02-07

viscomjim
Guru

Joined: 08/01/2014
Location: United States
Posts: 925

Posted: 02:59pm 06 Feb 2015

Copy link to clipboard

Print this post

I would love nothing more than to be part of the beta testing, but I don't have the facilities to prototype one of these chips. I would love to start with the 64 pinner. Hopefully one of the Grogster PCBoardster (or other) boards will be available soon that can be plugged into a breadboard and played with, without too much difficulty. I am drooling just reading about this beast... especially with the Matherp display gallery!!!

hitsware
Guru

Joined: 23/11/2012
Location: United States
Posts: 535

Posted: 04:07pm 06 Feb 2015

Copy link to clipboard

Print this post

> Perhaps I should start a new thread,
> but,
> why does the Pi take soooooo long to
> startup with all that Linux garbage
> when the processor is that fast?

It's meant to be left on ....
'Boot time' is beat to death

But not always mission critical

Edited by hitsware 2015-02-08

palcal

Guru

Joined: 12/10/2011
Location: Australia
Posts: 1993

Posted: 08:54pm 06 Feb 2015

Copy link to clipboard

Print this post

I have ordered the parts and boards. Can't wait to test some displays.
Paul

"It is better to be ignorant and ask a stupid question than to be plain Stupid and not ask at all"

Grogster

Admin Group

Joined: 31/12/2012
Location: New Zealand
Posts: 9610

Posted: 01:00am 07 Feb 2015

Copy link to clipboard

Print this post

@ Jim - boards ordered - I will update the forums, as soon as the factory lets me know they are done - should be sometime this coming week, with any luck, and I get DHL delivery for faster shipping time. Watch this space.......

Smoke makes things work. When the smoke gets out, it stops!

TassyJim

Guru

Joined: 07/08/2011
Location: Australia
Posts: 6283

Posted: 03:50pm 11 Feb 2015

Copy link to clipboard

Print this post

My PIC32MX470F512 chip arrived today. Just the one chip.
I was a bit taken back by the packaging.

Now I have to wait for the board to put it on.

Jim

VK7JH
MMedit

Grogster

Admin Group

Joined: 31/12/2012
Location: New Zealand
Posts: 9610

Posted: 04:05pm 11 Feb 2015

Copy link to clipboard

Print this post

I got my chips earlier this week, but they came in one of those little black Microchip plastic carriers - your shipping method looks pretty big!

Factory is in the middle of Chinese holiday at the moment, so says the S2U website, so I am guessing that although I got my order in, it was right at the start of that, so this means that we have to wait longer then usual for the boards, which is annoying. The website says they are back at work on the 26th, so it is likely to be after that before the boards arrive - a shame, but what can you do? C'est La Vie.....

Smoke makes things work. When the smoke gets out, it stops!

bigmik

Guru

Joined: 20/06/2011
Location: Australia
Posts: 2950

Posted: 08:00pm 11 Feb 2015

Copy link to clipboard

Print this post

Hey Jim,

Mine came yesterday too..

Pictured next to one of those ethernet to serial adapters that I ordered at least 2 months ago (which finally arrived today)..

Yours looks like the 64pin baby..

I ordered the 100pin one..

I hope to mount mine on a UBW32 board.. I need to check pinouts first and work out if its worth doing.

Mick

Mick's uMite Stuff can be found >>> HERE (Kindly hosted by Dontronics) <<<

palcal

Guru

Joined: 12/10/2011
Location: Australia
Posts: 1993

Posted: 10:39pm 11 Feb 2015

Copy link to clipboard

Print this post

I ordered mine from Microchip as Samples. They are sending 4 x 64pin and 2 x 100 pin.
They were very quick, posted in 2 days from Thailand so should have them soon.
Might as well get freebies if you can.
Paul

"It is better to be ignorant and ask a stupid question than to be plain Stupid and not ask at all"

Grogster

Admin Group

Joined: 31/12/2012
Location: New Zealand
Posts: 9610

Posted: 12:36am 12 Feb 2015

Copy link to clipboard

Print this post

Excellent thinking, Paul - I wish I had thought of that!

Smoke makes things work. When the smoke gets out, it stops!

matherp
Guru

Joined: 11/12/2012
Location: United Kingdom
Posts: 10315

Posted: 05:39am 14 Feb 2015

Copy link to clipboard

Print this post

4.6b changes including the enhanced RTC and IR functions have been incorporated for the MX470s. Beta testers, please check your spam folders if you haven't received it.

MicroBlocks

Guru

Joined: 12/05/2012
Location: Thailand
Posts: 2209

Posted: 11:14am 19 Feb 2015

Copy link to clipboard

Print this post

I have a few observations for the 64 pin version. (I'll check the 100 pin later).

In the pin description of the 64 pin chip there is no SPI out.

The I2C on pins 31 and 32 can not be remapped to other pins. Why is the choice made to map the COM1 pins to those specific pins as it will prevent future use of the second I2C?
In particular when more devices are used, like fram, flash, rtc, expanders, motor controllers, etc a second I2C will be very useful. Preventing that for future updates seems wrong to me.
My suggestion would be to map COM1:RX to RPG7 (pin 5) and COM1:TX to RPG8 (pin 6). This will then also combine nicely with the COM1:Enable that is already mapped to RPG9 (pin 8) and it will help routing on the PCB as they are close together.

Which pins are allocated for the 4th UART, or is that peripheral used by the IR?
Which pins are allocated for the second SPI? This might be necessary to support an SD card. Otherwise it needs to be shared with the touchscreen or other usefull parts like WiFi/RF modules and other sensors.

The VBUSON pin 11. Care should be taken because this pin will be unavailable once the decision is made to support USB. I am assuming that when USB is supported it will be in the device mode. The datasheet specifically mentions that this pin is 'reserved' and under control of the USB peripheral. In other modes like host and OTG this pin is used to provide power to the VBUS under software control.

If you are using a complete port[0-15] for the databus to connect to the screen there is only one of those available on the 64 pin chip. That is port B. Port B however has the VBUSON on B5. When used for controlling the screen this will break once USB will be used.
Even with the 100 pin chip, port B will have the same problem. However port D has all 16 bits available on the 100 pin. Does this mean that the only choice will be the 100 pin or would it be possible to use port B[8-15] and port D[0-7]. In that case it will work on both 64 and 100 pin device. Would using two 'half' ports be a lot slower?

If you give me a list of which pins are used for the screens i could double check them for you to see if there are any conflicts.

Microblocks. Build with logic.

WhiteWizzard
Guru

Joined: 05/04/2013
Location: United Kingdom
Posts: 2944

Posted: 03:08pm 19 Feb 2015

Copy link to clipboard

Print this post

Nice bit of 'detective' work there TZA

IMHO these are all very good points raised which would be great to get feedback about from Peter and/or Geoff. I'm sure they have both considered the points raised but if not then this is certainly the time to try 'resolve' as many things as possible relating to pin-outs.

Guess it is best to hold-off releasing any PCBs until the pinouts become confirmed.

matherp
Guru

Joined: 11/12/2012
Location: United Kingdom
Posts: 10315

Posted: 11:04pm 19 Feb 2015

Copy link to clipboard

Print this post

Quote In the pin description of the 64 pin chip there is no SPI out.

I assume you are not on the beta program so the information you have is out-of-date. Latest version attached.

Quote The I2C on pins 31 and 32 can not be remapped to other pins. Why is the choice made to map the COM1 pins to those specific pins as it will prevent future use of the second I2C?

There are no working PPS pins left on the 64-pin (other than port-B pins). Which functions should I sacrifice to free up the 2nd I2C?

Quote Which pins are allocated for the second SPI?

See list

Quote Which pins are allocated for the 4th UART, or is that peripheral used by the IR?

There is no 4th UART. IR on the 470 uses INT0 which is a non-PPS pin

Quote The VBUSON pin 11.

Quote If you give me a list of which pins are used for the screens i could double check them

No way round this.
It is perfectly possible to use non-contiguous pins in the Cfunctions. I've already run 16-bit parallel on the 44-pin which only has 10 pins on the C-port. It is just less optimal. If you want to use the 64-pin with a 16 bit parallel and USB just wire in another pin to replace RB5 (VBUSON) in the 16-bit port.

Quote My suggestion would be to map COM1:RX to RPG7 (pin 5) and COM1:TX to RPG8 (pin 6)

Some of the mappings simply don't work. The original obvious allocation of F1 as console-RX on the 64-pin hit some sort of silicon error, then some of the logical alternatives didn't work on the 100-pin. I spent a huge amount of time finding a set that do work based on the testing I've done. We desperately need the beta program to get fully underway to really check things out. I'm not willing to make changes until there is a body of testers available. We could then allocate different hardware functions to different testers so that any change can be tested in parallel to check for unforeseen problems.

Edited by matherp 2015-02-21

MicroBlocks

Guru

Joined: 12/05/2012
Location: Thailand
Posts: 2209

Posted: 11:12pm 19 Feb 2015

Copy link to clipboard

Print this post

Unfortunately i can not beta test right now. Stuff is on its way and then i can join.
I expect chips to arrive early next week.

Is there a repository where the source code can be worked upon.
Github for example works great for this and it allows projects to be private so that not everyone can access it. I have a paid account that can hold private projects if nothing exists at the moment to get things rolling.
It will also allow changes to be committed easily and merge with the master once it is tested and approved. The dictator model is probably best. :)

Which software stack are you using? I can setup a similar environment to prevent differences in compiler versions and settings.

Edited by TZAdvantage 2015-02-21

Microblocks. Build with logic.

matherp
Guru

Joined: 11/12/2012
Location: United Kingdom
Posts: 10315

Posted: 11:59pm 19 Feb 2015

Copy link to clipboard

Print this post

Quote Is there a repository where the source code can be worked upon.

No. I don't think any sort of "group" development is appropriate, (plus I hate and despise Github

).

I obtained permission from Geoff to undertake the development and he, subject to auditing my work, has indicated it will become part of his formal "Micromite" release once the beta is successfully completed. At this point I will stop developing the code as part of the handover.

You are of course perfectly free to download the MX170 code from Geoff's site and develop it yourself subject to meeting the licence conditions.

robert.rozee
Guru

Joined: 31/12/2012
Location: New Zealand
Posts: 2442

Posted: 01:53am 20 Feb 2015

Copy link to clipboard

Print this post

xxxxxxxxxxx said Is there a repository where the source code can be worked upon.

expanding on the words of matherp above - there have been long discussions over the years on this forum regarding the availability and licensing of the micromite source code. Geoff Graham has always been extremely accommodating in making the code available on a one-to-one basis to anyone who would like to look at it, subject to a few copyright provisos to protect Geoff's own interests.

yet every now and then one of a very small group of people 'needle' the author a little over access to the mmbasic source code. never anything major, just a little reminder 'jab'. as those who have been around here for a while know, a public repository is inconsistent with the licensing of Geoff's code.

i am upset as the 'needling' is undignified, certainly unhelpful, and reflects badly on the character of those who choose to undertake it.

robert rozeeEdited by robert.rozee 2015-02-21

MicroBlocks

Guru

Joined: 12/05/2012
Location: Thailand
Posts: 2209

Posted: 02:36am 20 Feb 2015

Copy link to clipboard

Print this post

That was uncalled for, now i am upset.
When someone is beta testing some software how else would you help each other. By just loading a hex file? That will give you only 'Doesn't work'. Not the why.
What i suggested is a CLOSED repository where some experimenting can take place without disturbing someone elses experiments. In a repository you just make a new branch and have your own local copy to work with.
The dictator model allows then for ONE person to commit those changes to a master.
The dictator keeps full control, as it should be.
I understand Geoffs reasoning very well, as with open source it tends to go off in all directions and focus is lost while all the blame squarly goes back to the initiator/creator.

Geoff already freely sends the sourcecode to any one who asks. No 'needling' necessary.

Microblocks. Build with logic.

Page 3 of 4

Print this page