 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : port(21,2) not simultaneous on picomite?
Page 1 of 2   |
|||||
| Author | Message | ||||
crez Senior Member  Joined: 24/10/2012 Location: AustraliaPosts: 152 |
 setpin 21,dout setpin 22,dout do port(21,2) = 1 port(21,2) = 2 loop 1k resistor from pin 21 to point A 2k resistor from pin 22 to point A I expected to see pulse changing from 1/3 to 2/3 vcc. Waveform at point A shows brief (around 1 microsecond) periods when both outputs are high and also both outputs low. Waveform also shows occasional longer periods (around 15 microseconds) when both outputs go low. It looks like the pins are updated sequentially, and occasionally an interrupt separates the two updates. |
||||
| matherp Guru  Joined: 11/12/2012 Location: United KingdomPosts: 10310 |
Nor on any other version of MMBasic |
||||
phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2640 |
"Waveform also shows occasional longer periods (around 15 microseconds)" Have also seen this when receiving a data stream. Rising edges sometimes recorded 5 to 15uS later than when they actually happened. One possibility is the USB interface requiring attention occasionally. In that instance, with a little arithmetic, just the falling edges could be used to get clean data. . Edited 2022-06-06 08:12 by phil99 |
||||
bigmik Guru Joined: 20/06/2011 Location: AustraliaPosts: 2950 |
Crez, What happens if you put a small capacitor at point A to GND? I would experiment with 10nf and Increase or decrease till you get the effect you wish (or it totally fails and you give up) Regards, Mick Mick's uMite Stuff can be found >>> HERE (Kindly hosted by Dontronics) <<< |
||||
| crez Senior Member Joined: 24/10/2012 Location: AustraliaPosts: 152 |
I have done a bit of machine code programming in the past and could either change a port one bit at a time or do a parallel load, which changed 8 bits simultaneously. I thought the port() command would be like the latter. Mick, you are right, a capacitor would do the trick, and a few extra microseconds delay would not hurt at all. Without it the 3v spikes would cause false triggers on the device monitoring this output. |
||||
Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 7937 |
I think, if you have a look at the RP2040 Datasheet, that the RP2040 has one single GPIO port which is 32 bits wide. I don't see how it can be accessed as actual physical ports at all (unless it's all the bits at once), unlike for example a PIC that may have ports A, B C etc. All versions of MMBasic (AFAIK) use a system of "virtual" ports which is more flexible than the pysical ports but has the problem noted here. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Volhout Guru Joined: 05/03/2018 Location: NetherlandsPosts: 5089 |
Hi Mick, The PIC has 8 bit wide registers. The IO pins are grouped into groups of 8. and for each group of 8 there is an 8 bit DATA register, and an 8 bit DATA DIRECTION register (and in more modern PIC's there is an 8ndividual 8 bit pullup register, and 8 bit DATA-IN register etc..). But in essence, and byte write to a single register can set 8 databits at the same time. The RP2040 (and many ARM based chips I have dealt with in the recent past) have PIN registers. So each individual pin has it's own input register, and it's own output register. So unless you attach the pin to a alternate function (read: hardware inside the chip), software individually writes to each pin. And there will be time differencen between behaviour of individual pins. Appart from that: The PORT command allows you to map ranges of pins to a port. These ranges can skip pins (port 2,2,16,4 is completely valid), and can also spread over different (PIC style) ports. So it s inevitable that the port command shows small delays between pins that toggle (and pins that are read). PIO In the specific shown case (2 pins) the PIO can help solve the problem. A small PIO program can parallel write i.e. 5 pins, and can be controlled from a PIO WRITE. HARDWARE In many cases the pins do not drive H bridges of similar directly. Hardware (a latch function) can also solve the problem and make GPIO changes simultaneous. PicomiteVGA PETSCII ROBOTS |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 7937 |
I've played with PICs a few times in the past. :) One of my favourite things was the interrupt on change facility, to issue an interrupt if any of the bits in a port changes status. Very neat. The larger chips also support parallel IO using ports, also very neat. You do get the port clashes though, where the pins you have to use for something else prevent you from using a port. :( The venerable (and still made, but more up to date now) Z80 had special command lines for port I/O, allowing 256 8-bit ports that took up no memory space. Very useful. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Tinine Guru Joined: 30/03/2016 Location: United KingdomPosts: 1646 |
For the PIC32MX170 + an interactive (dialect of) BASIC. This is where ByPic shines. Direct access to the "metal". Craig |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 7937 |
Pity it's no longer supported. :( Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Tinine Guru Joined: 30/03/2016 Location: United KingdomPosts: 1646 |
How come the concept of a finished, robust, working product doesn't work in our world? I just write code and [shock-horror] it does what it's supposed to  "Yeah, your new car looks awesome but what's with the square wheels?" "Well that's why I bought it...because it's still under development" Craig |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 10310 |
crez I think I've found a way of doing simultaneous update. Are you using normal version or VGA. Let me know and I'll post a test version. If anyone else wants to test let me know which version |
||||
| Michal Senior Member Joined: 02/02/2022 Location: PolandPosts: 125 |
I would like a PicoMite version Michal |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 10310 |
Try this and let me know PicoMite.zip |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 7937 |
Because it's not within the capacity of most people to know that something genuinely is finished and has no hidden errors. If code is no longer supported and a certain obscure combination of data or sequence of instructions after the device has been running for a certain time then you have no-one to fall back on to sort out the problem. That sort of code is fine where there is no risk as failures just mean pressing the reset button, but it shouldn't be used otherwise. If you have the source code and can debug it yourself then that's a different matter - you can fix it. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Tinine Guru Joined: 30/03/2016 Location: United KingdomPosts: 1646 |
This excuse has been used since forever but in reality, if one had the skills to do such a thing, they wouldn't be using such a HLL in the first place. The other reality is that there are endless ways to approach a coding problem...just take another approach. There is a comprehensive collection of working-code samples for the ByPic and because it deals directly with the hardware, one only needs to use the MX170 datasheet to understand the registers and one can accomplish pretty-much anything.  Craig |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 7937 |
It's true that having the source code doesn't automatically keep you safe, but it does mean that *someone* can do some debugging. If you can't get the source then it's highly unlikely that anyone can ever fix it so the bug will lurk forever. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 10310 |
Sorry to have a winge but when I post test software specifically to address the issue raised by the OP it would be helpful if the relevant posts weren't swamped by off-topic discussions Edited 2022-06-08 21:59 by matherp |
||||
| Volhout Guru Joined: 05/03/2018 Location: NetherlandsPosts: 5089 |
Just tried it. Have not been able to detect even 1ns difference when running the OP test program. This is marvelous !!! Can we have that please in all pico versions (also the VGA version). Thanks a lot for making this possible. It opens a whole new area of possibilities. Volhout PicomiteVGA PETSCII ROBOTS |
||||
| crez Senior Member Joined: 24/10/2012 Location: AustraliaPosts: 152 |
Thanks Peter, I've never been referred to as the OP before  .The Picomite version suits me. I will get a chance to try it on the weekend. |
||||
| Page 1 of 2 |
|||||
 Print this page |
 |
The Back Shed's forum code is written, and hosted, in Australia. | © JAQ Software 2025 |
