• AVR Freaks

Helpful ReplyHot!Is MCHP still developing the PIC32?

Page: << < ..678910 > Showing page 9 of 10
Author
jdeguire
Super Member
  • Total Posts : 599
  • Reward points : 0
  • Joined: 2012/01/13 07:48:44
  • Location: United States
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/08/06 10:21:28 (permalink)
1 (1)
Was anyone else expecting to read "CAN-FD does not function" in the errata for the part with "CAN-FD" in the title, or was that just me?
JPortici
Super Member
  • Total Posts : 1175
  • Reward points : 0
  • Joined: 2012/11/17 06:27:45
  • Location: Grappaland
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/08/06 10:23:58 (permalink)
5 (1)
Quite :D
DominusT
Super Member
  • Total Posts : 1456
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: online
Re: Is MCHP still developing the PIC32? 2020/08/09 09:24:22 (permalink)
4.5 (2)
jdeguire
Was anyone else expecting to read "CAN-FD does not function" in the errata for the part with "CAN-FD" in the title, or was that just me?


Only once did I make the mistake of trying to create a product, with a new MCU. I think that everything new that MCHP produces, you have to wait at least a year for others to experiment or design a prototype to determine how real the "wonderful virtues" are.
wdy
Starting Member
  • Total Posts : 34
  • Reward points : 0
  • Joined: 2016/10/05 06:32:30
  • Location: 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/08/10 05:32:29 (permalink)
5 (2)
jdeguire
Was anyone else expecting to read "CAN-FD does not function" in the errata for the part with "CAN-FD" in the title, or was that just me?

My favorite is how almost all MCUs targeted at low power operation have critical errata connected with going to or exiting sleep. Or simply consume too much power anyway.
The XLP series is no exception - boldly goes with the flow.
LdB_ECM
Super Member
  • Total Posts : 454
  • Reward points : 0
  • Joined: 2019/04/16 22:01:25
  • Location: 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/08/10 07:39:11 (permalink)
2 (1)
I see this thread is still going .. so perhaps it's time someone explained what really limits where Microchip can go commercially
 
This is the 2018 ALLWINNER CPU V3S reference designs ... cost approx $7 USD
https://github.com/petit-miner/Blueberry-PI
Specs on front page says it all except the support
 
Mainline 4.x Linux support, FreeRTOS support and obviously stock standard GCC compiler/assembler for baremetal
 
Commercially what many of us are doing is hanging SPI ADC's and DAC's off it to go into the embedded market.
 
DominusT
Super Member
  • Total Posts : 1456
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: online
Re: Is MCHP still developing the PIC32? 2020/08/10 08:13:14 (permalink)
4.67 (3)
LdB_ECM
I see this thread is still going .. so perhaps it's time someone explained what really limits where Microchip can go commercially
 
This is the 2018 ALLWINNER CPU V3S reference designs ... cost approx $7 USD
https://github.com/petit-miner/Blueberry-PI
Specs on front page says it all except the support
 
Mainline 4.x Linux support, FreeRTOS support and obviously stock standard GCC compiler/assembler for baremetal
 
Commercially what many of us are doing is hanging SPI ADC's and DAC's off it to go into the embedded market.
 


In my previous job, we had a product that worked very well with the PIC32MX, the MCU managed an LCD screen, a QTouch keyboard and communication with a server via Ethernet.
 
Another development section proposed to use the raspberrypi to design a similar product, but with a TFT touch screen and with Linux.
 
Apparently it was an evolution of our previous product, it worked very well, it had a sleek design and it looked much more sophisticated than the previous version.
 
But about 4 months after installing the first beta version products, the problems began. It stopped working, it crashed, the operating system had to be reinstalled and after investigating what happened, it was discovered that the temperature at which the raspberry elements operated plus the ambient temperature (tropical zone) was made to abandon the idea to continue with the project.

They tried to put heat sinks on the elements and fans, but in the end the same thing happened and it started to cost more than the original product.

Maybe it was not designed correctly or maybe these kinds of devices are for educational projects, not recommended for industrial environments.

Someone recommended us that if we want something similar to raspberrypi and similar but for rough environments, it can be Digi's ConnectCore
 
https://www.digi.com/products/browse/connectcore
LdB_ECM
Super Member
  • Total Posts : 454
  • Reward points : 0
  • Joined: 2019/04/16 22:01:25
  • Location: 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/08/10 09:01:40 (permalink)
4 (1)
The Pi foundation is a not for profit charity organization building SOC's for educational single board computers .. you would use them in an industrial design at your own peril :-)
https://www.raspberrypi.org/about/
 
That is a little different to using a commercial SOC on a Pi footprint so you can play with Pi Hats and peripheral market. We aren't talking about buying a fixed board here but laying up your own design just like you would a PIC32.
post edited by LdB_ECM - 2020/08/10 09:05:28
DominusT
Super Member
  • Total Posts : 1456
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: online
Re: Is MCHP still developing the PIC32? 2020/08/10 09:26:01 (permalink)
5 (1)
LdB_ECM
The Pi foundation is a not for profit charity organization building SOC's for educational single board computers .. you would use them in an industrial design at your own peril :-)
https://www.raspberrypi.org/about/
 
That is a little different to using a commercial SOC on a Pi footprint so you can play with Pi Hats and peripheral market. We aren't talking about buying a fixed board here but laying up your own design just like you would a PIC32.


Good observation.

What I comment happened several years ago when raspberry pi and computational modules just appeared and presented it as something wonderful and extraordinary. Then other similar products appeared.

I think it is possible and a good idea to use an operating system, but in other applications I don't. Someone wrote in this forum that he doesn't understand why there are obsessed people who want to put Linux in everything, even in the temperature control of a coffee maker if possible.
dan1138
Super Member
  • Total Posts : 3917
  • Reward points : 0
  • Joined: 2007/02/21 23:04:16
  • Location: 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/08/10 09:54:03 (permalink)
2 (1)
DominusT
... Someone wrote in this forum that he doesn't understand why there are obsessed people who want to put Linux in everything, even in the temperature control of a coffee maker if possible. ...

Such a narrow minded attitude would deprive the future of products like the Talkie Toaster. :)
DominusT
Super Member
  • Total Posts : 1456
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: online
Re: Is MCHP still developing the PIC32? 2020/08/10 09:58:49 (permalink)
5 (1)
dan1138
DominusT
... Someone wrote in this forum that he doesn't understand why there are obsessed people who want to put Linux in everything, even in the temperature control of a coffee maker if possible. ...

Such a narrow minded attitude would deprive the future of products like the Talkie Toaster. :)


As long as it is not cheap and easy to do that, I think it is not necessary yet. It would be as much as claiming that those who don't want to migrate to an FPGA from an MCU are also narrow minded
NorthGuy
Super Member
  • Total Posts : 6350
  • Reward points : 0
  • Joined: 2014/02/23 14:23:23
  • Location: Northern Canada
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/08/10 13:34:57 (permalink)
5 (5)
DominusT
As long as it is not cheap and easy to do that, I think it is not necessary yet. It would be as much as claiming that those who don't want to migrate to an FPGA from an MCU are also narrow minded



I don't think there's uni-directional migration path leading somewhere.
 
A guy who thinks that everything must be done with PIC16 is equally narrow minded as the guy who think that everything must be done with Raspberry Pi.
 
Just now, not specially preparing, I look at my table now and I see two PIC16s, PIC24, ESP-32-S2, RPi with ARM and 64-bit Linux. I use them all, each for different purposes.
DominusT
Super Member
  • Total Posts : 1456
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: online
Re: Is MCHP still developing the PIC32? 2020/08/10 13:53:16 (permalink)
0
NorthGuy
DominusT
As long as it is not cheap and easy to do that, I think it is not necessary yet. It would be as much as claiming that those who don't want to migrate to an FPGA from an MCU are also narrow minded



I don't think there's uni-directional migration path leading somewhere.
 
A guy who thinks that everything must be done with PIC16 is equally narrow minded as the guy who think that everything must be done with Raspberry Pi.
 
Just now, not specially preparing, I look at my table now and I see two PIC16s, PIC24, ESP-32-S2, RPi with ARM and 64-bit Linux. I use them all, each for different purposes.


I agree with that idea, what I am trying to say is that no matter what you make your product with, the important thing is to sell a product with the best possible quality, with the best stability and the cheapest to make a profit. Obviously those three characteristics cannot be fulfilled exactly, but I think that you have to make the best effort.
pboddie
Junior Member
  • Total Posts : 34
  • Reward points : 0
  • Joined: 2017/05/09 06:22:45
  • Location: 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/08/10 13:54:09 (permalink)
5 (1)
DominusT
What I comment happened several years ago when raspberry pi and computational modules just appeared and presented it as something wonderful and extraordinary. Then other similar products appeared.



The Compute Module was an opportunistic way for the Raspberry Pi people to offload yet more Broadcom SoCs that they presumably have as surplus due to product lifespan requirements. Arguably, a repackaging of the functionality in a fairly familiar profile is a good thing and makes it easier to design products around the RPi "platform".
 
But like practically all of the RPi stuff, if you use their stuff, you are completely dependent on one single vendor who might not even be able to supply replacements, spares, new parts for long periods of time (considering the availability of certain products like the Pi Zero), and whose product architecture is also largely proprietary, with even the bits considered notionally "open" like the instruction set architecture (which is not really open as such, but anyway) being rather esoteric in the sense that the ISA variant is rather out of step with other common SoCs.
 
Designing around the Compute Module makes any hardware producer virtually some kind of niche sub-project of the Raspberry Pi effort and completely at their mercy. But there will always be people who like to take risks, I suppose.
pboddie
Junior Member
  • Total Posts : 34
  • Reward points : 0
  • Joined: 2017/05/09 06:22:45
  • Location: 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/08/10 14:07:48 (permalink)
4.5 (2)
DominusT
It would be as much as claiming that those who don't want to migrate to an FPGA from an MCU are also narrow minded



Where FPGAs are tempting for me (as someone who isn't designing hardware for a living) is that they potentially provide the level of control that microcontrollers or SoCs are unable to guarantee, either due to architectural constraints in the latter kinds of products or due to vague or evasive documentation about how those products work.
 
So, in my experience of doing things with the PIC32MX that I probably shouldn't have been doing, it was a case of trying to get behavioural guarantees out of the documentation, measuring or observing the product behaviour, and then thinking that for my particular situation it might ultimately be easier to break out the Verilog and just specify how the hardware should be doing things.
 
But I can easily see that for some situations the FPGA would be augmenting the microcontroller or SoC, and I have seen projects involving, say, a Raspberry Pi Zero where a FPGA or CPLD is needed as necessary "glue" to share out the functionality appropriately.
 
This is where Microchip and others let themselves down: if the peripherals were just a bit better (and better documented) then programmable logic really would be out of the question. Instead, people can easily be driven to programmable logic out of frustration.
NorthGuy
Super Member
  • Total Posts : 6350
  • Reward points : 0
  • Joined: 2014/02/23 14:23:23
  • Location: Northern Canada
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/08/10 19:57:45 (permalink)
5 (4)
pboddie
But I can easily see that for some situations the FPGA would be augmenting the microcontroller or SoC, and I have seen projects involving, say, a Raspberry Pi Zero where a FPGA or CPLD is needed as necessary "glue" to share out the functionality appropriately.

 
Zynq would probably be better than such monstrosity.
 
pboddie
This is where Microchip and others let themselves down: if the peripherals were just a bit better (and better documented) then programmable logic really would be out of the question. Instead, people can easily be driven to programmable logic out of frustration.



I think Microchip peripherals are one of the best. They work well for many situations and are relatively well documented. Whatever you do, you  always will encounter some limitations and restrictions. This is unavoidable. If you make peripheral too complex by trying to cover all the possible use cases, you're making the peripheral difficult to use at the same time. It's important to strike a balance.
 
pboddie
Junior Member
  • Total Posts : 34
  • Reward points : 0
  • Joined: 2017/05/09 06:22:45
  • Location: 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/08/11 10:33:20 (permalink)
4.5 (2)
NorthGuy
pboddie
But I can easily see that for some situations the FPGA would be augmenting the microcontroller or SoC, and I have seen projects involving, say, a Raspberry Pi Zero where a FPGA or CPLD is needed as necessary "glue" to share out the functionality appropriately.

 
Zynq would probably be better than such monstrosity.

 
I don't really know anything about Zynq. From what I understand, it is a product that combines an ARM core with an FPGA. But it all depends on what people are trying to achieve. In one project I am thinking of, the need was to generate HDMI output from the RGB signal of old microcomputers and to do so relatively cheaply.
 
That favoured the use of very cheap and "available" Raspberry Pi boards because they provide HDMI already. Going for a full-on FPGA solution would need to involve building in HDMI support and then having to go and pay for the right to put it in a product. Instead, the RPi is just a component that provides HDMI ready to go and that people can obtain themselves. (HDMI is one of those technologies with a licensing cartel and vendors playing "secret squirrel" with their implementations.)
 
But in such a project there was still a need for some kind of programmable logic because the RPi, even when running a bare-metal environment, just cannot do the signal processing in a timely enough fashion. In other words, cost and technical feasibility were the driving factors. I doubt that any kind of Zynq board would deliver the same thing at a comparable cost.
 
NorthGuy 
pboddie
This is where Microchip and others let themselves down: if the peripherals were just a bit better (and better documented) then programmable logic really would be out of the question. Instead, people can easily be driven to programmable logic out of frustration.



I think Microchip peripherals are one of the best. They work well for many situations and are relatively well documented. Whatever you do, you  always will encounter some limitations and restrictions. This is unavoidable. If you make peripheral too complex by trying to cover all the possible use cases, you're making the peripheral difficult to use at the same time. It's important to strike a balance.

 
I agree that the design of these things is all about compromise, but given that these products must have some kind of well-defined behaviour, it is infuriating not to see it described very well. My ill-advised experiments had me looking at how the DMA mechanism worked, and one of the regulars here practically told me that what I had actually demonstrated was impossible, or words to that effect.
 
What that showed was that either the hardware is not behaving as it should (not particularly likely, despite remarks about errata) or that people do not have a complete understanding of how it works (much more likely given how vague the documentation can be). Ultimately, one is left with the impression of being powerless to either influence the behaviour of the hardware or to obtain usable insights into how it works. Every time you need to play guessing games about what the product is doing, you lose confidence in it, and that only encourages people to look at other options.
DominusT
Super Member
  • Total Posts : 1456
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: online
Re: Is MCHP still developing the PIC32? 2020/10/13 06:04:24 (permalink)
kseg
Starting Member
  • Total Posts : 66
  • Reward points : 0
  • Joined: 2016/05/26 07:30:47
  • Location: 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/10/13 07:37:01 (permalink)
0
Great.
Does it mean PIC32MZ not gone?
DominusT
Super Member
  • Total Posts : 1456
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: online
Re: Is MCHP still developing the PIC32? 2020/10/13 08:21:12 (permalink)
0
kseg
Great.
Does it mean PIC32MZ not gone?


I thought there was going to be a hybrid between PIC32 peripherals with the ARM core, but maybe it will continue with the MIPS core.
Howard Long
Super Member
  • Total Posts : 836
  • Reward points : 0
  • Joined: 2005/04/04 08:50:32
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/10/13 12:43:48 (permalink)
0
DominusT
https://www.microchip.com/design-centers/wireless-connectivity/embedded-wi-fi/pic32mz-w1
 


That product's been in gestation for about four years! I thought it was vapourware.
Page: << < ..678910 > Showing page 9 of 10
Jump to:
© 2020 APG vNext Commercial Version 4.5