• AVR Freaks

Helpful ReplyHot!Is MCHP still developing the PIC32?

Page: << < ..678 Showing page 8 of 8
Author
NorthGuy
Super Member
  • Total Posts : 6214
  • Reward points : 0
  • Joined: 2014/02/23 14:23:23
  • Location: Northern Canada
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/22 13:36:18 (permalink)
3 (1)
dlindbergh
I wonder if we're going to see any new PIC32 parts?  ;-)



There are lots of PIC32MK and some PIC32CM on the Microchip feed, but nothing available yet.
 
DominusT
Super Member
  • Total Posts : 1418
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/22 19:11:16 (permalink)
0
NorthGuy
dlindbergh
I wonder if we're going to see any new PIC32 parts?  ;-)



There are lots of PIC32MK and some PIC32CM on the Microchip feed, but nothing available yet.
 


And also the PIC32MZW1 which reminds me of the PIC32WK/PIC32WM that announced it and then hid it.
hevats
Junior Member
  • Total Posts : 57
  • Reward points : 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/22 21:02:59 (permalink)
0
EngSam
questions on Embedded Linux:
Do you use beagle bone or RSpi in your products directly ?
I see these board similar to arduino in terms using them in an industrial product. I have heard how unreliable they are especially running the code from SD card and they suffer from Random REBOOT ing etc.... maybe if they make a more rugged version with higher temperature range and compliance.

We have been using beaglebone black, although we did some prototypes with RPi Compute 3 without any issues. As far as being "rugged", beaglebone black is quite a solid board. There is a industrial temperature range version available from Arrow. As I mentioned earlier we do not use SD cards. We use the beaglebone bone directly, with custom built plug-in modules. We also use QCA based custom designed boards (QCA9331 / AR9331).
 
I see the arduino comparison quite often in professional forums. IMHO some of the flack is misguided. The problems arise due to poor hardware bought cheap on ebay, aliexpress and likes. Then using libraries written by kids in school (I'm not suggesting that all of these kids write bad code). Same goes for overall design of the final product, its often half baked. If you look fundamentally, all an Arduino is an AVR CPU with a GCC toolchain, I don't see any issues with that. The board design is quite simple as well. We haven't used AVRs ourselves but I doubt they would suffer from reboot or temperature issues.
 
Stefiff
And I was very impressed by the words of the man from Microchip. He said this in front of the whole hall, literally the following.

    We know that it is difficult to use BGA, and that is why we try to make your work easier. It is not just to draw and have the circuit board made. We have personally reworked this board several times. We had it made in several different companies. Just because they will make your board does not mean that it will work.

    We had to repeat the whole process several times until we got a working board that met all the requirements.

I am not sure why the Microchip man said what he said. I personally know quite a few people at Microchip who are not at all intimidated by BGAs. Maybe he was referring to Linux more than the BGA itself, which is understandable. Microchip, and for that matter ST as well, do not have large exposure to Linux SoCs, neither do they have the scale to make their offerings cost effective. This domain is currently in the grip of Mediatek, Broadcom, QCA, TI and the likes.
Stefiff
Linux turned me down.

Maybe you didn't try hard enough :) There is only so much one can do in baremetal, after a point you need an OS.  IMHO change to an OS in embedded products is inevitable, be it Linux or others. There would be people who would change early, people would change with the tide and people who would stand their ground. I have a friend who still uses DOS based tools! Ourselves, we have been slowly shifting everything to Linux... PCs, open / free toolchains, KiCAD for PCB, Libre Office for documents, even started using FreeCAD for basic 3D and sharing of ideas.
Stefiff
The easiest way is to give you the files and you can order it somewhere in China. But that doesn't mean it will work. Not everyone can make a working board for you, even in China.

The advantage of sending files to China has mostly been price, that too has started to erode. The quality of a lot of such fabs has been questionable from the start. Likewise there are a whole bunch of fabs spread across the globe who do a decent job (US, europe including eastern europe, thailand. taiwan, malaysia, some even starting in india). Having some stake in the PCB ourselves, I would say that manufacturing quality depends a lot on good design. Multi-layer designs can get quite complex, a designer who knows the manufacturing process can avoid the common pitfalls. Like all design, PCB design is an art, just like programming. Start with the assumption that your board would be made by monkeys, better to have a monkey proof design. Or maybe the solution could be to make / buy modules for the core, then you can use a much simpler base board.
 
dlindbergh
This has become quite a thread.
 
I wonder if we're going to see any new PIC32 parts?  ;-)

Its unfair to question the wavering nerdiness of a thread while having nerdfever.com as you signature :P
wdy
Starting Member
  • Total Posts : 28
  • Reward points : 0
  • Joined: 2016/10/05 06:32:30
  • Location: 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/22 23:53:15 (permalink)
0
hevats
EngSam
questions on Embedded Linux:
Do you use beagle bone or RSpi in your products directly ?
I see these board similar to arduino in terms using them in an industrial product. I have heard how unreliable they are especially running the code from SD card and they suffer from Random REBOOT ing etc.... maybe if they make a more rugged version with higher temperature range and compliance.

We have been using beaglebone black, although we did some prototypes with RPi Compute 3 without any issues. As far as being "rugged", beaglebone black is quite a solid board. There is a industrial temperature range version available from Arrow. As I mentioned earlier we do not use SD cards. We use the beaglebone bone directly, with custom built plug-in modules. We also use QCA based custom designed boards (QCA9331 / AR9331).

Off-topic, but I'd personally suggest to steer away from the BBB. The design has problems (you can search the forums) and I even had a consulting job for a client finding issues and remedies on BBB. The client had about few hundred units installed in the field.
 
On the main thread topic - I wish we'd see more new MX and MZ parts. And most of all, the old ones with fixed errata, or at least most of it. Some issues are very basic, it is a shame to employ newbies on such important positions. On the other hand newer silicon masks (not just metal layer fixups) are costly, so not holding my breath down. But, Microchip, really - how about not throwing fresh newbies at design positions? Microchip, and sadly, most other current silicon vendors, are making the design process look more complex than it really is.
EngSam
New Member
  • Total Posts : 18
  • Reward points : 0
  • Joined: 2019/04/17 09:28:54
  • Location: 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/23 04:30:18 (permalink)
4 (1)
NorthGuy
dlindbergh
I wonder if we're going to see any new PIC32 parts?  ;-)



There are lots of PIC32MK and some PIC32CM on the Microchip feed, but nothing available yet.
 




Hopefully PIC32MK Errata gets some serious attention , Non-working UART interrupt is simply unacceptable!! 
I had high hopes for this part since it was the first PIC32 with FPU for motor control , so i can finally make my dspic33 designs rest in peace. Instead we got a chip with 58 problems including prefetch cashe although was corrected in revision 2. When Atmel was acquired including SAM E5 and E70 that can be used for motor control I thought (and still think) they would never spend much resources on this line. 
 
Off-topic : when buying the processors from digikey or mouser can you know which hardware revision they are ? ( my initial assumption (very optimistic) that when a new revision is made , microchip takes the stock from distributors and replaces it with new one) 
NorthGuy
Super Member
  • Total Posts : 6214
  • Reward points : 0
  • Joined: 2014/02/23 14:23:23
  • Location: Northern Canada
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/23 06:10:31 (permalink)
5 (1)
wdy
On the main thread topic - I wish we'd see more new MX and MZ parts. And most of all, the old ones with fixed errata, or at least most of it. Some issues are very basic, it is a shame to employ newbies on such important positions. On the other hand newer silicon masks (not just metal layer fixups) are costly, so not holding my breath down. But, Microchip, really - how about not throwing fresh newbies at design positions? Microchip, and sadly, most other current silicon vendors, are making the design process look more complex than it really is.



Modern MCU designs are written in Verilog, as software. The peripheral modules IP are purchased from other vendors (if not downloaded from opencores.org).
 
Hence little documentation for some (modules like USB or DDR2 have almost none).
 
Hence the increased complexity. It is now easy to achieve. Since the most important thing is time to market, there's no time to work on details.
 
Hence huge errata. Complex software designs have more bugs. Complex Verilog designs are no exception.
 
Hence the software-like development cycle. Bugs in older versions are not fixed. Instead, newer versions are created which add more new features and hence new bugs.
 
There's no reason to single out Microchip. Everyone does this. You will never get mature products with little or no errata (as in some PIC16s). Looks like it'll be more of that as time goes by.
 
Such are the times. Not simplicity, but complexity is sought in the designs.
 
NKurzman
A Guy on the Net
  • Total Posts : 18848
  • Reward points : 0
  • Joined: 2008/01/16 19:33:48
  • Location: 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/23 06:54:26 (permalink)
0
No when a new revision comes out, Old stock is not destroyed.
If you want a specific Rev from a distributor you would need to contact them directly and specify it. Otherwise you’ll get whatever Rev they pull out of the bin.
I believe the rev number is marked on the box.
JPortici
Super Member
  • Total Posts : 1107
  • Reward points : 0
  • Joined: 2012/11/17 06:27:45
  • Location: Grappaland
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/23 07:07:09 (permalink)
4 (1)
EngSam
Off-topic : when buying the processors from digikey or mouser can you know which hardware revision they are ? ( my initial assumption (very optimistic) that when a new revision is made , microchip takes the stock from distributors and replaces it with new one)



hahahaha no they don't.
maybe you can ask for parts with a minimum datecode, nothing else. Altough usually mouser/digkey sell enough to only have "fresh" lots.
RS will just not care and send you 5yo parts
realexander
Super Member
  • Total Posts : 235
  • Reward points : 0
  • Joined: 2006/04/08 09:50:42
  • Location: 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/23 07:10:44 (permalink)
0
NKurzman
No when a new revision comes out, Old stock is not destroyed.



I don't think so. I've gotten free sample parts that were old revs. That may be why MCHP has such a generous sampling program.
 
DominusT
Super Member
  • Total Posts : 1418
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/23 14:57:32 (permalink)
0
A possible silly question:


With so many 'silicon bugs' without knowing if there will be a solution or not, shouldn't we think about migrating to FPGAs?
dlindbergh
Super Member
  • Total Posts : 483
  • Reward points : 0
  • Joined: 2006/09/11 18:17:11
  • Location: Boston USA
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/23 15:10:15 (permalink)
5 (2)
DominusTWith so many 'silicon bugs' without knowing if there will be a solution or not, shouldn't we think about migrating to FPGAs?

 
You can if you like, but that's just saying you're going to do the job of design/debug that Microchip didn't/couldn't successfully do. Sounds like the "hard way".
 
NorthGuy
Super Member
  • Total Posts : 6214
  • Reward points : 0
  • Joined: 2014/02/23 14:23:23
  • Location: Northern Canada
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/23 15:49:20 (permalink)
4 (2)
DominusT
With so many 'silicon bugs' without knowing if there will be a solution or not, shouldn't we think about migrating to FPGAs?



Good idea of you're filthy rich.
MisterHemi
Super Member
  • Total Posts : 272
  • Reward points : 0
  • Joined: 2017/11/02 12:24:21
  • Location: Commerce, CA USA
  • Status: online
Re: Is MCHP still developing the PIC32? 2020/07/23 15:53:09 (permalink)
5 (2)
DominusT
A possible silly question:


With so many 'silicon bugs' without knowing if there will be a solution or not, shouldn't we think about migrating to FPGAs?




I'm planning to use FPGAs for things I can't do with microcontrollers or where it seems like a reasonable solution.
 
However as dlindbergh said to create a microcontroller/microprocessor in Verilog/VHDL would take some work.
There's some existing examples but you'd have to add your own peripherals and test them.
 
Not something I wouldn't consider but understand the work involved. Maybe it's a viable solution depending upon the demands of your project(s).
 
Here's one i've looked at:
www.fpga4student.com/2017/06/32-bit-pipelined-mips-processor-in-verilog-1.html
 
github.com/diadatp/mips_cpu

My configuration:
MacBook Pro (Retina, 15-inch, Mid 2015) with MacOS Mojave (10.14.6) and MPLAB X IDE v5.30
 
Curiosity PIC32MZ EF 1 & 2, PIC24F Curiosity, XPRESS EVAL BOARD (PIC16F18855), SAMA5D3 Xplained and various custom boards.
DominusT
Super Member
  • Total Posts : 1418
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/23 16:15:02 (permalink)
0
MisterHemi
DominusT
A possible silly question:


With so many 'silicon bugs' without knowing if there will be a solution or not, shouldn't we think about migrating to FPGAs?




I'm planning to use FPGAs for things I can't do with microcontrollers or where it seems like a reasonable solution.
 
However as dlindbergh said to create a microcontroller/microprocessor in Verilog/VHDL would take some work.
There's some existing examples but you'd have to add your own peripherals and test them.
 
Not something I wouldn't consider but understand the work involved. Maybe it's a viable solution depending upon the demands of your project(s).
 
Here's one i've looked at:
www.fpga4student.com/2017/06/32-bit-pipelined-mips-processor-in-verilog-1.html
 
github.com/diadatp/mips_cpu




I don't know much about FPGAs, but isn't there something like "libraries" where you can add peripherals with your "MPU"?

Obviously you could also create the ones you need or modify the ones offered by the "libraries"
MisterHemi
Super Member
  • Total Posts : 272
  • Reward points : 0
  • Joined: 2017/11/02 12:24:21
  • Location: Commerce, CA USA
  • Status: online
Re: Is MCHP still developing the PIC32? 2020/07/24 09:38:17 (permalink)
0
DominusT
MisterHemi
DominusT
A possible silly question:


With so many 'silicon bugs' without knowing if there will be a solution or not, shouldn't we think about migrating to FPGAs?




I'm planning to use FPGAs for things I can't do with microcontrollers or where it seems like a reasonable solution.
 
However as dlindbergh said to create a microcontroller/microprocessor in Verilog/VHDL would take some work.
There's some existing examples but you'd have to add your own peripherals and test them.
 
Not something I wouldn't consider but understand the work involved. Maybe it's a viable solution depending upon the demands of your project(s).
 
Here's one i've looked at:
www.fpga4student.com/2017/06/32-bit-pipelined-mips-processor-in-verilog-1.html
 
github.com/diadatp/mips_cpu




I don't know much about FPGAs, but isn't there something like "libraries" where you can add peripherals with your "MPU"?

Obviously you could also create the ones you need or modify the ones offered by the "libraries"




There are a lot of examples on the internet including some MIPS and RISC-V processors. Companies such as Xilinx have their Microblaze processor IP in Vivado for use with  their FPGAs, etc.
 
So, in a sense, there are some libraries. I'm just barely getting started with FPGAs.

My configuration:
MacBook Pro (Retina, 15-inch, Mid 2015) with MacOS Mojave (10.14.6) and MPLAB X IDE v5.30
 
Curiosity PIC32MZ EF 1 & 2, PIC24F Curiosity, XPRESS EVAL BOARD (PIC16F18855), SAMA5D3 Xplained and various custom boards.
ric
Super Member
  • Total Posts : 27953
  • Reward points : 0
  • Joined: 2003/11/07 12:41:26
  • Location: Australia, Melbourne
  • Status: online
Re: Is MCHP still developing the PIC32? 2020/07/24 16:18:48 (permalink) ☄ Helpfulby DominusT 2020/07/24 18:02:39
5 (3)
As per post #152, have you investigated the cost of an FPGA big and fast enough to hold your processor core and all the peripherals you need?
Be ready for a shock!

I also post at: PicForum
Links to useful PIC information: http://picforum.ric323.co...opic.php?f=59&t=15
NEW USERS: Posting images, links and code - workaround for restrictions.
To get a useful answer, always state which PIC you are using!
JPortici
Super Member
  • Total Posts : 1107
  • Reward points : 0
  • Joined: 2012/11/17 06:27:45
  • Location: Grappaland
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/24 21:55:41 (permalink)
4 (1)
And the cost of the PMIC, the oscillator chip, the level translator ICs, the impoved layout... YIKES!
The suitable replacement FPGA for any of my projects you cost about as much as their current BOM if not more (a 15€ FPGA will NOT be able to replace -say- the largest dsPIC, running at full capacity
wdy
Starting Member
  • Total Posts : 28
  • Reward points : 0
  • Joined: 2016/10/05 06:32:30
  • Location: 0
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/25 00:20:37 (permalink)
5 (1)
FPGAs have their place in the Universe, just not as a replacement of cheap MCUs. First, the product has to have large enough added value, so the price can be swallowed. And second, power consumption, for obvious reasons.
 
In some cases, proper MCU+FPGA combination is one of the best solutions.
 
NorthGuyModern MCU designs are written in Verilog, as software. The peripheral modules IP are purchased from other vendors (if not downloaded from opencores.org).
 
Hence little documentation for some (modules like USB or DDR2 have almost none).

Tell me, so often finding and fixing bugs in third party IPs of the sorts, for which there is no access to documentation due to licensing. :)
 
NorthGuyHence the increased complexity. It is now easy to achieve. Since the most important thing is time to market, there's no time to work on details.
 
Hence huge errata. Complex software designs have more bugs. Complex Verilog designs are no exception.
 
Hence the software-like development cycle. Bugs in older versions are not fixed. Instead, newer versions are created which add more new features and hence new bugs.
 
There's no reason to single out Microchip. Everyone does this. You will never get mature products with little or no errata (as in some PIC16s). Looks like it'll be more of that as time goes by.

What you describe is more or less the truth. Rushed everything, because engineering values and dignity is a thing of the past. And, no, not singling out Microchip - I mentioned that it is a common trend. But since this is MCHP forum, I felt the need to at least criticize them, given that sometimes employees wander around.
 
NorthGuySuch are the times. Not simplicity, but complexity is sought in the designs.

Indeed. Monetization first, everything else second (or does not matter). Add the sheer amount of people who enter the business out of purely commercial interest and with mediocre skills, and you have the recipe for modern success.
 
And back to what I was saying - Microchip need to up their game. Their IPs are simple enough to be designed and tested properly. No excuse IMHO. The idea to share the same IPs between different cores is good, but it needs proper execution.
DominusT
Super Member
  • Total Posts : 1418
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: offline
Re: Is MCHP still developing the PIC32? 2020/07/25 11:20:08 (permalink)
4 (1)
wdy
 I mentioned that it is a common trend. But since this is MCHP forum, I felt the need to at least criticize them, given that sometimes employees wander around.
 

 
LoL: LoL
 
Yes, I was finding out which are the cheapest FPGA's and I found a forum where the topic of substituting MCUs for FPGA comes up and someone indicated that an FPGA whose cost is between 10 to 12 dollars could "emulate" an 8 bit MCU in the best of cases.
 
 
 
Page: << < ..678 Showing page 8 of 8
Jump to:
© 2020 APG vNext Commercial Version 4.5