• AVR Freaks

Helpful ReplyHot!Future of PIC32

Page: 12345 > Showing page 1 of 5
Author
karpouzi
Starting Member
  • Total Posts : 29
  • Reward points : 0
  • Status: offline
2018/12/21 02:22:07 (permalink)
5 (1)

Future of PIC32

There has been very little news from Microchip on the PIC32 front for the last 20 months. Does anyone know of any plans for new devices? Are Microchip switching attention to SAM? I would like to know if the family has a future and it's worth investing efforts into it.
#1
Mysil
Super Member
  • Total Posts : 3324
  • Reward points : 0
  • Joined: 2012/07/01 04:19:50
  • Location: Norway
  • Status: offline
Re: Future of PIC32 2018/12/21 04:19:33 (permalink)
3.71 (7)
Hi,
This public forum is the wrong place if you expect an official answer or disclosure of internal information.
You would have to ask microchip sales, or development support.
 
They are a commercial company, so future products will depend on number of devices sold.
How many million chips have you bought?
I mostly buy 1 device at a time, so will hardly make any impact on the bottom line of accounting.
 
Development cycle for 32 bit microcontroller with integration to peripherals, is a lot of work and money.
Experience from errata documents, indicate that new products should be carefully designed and tested, before put into production.
 
Microchip have a track record of producing and delivering existing devices for a long time,
so I do not fear for PIC32M... chips to not be available.
 
Is there anything specific you want or need?
Then tell Microchip, directly or in this forum, what the you need.
 
Revision of existing designs to remove errata?
New features in signal processing accelerator?
More complicated peripherals, like motor control PWM and position encoder interface?
Even higher instruction frequency?
 
Every feature or property have a cost of some kind. Higher procesing frequency may cause more power consumption.
More memory will need more silicon area, and cause higher device price.
More extensive PPS routing will need more signals to be routed all around the chip, causing additional possibilities for picking up noise or interference.
Stronger pin output drive for high frequency communication signals, may cause more EMC noise.
 
What are your priority?
 
    Mysil
 
#2
DominusT
Super Member
  • Total Posts : 1292
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: offline
Re: Future of PIC32 2018/12/26 06:03:42 (permalink)
0
karpouzi
There has been very little news from Microchip on the PIC32 front for the last 20 months. Does anyone know of any plans for new devices? Are Microchip switching attention to SAM? I would like to know if the family has a future and it's worth investing efforts into it.

Maybe it's my idea, but I think MCHP is giving more impetus to the 32-bit MCUs of atmel.
#3
JPortici
Super Member
  • Total Posts : 678
  • Reward points : 0
  • Joined: 2012/11/17 06:27:45
  • Location: Grappaland
  • Status: online
Re: Future of PIC32 2018/12/28 03:13:46 (permalink)
5 (3)
Rant mode ON.
 
This is what i fear as well.
I evaluated some atmel parts in the past for projects where any small M0 or a pic32MX would suffice and i always went back to a PIC32MX documentation quality and flexibility.
The documentation for ATSAM is a joke to me. Pages and pages of doxygen style register description that leave out a LOT of information, then other pages are spend discussing things of dubious utility on detail (I was amused that how on some chips like the C21 there is a lot of detail on the power domains, saying which peripherals are powered by which VDD/VSS pair, which IOs are powered by which VDD/VSS pairs... but then written in veeeery fine printing you find that you can't power off a section of the chip or use different logic levels for IO banks. Why bother discussing it then?)
And now the same style of shitty documentation is getting on the new 8 bit parts and even on the new 16 bit reference manual chapters. FFS.
 
PIC32MZ is a wonderful chip, but at least it's mature enough that even if support is abandoned it won't be a big deal.
PIC32MZ DA has a lot of potential but shitty support, of course one will rather use something else (see the linux on pic32 threads)
PIC32MK is a potentially great chip with shitty support, because everything you can't do on your own doesn't work. Or work with a very specific configuration, like in the harmony examples, change one bit and doesn't work anymore. Granted that it really makes sense if you need the FPU and dual USB, otherwise a dsPIC will outperform it with a blink of an eye, but there is no software support for say dual USB, then no support = no sales = even less support because it's not generating enough revenue.
 
Harmony v3 is available guys, you can download it from inside MPLAB X (download MHC for V3 and the plugin will get all the stuff from the plugin repo) but for now it supports ONLY ATSAM devices!
 
Now, i understand that MIPS has lost in the microcontroller world war.. I don't really care about the core, although it would be interesting to see a PIC32R (either as in RISC or as in Cortex R) but i would love it to be a PIC32 and not an ATSAM. Because to me in the same class of device the analog peripherals were always better performing and more flexible, the digital peripherals were more, more powerful and more flexible. That's it.
 
Rant mode OFF.
#4
DominusT
Super Member
  • Total Posts : 1292
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: offline
Re: Future of PIC32 2018/12/28 05:53:38 (permalink)
0
Jack_M
Rant mode ON.
 
This is what i fear as well.
I evaluated some atmel parts in the past for projects where any small M0 or a pic32MX would suffice and i always went back to a PIC32MX documentation quality and flexibility.
The documentation for ATSAM is a joke to me. Pages and pages of doxygen style register description that leave out a LOT of information, then other pages are spend discussing things of dubious utility on detail (I was amused that how on some chips like the C21 there is a lot of detail on the power domains, saying which peripherals are powered by which VDD/VSS pair, which IOs are powered by which VDD/VSS pairs... but then written in veeeery fine printing you find that you can't power off a section of the chip or use different logic levels for IO banks. Why bother discussing it then?)
And now the same style of shitty documentation is getting on the new 8 bit parts and even on the new 16 bit reference manual chapters. FFS.
 
PIC32MZ is a wonderful chip, but at least it's mature enough that even if support is abandoned it won't be a big deal.
PIC32MZ DA has a lot of potential but shitty support, of course one will rather use something else (see the linux on pic32 threads)
PIC32MK is a potentially great chip with shitty support, because everything you can't do on your own doesn't work. Or work with a very specific configuration, like in the harmony examples, change one bit and doesn't work anymore. Granted that it really makes sense if you need the FPU and dual USB, otherwise a dsPIC will outperform it with a blink of an eye, but there is no software support for say dual USB, then no support = no sales = even less support because it's not generating enough revenue.
 
Harmony v3 is available guys, you can download it from inside MPLAB X (download MHC for V3 and the plugin will get all the stuff from the plugin repo) but for now it supports ONLY ATSAM devices!
 
Now, i understand that MIPS has lost in the microcontroller world war.. I don't really care about the core, although it would be interesting to see a PIC32R (either as in RISC or as in Cortex R) but i would love it to be a PIC32 and not an ATSAM. Because to me in the same class of device the analog peripherals were always better performing and more flexible, the digital peripherals were more, more powerful and more flexible. That's it.
 
Rant mode OFF.


The question is why MCHP chose MIPS when they entered the world of 32bit MCUs. They could have chosen ARM from the beginning and I don't think that they have only decided to use MIPS to be "different" to the majority of manufacturers.
 
#5
crosland
Super Member
  • Total Posts : 1575
  • Reward points : 0
  • Joined: 2005/05/10 10:55:05
  • Location: Bucks, UK
  • Status: offline
Re: Future of PIC32 2018/12/28 06:33:54 (permalink)
4.75 (4)
I suspect a MIPS license was way cheaper than an ARM license.
#6
andersm
Super Member
  • Total Posts : 2583
  • Reward points : 0
  • Joined: 2012/10/07 14:57:44
  • Location: 0
  • Status: online
Re: Future of PIC32 2018/12/28 07:11:49 (permalink)
5 (2)
The Cortex-M3 core would still have been quite new when Microchip started working on the first PIC32Ms, so it probably wasn't as obvious a choice as it seems in hindsight. The PIC32 family was announced in 2007. IIRC at that point there were still only two vendors with Cortex-M3 devices on the market (Luminary and ST).
post edited by andersm - 2018/12/28 07:15:58
#7
Howard Long
Super Member
  • Total Posts : 672
  • Reward points : 0
  • Joined: 2005/04/04 08:50:32
  • Status: offline
Re: Future of PIC32 2018/12/28 08:12:45 (permalink)
5 (3)
I'll add a further point here. In my experience and in practice, whether the core is ARM or MIPS makes little difference to me. The multitudes of implementation and toolchain differences between vendors means that you're not really gaining much from being tribal towards ARM or MIPS at that level.
 
The main place it makes a difference for me is when you drill down to hand tune pinch points in assembly language. When you get to that level, the differences between the various ARM Cortex cores and underlying vendor bus and memory architectures means you need to consider each one on its own merits and work on tuning them piecemeal anyway. You also have to ask how long you spend writing assembly language. Maybe a couple of days per year total?
 
I don't buy that the various CMSIS libraries are much of a benefit for ARM, even in DSP which is one of my key areas: not all of the functions have been optimised for all the target cores anyway.
 
I do agree that MIPS was an interesting choice for Microchip to make back then, I remember being a little surprised myself.
#8
JPortici
Super Member
  • Total Posts : 678
  • Reward points : 0
  • Joined: 2012/11/17 06:27:45
  • Location: Grappaland
  • Status: online
Re: Future of PIC32 2018/12/28 08:53:43 (permalink)
5 (1)
Agree 100%, as i said i don't care wether the new parts have MIPS core or an ARM core or a RISC-V Core or even a proprietary architecture, i want them to be PIC32 and not ATSAM, because of everything else that is around the core: Busses, interconnects, peripherals, documentation.
#9
NorthGuy
Super Member
  • Total Posts : 5431
  • Reward points : 0
  • Joined: 2014/02/23 14:23:23
  • Location: Northern Canada
  • Status: offline
Re: Future of PIC32 2018/12/28 09:40:33 (permalink)
0
Howard Long
... you're not really gaining much from being tribal towards ARM or MIPS at that level.



This is, of course, true, but the (sales) force has strong influence on a weak mind, thus most people wouldn't agree with your statement.
 
I think it would be interesting if Microchip extended their PIC24 into 32-bit back then (instead of licensing MIPS), but it's too late now anyway. Microchip is becoming general semiconductor company, with MCUs not being their main thing any more.
#10
marcov
Super Member
  • Total Posts : 248
  • Reward points : 0
  • Joined: 2006/10/08 01:59:40
  • Location: Eindhoven, NL.
  • Status: offline
Re: Future of PIC32 2018/12/28 09:52:49 (permalink) ☄ Helpfulby DominusT 2018/12/28 10:07:07
0
IMHO PIC32MK isn't that great. Its performance simply sucks, and its only redeeming feature, faster SPI, is now matched by dspic33c, which is much more performant.
#11
JPortici
Super Member
  • Total Posts : 678
  • Reward points : 0
  • Joined: 2012/11/17 06:27:45
  • Location: Grappaland
  • Status: online
Re: Future of PIC32 2018/12/28 11:42:24 (permalink)
0
but doesn't have usb.
I have a design which uses the 33CH plus a PIC16LF1455 as a usb <-> PC bridge with extended temperature range.
I have a project that would use two separate USBs and the MK would be perfect, if there was software support. Other offerings with two USBs aren't great either...
#12
marcov
Super Member
  • Total Posts : 248
  • Reward points : 0
  • Joined: 2006/10/08 01:59:40
  • Location: Eindhoven, NL.
  • Status: offline
Re: Future of PIC32 2018/12/29 07:02:24 (permalink)
0
I mostly use w5500 based ethernet for connection to the outside world (and CAN for simple slaves).
 
But yes, if you use USB, that is probably a factor.
 
But in my case, a 33e MU8xx already ran circles around the pic32mk.
#13
Brane313
Starting Member
  • Total Posts : 46
  • Reward points : 0
  • Joined: 2018/06/16 01:23:47
  • Location: 0
  • Status: offline
Re: Future of PIC32 2018/12/29 08:02:15 (permalink)
5 (1)
What's so wrong with PIC32MZ then ?
 
Move to ARM space doesn't make much sense. Competition there is insane, MC wouldn't make profit in that space.
 
Also, why move ? Architectures are pretty similar and at least within that space, MIPS has its decent member.
 
If anything, I'd expect them to move to RISC-V. It's quickly gaining attention and implementations.
gcc support is here, soon the Linux kernel will officially support it.
 
RISC-V branche is growing insanely fast. And it's very similar to MIPS. And Microchip is notable member of RISC-V org.
 
And Microsemi( Now Microchip) has leaning on it for their FPGAs.
 
 
 
 
 
#14
DominusT
Super Member
  • Total Posts : 1292
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: offline
Re: Future of PIC32 2018/12/29 15:43:48 (permalink)
0
Brane313
What's so wrong with PIC32MZ then ?
 
Move to ARM space doesn't make much sense. Competition there is insane, MC wouldn't make profit in that space.
 
Also, why move ? Architectures are pretty similar and at least within that space, MIPS has its decent member.
 
If anything, I'd expect them to move to RISC-V. It's quickly gaining attention and implementations.
gcc support is here, soon the Linux kernel will officially support it.
 
RISC-V branche is growing insanely fast. And it's very similar to MIPS. And Microchip is notable member of RISC-V org.
 
And Microsemi( Now Microchip) has leaning on it for their FPGAs.
 

Possibly the sales of Atmel MCUs are higher, which forced MCHP to give it more attention.
#15
LostInSpace
Super Member
  • Total Posts : 245
  • Reward points : 0
  • Joined: 2016/03/11 22:47:59
  • Location: 0
  • Status: offline
Re: Future of PIC32 2018/12/29 17:27:26 (permalink)
4 (1)
>>>>Move to ARM space doesn't make much sense.
 
Well, with the Atmel + MicroSemi purchase, Microchip is in the ARM space now.
 
>>>>Competition there is insane, MC wouldn't make profit in that space.
True, that's how they bought Atmel - Atmel never made much money at all, finally gave up.
 
However - In economic theory, no matter what the commodity market - there is always a single "low cost producer" of the commodity. The low cost producer makes money. That is Microchips aim, to become the low cost producer. They probably are already.
 
#16
Brane313
Starting Member
  • Total Posts : 46
  • Reward points : 0
  • Joined: 2018/06/16 01:23:47
  • Location: 0
  • Status: offline
Re: Future of PIC32 2018/12/29 18:43:43 (permalink)
0
I don't think so. NXP is very aggresive there, as well as STM.
#17
DominusT
Super Member
  • Total Posts : 1292
  • Reward points : 0
  • Joined: 2005/07/22 08:31:18
  • Status: offline
Re: Future of PIC32 2018/12/30 06:14:42 (permalink)
0
As a future product there are only ATSAM type MCUs, possibly that is the end of the PIC32
#18
NorthGuy
Super Member
  • Total Posts : 5431
  • Reward points : 0
  • Joined: 2014/02/23 14:23:23
  • Location: Northern Canada
  • Status: offline
Re: Future of PIC32 2018/12/30 07:29:58 (permalink)
5 (1)
DominusT
As a future product there are only ATSAM type MCUs, possibly that is the end of the PIC32



There's a chance they may design RISC-V based PIC32s. RISC-V might be the next ARM.
 
#19
marcov
Super Member
  • Total Posts : 248
  • Reward points : 0
  • Joined: 2006/10/08 01:59:40
  • Location: Eindhoven, NL.
  • Status: offline
Re: Future of PIC32 2018/12/31 04:31:55 (permalink)
0
Brane313
What's so wrong with PIC32MZ then ?

 
No hardware QEI.
 
#20
Page: 12345 > Showing page 1 of 5
Jump to:
© 2019 APG vNext Commercial Version 4.5