Author Topic: Good cheap uC with low-jitter USB HID capability?  (Read 13709 times)

0 Members and 1 Guest are viewing this topic.

Offline lpc32Topic starter

  • Frequent Contributor
  • **
  • Posts: 454
Good cheap uC with low-jitter USB HID capability?
« on: May 01, 2014, 08:18:55 pm »
What are some good cheap uCs for implementing a low-jitter USB HID device? I don't need a very fast processor, but low jitter is important (if that's an issue at all).

Would USB 2 High Speed help in assuring better worst case jitter?

On uCs with an integrated USB controller, how do you use it in practice? Minimally just connecting the chip directly to a USB connector?
 

Offline electr_peter

  • Supporter
  • ****
  • Posts: 1300
  • Country: lt
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #1 on: May 01, 2014, 08:29:51 pm »
Can you give a bit more detail about your application? Why the jitter is important for you?
I am not sure if this is more accurate representation of your question: "which µC can maintain reliable USB 2.0 connectivity with internal oscillator only (without external crystal)?"
 

Offline lpc32Topic starter

  • Frequent Contributor
  • **
  • Posts: 454
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #2 on: May 01, 2014, 08:36:57 pm »
I want to send sensor data to PC, with report latency consistent to < 1ms. I don't know how consistent Windows would be, but at least I can remove variability from the device side.

Jitter wouldn't matter if I could somehow synchronize the PC and device clocks, but I don't see how I can do that.
 

Offline electr_peter

  • Supporter
  • ****
  • Posts: 1300
  • Country: lt
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #3 on: May 01, 2014, 09:03:45 pm »
Jitter doesn't matter at all for low latency applications (as long as communication works).

For <1ms latency USB as a protocol is not suited well (if we are talking about random data burst, not continuous transmission). This is mainly to due to USB protocol specifics (not real time by any means).
I can't tell exactly what kind of latency you could expect, but I doubt that <1ms is doable. You could try to do that and test performance - you should mainly work to setup both µC and PC sides to communicate as fast as possible. Windows, for example, hardly allows "direct metal" access to USB or USB serial ports - these have internal time outs, buffers to fill in, coffee breaks, etc. that you cannot control with <1ms.

I would suggest to do everything critical in µC (maybe add some RAM for temporal storage) then send all data for PC - this way you can control latency in critical areas.
 

Offline lpc32Topic starter

  • Frequent Contributor
  • **
  • Posts: 454
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #4 on: May 01, 2014, 09:10:17 pm »
Latency doesn't matter as much. It's consistency I'm after: I need the relative timing to be as consistent as possible so that I can timestamp the data.

But what USB-capable controllers are a good choice, also in terms of hardware simplicity?


 

Offline electr_peter

  • Supporter
  • ****
  • Posts: 1300
  • Country: lt
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #5 on: May 01, 2014, 09:37:09 pm »
Latency doesn't matter as much. It's consistency I'm after: I need the relative timing to be as consistent as possible so that I can timestamp the data.

But what USB-capable controllers are a good choice, also in terms of hardware simplicity?

If you like to have consistent delivery of data, you should match send data size with buffer size of receiving end (PC). It depends on your data size and reconfigurability of PC. See the links, they explain importance of matching data size to buffer size (you may need to send "fake" 0 data just to fill in the buffer).
http://www.datatranslation.com/phpkb/question.php?ID=627
http://www.digi.com/support/kbase/kbaseresultdetl?id=3418

In terms of devices, I would suggest to look at Maxim MAX232 or FTDI line of serial to USB converters. Very simple to use and configure (there are fake ones on eBay, beware).
If you have USB controller build in µC, it should suffice.

Other option would be to timestamp data at µC side and add timing info to data stream.
 

Offline elgonzo

  • Supporter
  • ****
  • Posts: 688
  • Country: 00
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #6 on: May 01, 2014, 09:48:48 pm »
If you need to timestamp your data with millisecond accuracy or better, do it on the MCU side and send the data together with the time-stamps through USB.

Depending on your application scenario the MCU choice depends on more factors than just having USB support in hardware.
One such factor certainly is driver support, where FTDI is rather good.

But you might also need to consider possible license restrictions when using 3rd-party USB libraries (even if they are from the MCU manufacturer) and how they would apply to your case. As an example, look at DangerousPrototypes' BusPirate version 4 which  is stuck in hell and probably will never be finished properly. As Ian Lesnet from DangerousPrototypes fame recently stated: "I'm frankly sick to death of dealing with USB on PICs." Reason behind this is to a great deal Microchips unfriendly licensing terms regarding their PIC USB libraries, which sort of forced the PIC community to spend lots of efforts in developing their own USB libs with varying success...
« Last Edit: May 01, 2014, 09:55:03 pm by elgonzo »
 

Offline lpc32Topic starter

  • Frequent Contributor
  • **
  • Posts: 454
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #7 on: May 01, 2014, 10:05:51 pm »
Thanks for the links, Peter, I'll have a look.

I need the timestamps to be PC-relative, not device-relative, that's the problem. MAX232/FTDI would be CDC, I suppose, so no latency guarantees, unlike HID's Interrupt transfers. HID is also driverless, which is a plus. And I want the USB controller and processor to be the same chip.

My thinking is, if USB2 High Speed Interrupt transfers are nominally guaranteed to 125us, that's the most likely way to get low jitter. But I'll go with USB1's 1ms if that means a much cheaper uC, or something that's easier to work with, mostly hardware-wise, but also software.
« Last Edit: May 01, 2014, 10:10:06 pm by lpc32 »
 

Offline elgonzo

  • Supporter
  • ****
  • Posts: 688
  • Country: 00
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #8 on: May 01, 2014, 10:16:44 pm »
I need the timestamps to be PC-relative, not device-relative, that's the problem.

What is the problem? When you initialize your USB device for operation let the PC send a time information (essentially from the PC's clock) to the MCU, and the MCU adjust its clock or sets up a timer with the time -- voila, now both the MCU and the PC have the same time base. It seems to be possible to do control transfers to such a HID device in a comparatively(!) easy way with the help of libraries such as libUSB.

This is way, you will also avoid any kind of variations occurring during time taking on PC, which will very likely happen unless you use a real-time OS.

If you worry about the few microseconds (or perhaps rather milliseconds) difference between the clocks of the PC and the MCU (because of the time it takes to send the time info through USB and to setup the MCU clock), well, don't worry. Your PC clock is most likely anything but accurate in relation to any official time reference anyway, so a few milliseconds more divergence does not really matter...
« Last Edit: May 01, 2014, 10:47:08 pm by elgonzo »
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 3781
  • Country: de
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #9 on: May 01, 2014, 10:36:41 pm »
Thanks for the links, Peter, I'll have a look.

I need the timestamps to be PC-relative, not device-relative, that's the problem. MAX232/FTDI would be CDC, I suppose, so no latency guarantees, unlike HID's Interrupt transfers. HID is also driverless, which is a plus. And I want the USB controller and processor to be the same chip.


Ehm, CDC is also interrupt transfer. USB interrupt transfer has nothing to do with interrupts in the MCU sense. It just means that the USB host polls your device periodically and services the requests one by one.

The other types are bulk (set up transfer once and send a large amount of data over several data requests) and isochronous (set up once and then timing has to be followed - used for e.g. streaming audio).

Looking at the description of your application, I would say you need the isochronous transfer. But that is not possible to do without custom driver.

My thinking is, if USB2 High Speed Interrupt transfers are nominally guaranteed to 125us, that's the most likely way to get low jitter. But I'll go with USB1's 1ms if that means a much cheaper uC, or something that's easier to work with, mostly hardware-wise, but also software.

What matters more is the total roundtrip time. If you are using interrupt transfers, then each has to be set up first. That gives you some ~10ms end-to-end latency at best, regardless of whether you are doing USB 1 or USB 2.0 - the time is going to be completely dominated by the transfer setup overhead. E.g. in Windows the task scheduler runs every 10ms, so your driver likely won't be scheduled to set up an interrupt transfer every 1ms ...


 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 3781
  • Country: de
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #10 on: May 01, 2014, 10:53:59 pm »
Have a look here, these two pages describe what is what in USB:

http://www.usbmadesimple.co.uk/
http://www.beyondlogic.org/usbnutshell/usb1.shtml

That is for USB 1.x, USB 2.0 is quite a bit more complex and unless you need the extra bandwidth (transferring LOTS of data), don't bother. Chips that can handle that are a few and far between and the board design starts to be nontrivial (the signal integrity at 480Mbps is quite a big deal). If you only need to send few bytes of data with a predictable latency, then full speed USB 1.x and isochronous transfer will do the job just fine.

« Last Edit: May 01, 2014, 10:56:23 pm by janoc »
 

Offline Galaxyrise

  • Frequent Contributor
  • **
  • Posts: 531
  • Country: us
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #11 on: May 01, 2014, 11:03:08 pm »
I need the timestamps to be PC-relative, not device-relative, that's the problem.
Add a one-time computation when the PC receives the first event, computing the mapping input time x to PC time y.  You're not going to get a better source of relative timings than the MCU, so you should preserve that information.  Depending on your setup, the PC can handle the device rebooting by recomputing the offset if MCU time ever goes backwards.
I am but an egg
 

Offline lpc32Topic starter

  • Frequent Contributor
  • **
  • Posts: 454
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #12 on: May 02, 2014, 12:08:14 am »
elgonzo: The issue is indeed the few ms due to the transfer. If jitter is low (admittedly, particularly from the Windows side) I might be able to account for the roundtrip and sync the clocks more or less: a few pings, throwing out high outliers, maybe minding both the average and minimum. Ultimately the timestamping I need is the PC time. I don't care about world time.

janoc: CDC data is Bulk and possibly Isochronous. Or do you mean the control endpoint(s)? Iso is allowed to drop data, and I think is more likely to add jitter due to buffering (assuming it just needs to guarantee a semi-local data rate rather than max latency per transfer). I don't need to transfer data most of the time, but when I do it's just a few bytes and the latency needs to be as consistent as possible. HID is both Interrupt and driverless in Windows, so seems the most suitable. (Though I think continuous but minimal data should work just as well, as mice are HID?)

Galaxyrise: Ideally I need to sync the clocks to <1ms, so the USB jitter starts to dominate.

But anyway, this is all software. I need advice with hardware! :)

Any recommended chips? Cheap, easy package, as much of USB/HID handled by hardware rather than software (if that's gonna be useful), minimal auxiliary components, enough memory, high frequency/resolution timer, easy to work with/good docs/tools, and if possible USB2 High Speed.
« Last Edit: May 02, 2014, 12:13:25 am by lpc32 »
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 3781
  • Country: de
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #13 on: May 02, 2014, 07:02:38 pm »
janoc: CDC data is Bulk and possibly Isochronous.

It cannot be both, AFAIK. CDC is either interrupt (the common serial over USB implementations) or bulk (e.g. some USB Ethernet adapters). The common FTDI USB-to-UART chips implement interrupt transfers, AFAIK.

Or do you mean the control endpoint(s)? Iso is allowed to drop data, and I think is more likely to add jitter due to buffering (assuming it just needs to guarantee a semi-local data rate rather than max latency per transfer).

You are mixing everything together. All transfers are going to be buffered on the OS side, but the isochronous transfer is your best bet if you want to limit the jitter and have a defined deadlines when to deliver the data to/from the device (otherwise things like audio streams would have dropouts).

I don't need to transfer data most of the time, but when I do it's just a few bytes and the latency needs to be as consistent as possible. HID is both Interrupt and driverless in Windows, so seems the most suitable. (Though I think continuous but minimal data should work just as well, as mice are HID?)

But mice are certainly not working on ~1ms latency/jitter limit.

Galaxyrise: Ideally I need to sync the clocks to <1ms, so the USB jitter starts to dominate.

But anyway, this is all software. I need advice with hardware! :)

Any recommended chips? Cheap, easy package, as much of USB/HID handled by hardware rather than software (if that's gonna be useful), minimal auxiliary components, enough memory, high frequency/resolution timer, easy to work with/good docs/tools, and if possible USB2 High Speed.

I doubt you will find a general purpose chip that does HID in hardware. Most microcontrollers have general purpose USB transceivers and is up to you to write firmware that does what you want - e.g. implements HID or CDC or whatever.

The USB2 issue - you are insisting on USB2 - no idea why - it is a complete waste for your application as your latencies will be completely dominated by the transfer setup times. With HID you are limited to interrupt transfers (bulk and iso are not allowed for HID). So each transfer must wait until your host sends a control transfer and then starts a data transfer. That will make you fully dependent on the OS latencies (the control transfers are application/driver-initiated). In addition, you may demand the 125us polling frequency from the host, but there is no guarantee you will get it:

http://www.embeddedsys.com/subpages/resources/images/documents/InterruptsAndUSB.pdf

And no matter what USB host controller does, that is still only latency to the host's buffer. Afterwards you are at the mercy of the OS when it decides to actually give time to your driver.

I strongly suggest you rethink your design, because you are attempting to do something that may not be possible with the tools you want to use.
« Last Edit: May 02, 2014, 07:19:56 pm by janoc »
 

Online mariush

  • Super Contributor
  • ***
  • Posts: 4983
  • Country: ro
  • .
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #14 on: May 02, 2014, 07:31:02 pm »
Here's an idea... make your project output a stereo sound, 48Khz 16 bit stereo...  you'd be able to safely pack 8-10 bits and some crc info in that 16 bit sample, and have redundancy in each of those 48000 samples  a second. Use second channel for redundancy or for signalling that new data packet comes etc etc, use your brains.

Write some simple software that records line in input... if you need to synchronize, maybe use send a "ping" through usb (or maybe serial to mcu for less latency) to your microcontroller and your mcu sends back an "aknowledge" through the audio cable, and when it comes your software knows the time it takes for signal to go to mcu, be processed and how much time it takes for the soundcard to get you the sound.

See http://hackaday.com/2011/07/10/sound-card-microcontrollerpc-communication/
 

Offline amyk

  • Super Contributor
  • ***
  • Posts: 8240
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #15 on: May 03, 2014, 12:41:14 am »
Simplest solution would be to put a PCI/PCIe parallel port card in the PC.
 

Offline Rerouter

  • Super Contributor
  • ***
  • Posts: 4694
  • Country: au
  • Question Everything... Except This Statement
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #16 on: May 03, 2014, 02:16:39 am »
Another flip on it, why not time-stamp on the local device and use a good crystal, if your running it for stupidly long periods of time them have the computer send back a current time packet every hour and your assured that it wont matter how unreliable the communication method is, it will get there and have a correct time stamp,
 

Offline lpc32Topic starter

  • Frequent Contributor
  • **
  • Posts: 454
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #17 on: May 03, 2014, 10:15:00 pm »
janoc: Regardless of USB class or transfer type I need to settle on a chip first before I could experiment. (BTW, some mice/drivers do have a 1ms polling rate.)

mariush: I considered line-in, but I don't think I can expect low jitter from audio recording.

amyk: I need it to work on any random computer, so no real serial/parallel ports.

Rerouter: Jitter is still the issue.

Anyway, no one has recommended chips so I'll go with some more specific questions... :)

I'm new to this. Should I completely rule out anything without a big community? Does that mean only PIC and AVR, or should I consider also ST, TI, Freescale, Silicon Labs, etc.?

Some MCUs don't need an external crystal for USB and apparently re-adjust based on USB communication from the host. Any big reasons to still prefer an external crystal? If part minimization is a factor, are there any reasons to look at chips that do need a crystal?

If I need to store dynamics settings, should I still consider chips with no EEPROM and just flash? (Some advertise 512-byte flash block size, which is workable enough.) Do all chips allow writing to the flash in-system?

Is 8-bit vs 16-bit a big deal?

Do chips of the same family with more instructions offer a big advantage?
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 3781
  • Country: de
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #18 on: May 05, 2014, 10:24:16 pm »
janoc: Regardless of USB class or transfer type I need to settle on a chip first before I could experiment. (BTW, some mice/drivers do have a 1ms polling rate.)

You are confusing the polling rate that you can request from the host with actual latency you will get. That are two very different things.

If you take the interrupt transfer (e.g. HID), your device gets polled first time - it signals the host that it wants to do transfer. However, when you are going to actually get the incoming control transfer from the host setting up the actual data transfer containing the HID report, you don't know - that does not happen right away (e.g. the next polling period)

That is completely dependent on the situation on the OS, how it handles interrupts from the USB host controller, load, task scheduling, etc. Common latencies there are on the order of 10ms or so. The faster polling rate does not help you anything. That's why these "gaming mice" with super fast polling are mostly a gimmick - beyond certain polling rate the host will not be able to service the device as fast as it is asking and the extra interrupts only increase system load.

See why the 125us limit of USB 2.0 is a red herring here? The overhead of the interrupt transfers needed for something like HID is going to blow your latencies out the water. USB 2.0 is good if you need to transfer a LOT of data fast, but that isn't your case.

Let's take the isochronous transfer instead - in that case you have only one control transfer setting it up and afterwards you are streaming data to the host at a defined speed, no matter what else is going on on the bus. Your transfer even gets priority over stuff like bulk or interrupt transfers, because isochronous mode guarantees constant latency. However, to use this, you cannot insist on HID, you will need to write your own driver and firmware.

Re chips: small USB2.0 micros are rare, I think only some Cypress PSoCs have USB 2 transcievers. USB 2.0 is way too fast and needs way too much memory to service on the small chips (another reason to stick to USB 1.1). You would have to step up to some bigger ARM SoC and similar, but then we are talking devices on the level of a BeagleBoard or something that could power a smartphone.

If you stick to USB 1.1, then you can get the 12MBps full speed transciever in many micros - I have good experience with the STM32F1 & STM32F3xx line, also the LPC17xx from NXP is decent.

PIC was also mentioned, while it does work, the tooling and the libraries from Microchip leave a lot to be desired - it is not very easy to make work, even if you know what you are doing. Moreover, their libraries were not compiling with their new compilers, only the old C18 one :palm: In addition, the smallest USB-enabled chips like 18F14k50 have a weird memory layout with only very tiny slice of the address space accessible to the USB hardware - makes for an "interesting" debugging and development, where you have to count every byte of your HID report in order to make sure it still fits in there.

Some of Atmel's ATmega and XMega chips also have USB hardware, but I don't have experience with those, I have only used the V-USB bitbanged low speed USB on Atmels.
« Last Edit: May 05, 2014, 10:27:34 pm by janoc »
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 8515
  • Country: us
    • SiliconValleyGarage
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #19 on: May 05, 2014, 10:46:07 pm »
Janoc got it right

IN SHORT : High speed usb (480 mbit/s) is good to open a pipe and blast a huge amount of data in one operation. repeatedly opening and closing the pipe to send a few bytes takes way too much overhead. and you can't leave the pipe open.

now , Low speed usb (1.5mbit/s) or full speed usb (12 megabit) per second is limited to a packet every 1 millisecond  you can't go faster than that hardware wise. miss the slot and you have to wait.

USB is a time multiplexed bus. every 1 milliseconds a packet can be transmitted. miss the begin by 1 microsecond and you wait another 1 milliseconds.

so what you are attempting is simply not possible
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 3781
  • Country: de
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #20 on: May 05, 2014, 10:46:28 pm »
I'm new to this. Should I completely rule out anything without a big community? Does that mean only PIC and AVR, or should I consider also ST, TI, Freescale, Silicon Labs, etc.?

Um, ARM chips like from ST, NXP and similar have probably orders of magnitude larger "community" behind than anything AVR (especially because AVRs don't have USB except for a few odd chips).

Some MCUs don't need an external crystal for USB and apparently re-adjust based on USB communication from the host. Any big reasons to still prefer an external crystal? If part minimization is a factor, are there any reasons to look at chips that do need a crystal?

If you want to keep jitter to a minimum, you must have a crystal, period. The need for precise timing and not using a crystal oscillator is just silly. Moreover, the crystal-less operation usually works with large limitations - e.g. only for low speed USB. There are some PICs that do clock recovery from USB 2.0, but I have no experience with those (also see about PICs below)

If I need to store dynamics settings, should I still consider chips with no EEPROM and just flash? (Some advertise 512-byte flash block size, which is workable enough.) Do all chips allow writing to the flash in-system?

Typically you don't want to write into flash too often - the flash has limited number of writes, with some chips even as low as 1000 writes. Also not all chips allow self-programming (writing into program flash), even though those are getting rare. Either get a chip with a built-in EEPROM or add an I2C/SPI external one - it costs peanuts.


Is 8-bit vs 16-bit a big deal?

For USB 1.1, the difference between hacking something up on a slow 8bit CPU which and a beefy 32bit ARM can be whether you application will work or will not work at all (e.g. because the CPU is too slow to send the data over USB fast enough).

For USB 2.0 there are only some Microchip PICs that support this on 8bit/16bit, but I wouldn't feel comfortable using an 8bit/16bit PIC for a low latency USB 2.0 device. Those chips simply don't have the power nor RAM to handle much traffic at those speeds.

Do chips of the same family with more instructions offer a big advantage?

That depends pretty much on your compiler (whether it can take advantage of the extra instructions, for ex.), which chip are we talking about, etc. It is not possible to answer in general.

 

Offline GeorgeOfTheJungle

  • Super Contributor
  • ***
  • !
  • Posts: 2699
  • Country: tr
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #21 on: May 05, 2014, 11:43:41 pm »
What are some good cheap uCs for implementing a low-jitter USB HID device? I don't need a very fast processor, but low jitter is important (if that's an issue at all).

Would USB 2 High Speed help in assuring better worst case jitter?

On uCs with an integrated USB controller, how do you use it in practice? Minimally just connecting the chip directly to a USB connector?

Cypress makes some very pretty USB µC (ARM) + Analog + CPLD programmable all-in-one chips: EZ-USB and PSoC 5 have built-in USB, and the PSoC 4 dev kits (the dev kit is $4, the chips are $1) although not themselves USB devices, can connect via USB through a bridge chip (that isn't a HID device): http://www.cypress.com/?rID=92146



If you haven't seen them already, I'd say check out their PSoC and EZ-USB intro videos, just in case.
The further a society drifts from truth, the more it will hate those who speak it.
 

Offline lpc32Topic starter

  • Frequent Contributor
  • **
  • Posts: 454
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #22 on: May 05, 2014, 11:55:23 pm »
I realize actual latency won't be that good, but my hope is that jitter would have lower bounds. If <=1ms polling rates are there, I hope there's some drive in software/hardware to try to implement it in practice.

The USB specs define Interrupt as "limited latency transfer", "may have response time bounds that the USB must support", and "guaranteed maximum service period". Iso is "sensitive to delivery delays" and has "guaranteed bandwidth and bounded latency". I have not tested this, can't yet, but my hunch is that Iso transfers would act more like average bitrate than constant bitrate, as the implied usage is audio/video and other streaming. Jitter won't matter much, it would mainly need no buffer underruns and latency that isn't horrible. But this is all a future issue.

Yeah, I see cheap USB 480Mb MCUs are non-existent. There are some at around $5 from ST, Atmel, etc., but I'll skip that for the time being. And yes, I've read complaints about Microchip's software, so these are out.

Would you consider the ST/NXP/Freescale ARM ecosystem/community/whatever to be equally newcomer-friendly as AVR? I'm leaning toward Atmel, maybe ATmega32u4, but going with ARM might have long-term advantages.

Re extended instruction sets, assembly is acceptable too (where needed). I just wonder how common it is, with the more basic chips, to run into situations where you have to start doing things tediously with multiple instructions, or ending up getting much larger binary code, or much slower code even for simple things (no hardware mul?).

BTW, another advantage would be programming (flash) and debugging interfaces that don't require more hardware.
« Last Edit: May 06, 2014, 12:01:14 am by lpc32 »
 

Offline lpc32Topic starter

  • Frequent Contributor
  • **
  • Posts: 454
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #23 on: May 05, 2014, 11:58:08 pm »
George, thanks, I'll have a look. Too bad it's not the PSoC5 board for $4. But I bet anyway shipping is $40. :)
 

Offline miguelvp

  • Super Contributor
  • ***
  • Posts: 5550
  • Country: us
Re: Good cheap uC with low-jitter USB HID capability?
« Reply #24 on: May 06, 2014, 07:45:00 am »
George, thanks, I'll have a look. Too bad it's not the PSoC5 board for $4. But I bet anyway shipping is $40. :)

The pioneer board not only has the PSoC 4 but also comes with the PSoC5LP that can be used as well:
http://www.element14.com/community/thread/23736/l/100-projects-in-100-days

http://www.element14.com/community/thread/25084/l/psoc-4-pioneer-kit-community-project036-what-i-can-use-the-psoc-5lp-too

but the kit is $25 unlike the PSoC 42xx/41xx prototyping kits for $4 or $1 for the chip. I bought my pioneer from newark since they drove the effort for all that sample code (no HID samples I'm afraid)
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf