Digistump Forums
The Digispark => Digispark (Original) Support => Topic started by: digistump on December 30, 2012, 07:20:56 pm
-
There seem to be some issues with Retina Mac Book Pros - either not seeing the Digispark at all, or it causing the system to hang. I am pretty certain these are related to the USB 3.0 ports used entirely on the Retina MBP - these are suppose to be backward compatible but other device makers have started to have issues as well, some issues are caused by drivers, some firmware, and some because of the plug design or other electrical issues.
The best solution at this point is to plug your Digispark into a USB hub - USB 2.0 or lower - which seems to solve the issues.
I do not have a Retina Mac Book Pro at this point, but I do have a computer with several USB 3.0 ports on it and I'll be doing some testing soon to see if I can isolate the cause and any possible solutions.
In the meantime please post anything related to Retina MBP compatibility here.
Thanks,
Erik
-
If anyone with a Retina MBP is reading this can you please try the following:
Plug a Digispark into the Retina MBP USB port using a USB extension cable - does that work any better? (This would mean it is related to the plug)
-
I\'ve seen issues when using an extension cable. I can test some of the various cables/hubs I have to see if there\'s some sort of pattern.
-
@buffington - it seemed other Retina MBP users couldn\'t get it to work at all, I think a good extension could help that as it might be the plug of the Digispark contacting the USB 3.0 pins - but with yours working I\'d think that an extension would cause more issues, as generally we recommend not to use one since most are cheap and cause too much of a voltage drop.
I just tested on a PC with USB 3.0 ports and it worked, but it is a tower computer so you can\'t push it in as far as with a laptop, which would prevent contact with the USB 3.0 specific pins.
We also could be dealing with a driver issue with the USB 3.0 controller - which might be more related to your issues, the hub would fix it because the hub is then the controller. I only think that is a possibility because I\'ve seen that some people have issues with USB 2.0 sound devices on Retina MBP because of the controller that is used.
Let us know if you find a pattern that works! - Thanks for experimenting with it!
-
Here\'s something that consistently happens when connected directly or with an extension (but not with a hub):
1. With the Digispark unplugged, upload a program.
2. When prompted, plug the Digispark in.
3. The MBP\'s trackpad will become unresponsive while the program is uploaded to the Digispark and remain unresponsive for as long as a minute.
Looking at Console.app, I see a handful of these messages that correspond with the period of inactivity:
12/30/12 9:08:08.000 PM kernel[0]: AppleUSBMultitouchDriver::_deviceGetReport - DeviceRequest for reportID 0x0 returned with result 0xe000404f - retrying
-
Is this true on all USB ports? Very strange...
-
Using the 27\" Cinema Display\'s hub, I see no issues of inactivity. Nor do I see any issues with any other hubs (a cheap, USB 2.0 no name 4 port powered hub, and the built in ports on a 24\" Dell LCD monitor).
One thing I just noticed was that I only see the unresponsive track pad when uploading a program that\'s different than that of the one already on the Digispark.
I tested this by adjusting a single delay value within the DigisparkRGB example. Uploaded it once, saw the delay (since it was replacing a different program), then tried to get the delay 10 more times by uploading again. No trackpad problems.
Then I changed the delay (which in itself I doubt is relevant, but it\'s a hard coded value). After uploading, I saw the trackpad problem. I left it the same and uploaded again. No problem. Changed it to a new value - saw the trackpad problem.
Of the 10 times I changed values, I saw trackpad problems each time.
-
After a few tries here I\'ve not seen any issues with the trackpad when using any other port aside from one of the two on the machine.
-
Thanks for the analysis!
Something with the Digispark upload is obviously clashing with the trackpad communication when they are on the same controller.
-
Hello,
It\'s not related to just retina MBP\'s. I have the latest 13\' MBP and here it\'s the same issue. i think all USB devices freezes each minute until a restart after i try to upload with the direct usb connection. But it also worked well with a USB Hub.
-
@saperlot for the freezing try just quitting the digispark arduino program. If that doesn't work try opening a terminal, type "sudo killall -9 micronucleus". Does this fix it? Arduino software seems to leave a zombie micronucleus program running which can be a little mean to the USB interface, especially if several copies end up running - maybe it should time out after five seconds or so.
-
@Bluebie - I think a 10-20 second time out would be a great feature for the uploader, I had hacked it in to the ruby version but haven\'t had a chance to add it to the C++ one yet
-
I opened my digispark for the 1st time tonight on a retina MBP, and ran into the same issue. Used a USB cable and then started working.
One catch is that I have to unplug the digispark everytime I want to upload. I noticed the trackpad issue once (altho my Magic Trackpad via BT seemed to work), but I can\'t reproduce anymore.
Anyway, really cool to have arduino on such a small USB chip! *stoked*
-
My digispark isn\'t detected on my 2009 13\" MBP. I don\'t have a USB cable around to see if that works.
-
@bobchadwick - this is likely due to the USB ports being recessed a bit more than on most machines, unfortunately the only good solution for that is to use a hub or cable.
Something like this is nice and portable - http://www.ebay.com/itm/USB-Male-to-USB-Female-M-F-Adaptor-Converter-/180544622456?pt=LH_DefaultDomain_0&hash=item2a094c4f78
But any half decent usb extension cable or hub will likely work.
-
Just wanted to note I'm also getting the trackpad freezes on any communication to the digispark, running the send.py program to send text to the USB2LCD program causes my trackpad to hang for at least a few seconds. I'm on a 15" 2008 Macbook. Works fine besides that though, guess I'll have to get a cable...
EDIT: I've found a USB cable, but the trackpad issues persist. It does seem that merely running the send.py program causes the hangs, as using it even without the digispark attached produces the hang.
-
alexnking - at this point we believe the hang is caused by libusb or the python implementation of it (pyusb) - or the way we've implemented that code. Once shipping is done (80% of the way there!) I'll be working on the software side more to straighten things like this out/implement better ways to talk to the USB device.
-
On my macbook air I get the trackpad/keyboard freezes too, and I find if I keep one finger on the trackpad wiggling the mouse around when I start the program and plug the digispark in, it doesn't freeze at all. I have no idea why. I wonder if holding down a modifier key on the keyboard would have the same effect.
-
@saperlot for the freezing try just quitting the digispark arduino program. If that doesn't work try opening a terminal, type "sudo killall -9 micronucleus". Does this fix it? Arduino software seems to leave a zombie micronucleus program running which can be a little mean to the USB interface, especially if several copies end up running - maybe it should time out after five seconds or so.
I have very similar problem with my 15' MBP (I think 2009 model). Arduino IDE gets stuck at "Plug in device now" prompt forever, and I see "AppleUSBMultitouchDriver" error messages in Console repeating forever.
Killing all micronucleus process does stop the error message. But still no go with loading the program into Digispark...
Oddly, I had successful load a few of the first tries, then it stopped working at all. No change in the way I'm connecting Digispark.
I noticed that the following error shows up when I connect Digispark the USB:
kernel USBF: 22348.769 [0xc589400] The IOUSBFamily is having trouble enumerating a USB device that has been plugged in. It will keep retrying. (Port 4 of Hub at 0x26200000)
This message repeat a few times following the Digispark connect. What does this all mean?
-
@theLEDartist - have you tried a different port/hub - the message means the trackpad and digispark are conflicting on the same port/internal hub
-
@theLEDartist - have you tried a different port/hub - the message means the trackpad and digispark are conflicting on the same port/internal hub
Tried different ports, and via an external hub - no difference. (except the port# reported in the error log.) My take on the situation is that my Mac OS (10.6.8 BTW) can not enumerate Digispark as an USB device, therefore Digispark uploader can not find the Digispark. Then micronucleus causes problems with the trackpad driver (by repeatedly resetting the driver somehow).
I have 3 Digisparks and they all do the same.
-
If you have access to a different USB hub, try that. I'm using a cheapo Belkin 4-port USB 2.0 "pocket hub" on my 2008 Macbook. I can't program my Digisparks when plugged directly into the computer, but they work fine through my hub.
-
theLEDartist-
I use 10.6.8 so that isn't it.
Was the hub you tried powered - that has worked for some linux users - I don't think any Mac folks have had to resort to that yet but it might be worth a shot.
Do you have access to a second machine to test? - on the very rare chance you got three bad ones or something like that.
-
Same issue here, MacBook Pro Retina, 15". Inserting basically froze the machine and eventually required a reboot to get it back and talking to me. Adding a small extension inline seems to have resolved the problem.
(https://lh5.googleusercontent.com/-Q19Ao5XwDGk/UR7Sxubz52I/AAAAAAAABLI/rnn-Sa911BY/s601/IMG_20130215_161922.jpg)
-
Just to add a (non-MBPR) data point, I have a laptop with USB 3.0 ports. My digispark is working just fine in the port, tested with the programmer and with the Cheerlights example python script. No extensions used at all. Laptop is a Lenovo x230.
-
theLEDartist-
I use 10.6.8 so that isn't it.
Was the hub you tried powered - that has worked for some linux users - I don't think any Mac folks have had to resort to that yet but it might be worth a shot.
Do you have access to a second machine to test? - on the very rare chance you got three bad ones or something like that.
I'm using a Powered (Pluggable 7 port, USB 3.0 powered hub) hub with a 4a power supply, also trying an extension that is USB 2, and cannot even get my debug console to give the kernel errors mentioned above. My hub fails to indicate it actually sees an active device, but the digispark DOES power on.
Using a retina Macbook pro 15". 10.8.2.
-
Dippyskoodlez -
Do you have multiple digispark? If so do you see this on more than one? If the computer doesn't see them at all it could be a bad Digispark. Also look at the connector on the Digispark and see if the USB pins seem to be centered?
Do you have access to a second machine to test with?
Thanks,
Erik
-
Dippyskoodlez -
Do you have multiple digispark? If so do you see this on more than one? If the computer doesn't see them at all it could be a bad Digispark. Also look at the connector on the Digispark and see if the USB pins seem to be centered?
Do you have access to a second machine to test with?
Thanks,
Erik
I'm getting the same behavior with all 3 of my Digisparks.
http://www.amazon.com/Plugable-Power-Adapter-VL812-Chipset/dp/B008ZGKWQI/ref=sr_1_1?ie=UTF8&qid=1362640299&sr=8-1&keywords=plugable+usb+3.0 (http://www.amazon.com/Plugable-Power-Adapter-VL812-Chipset/dp/B008ZGKWQI/ref=sr_1_1?ie=UTF8&qid=1362640299&sr=8-1&keywords=plugable+usb+3.0)
Ok, did some more troubleshooting:
Windows PC(NOT macbook Pro. ivybridge/Z77) : Everything works. No issues. Including hub.
OS X: Nothing will even detect the digispark. Hub, Extension, internal ports. Notta. ]With some wiggling, both USB ports will power one of my digisparks, but not detect.
rMBP+Windows(Win7 x64) Bootcamp install: INTERNAL ports will NOT detect digispark. Extension WORKS.
-
Dippyskoodlez -
Re: Internal ports on MBP - usually this is due to the usb end of the Digispark not going far enough into the port - this is common and has been addressed in the next revision.
Re: The USB 3.0 hub - there are some issues with 3.0 devices (with more than just the Digispark).
Re: Working in windows but not os x - strange! usually it works better in os x! - Just the extension with the Digispark (no hub) doesn't work on any port?
-
Dippyskoodlez -
Re: The USB 3.0 hub - there are some issues with 3.0 devices (with more than just the Digispark).
Re: Working in windows but not os x - strange! usually it works better in os x! - Just the extension with the Digispark (no hub) doesn't work on any port?
re: Hub; this hub actually works, and works well :) It just gives the same symptoms in OS X as the native devices. Every other device I use (x-fi, DAC, HDD's, etc all work great.)
I cannot get it [Digispark] to work at all in 10.8.2.
-
I have a non-retina MacBook Pro (medio 2012 model, OSX 10.8.3) that only has USB3.0 ports.
When I insert the DigiSpark, it gets powered, but does not show up. When I use an USB extension cable, the DigiSpark does show up.
How to check for your DigiSpark:
Click the Apple logo at the top left of your desktop, and select "About this Mac".
In the popup, click "More info" and then "System Report".
In the Sytem Report, click on 'USB' in the left column (under 'Hardware').
Insert your Digispark, then within 5 seconds hit Command-R to refresh the System Report.
Your DigiSpark should now show up as 'Vendor-Specific Device', and if you click on that, it will list "Product ID: 0x0753, Vendor ID: 0x16d0" which are the correct numbers for the DigiSpark bootloader. You must do the refresh within 5 seconds, as the boatloader shuts down after that.
-
This sounds like the "pins are too short" issue - our next revision addresses this with a longer connector
-
I am also using the MBP Ret. So far it seems to have been behaving with the digispark programming tool. and a usb cable.
I have 4 units that I am building into some prototypes. I will update if there are issues I haven't seen.
L
-
Well, crud. I just upgraded to a MBP Retina last weekend, and now that I've tried programming a Digispark for the first time, I find it doesn't work. Tried my cheap USB hub (even with external power) and a plain USB extension cable. Neither helped.
:(
I still have my old Macbook, so I'm not completely dead in the water. But I hope I can find a workaround so I don't have to keep swapping around. I was going to give the old laptop to one of my kids after I'm sure I've gotten everything I need off of it.
-
Dougal - have you tried all the port and removing any other USB devices? Witht he extension and or programming tool (if you have it)?
-
I tried both of the USB3 ports on the MBP, two different ports on the hub, even tried hooking the hub to the laptop with a USB extension cable. I dug up another extension cable last night, but haven't had a chance to try it yet.
Not sure when I'll have time to experiment much with it again, but hopefully soon.
-
I just had success uploading a sketch to a Digispark from my new MacBook Pro Retina!
Details here:
http://digistump.com/board/index.php/topic,1059.msg4454.html#msg4454
-
Pins are long enough for me... I think. The lights on the digispark turn on/start blinking.
I hit upload and:
Sketch uses 360 bytes (2%) of program storage space. Maximum is 14,844 bytes.
Global variables use 9 bytes of dynamic memory.
Running Digispark Uploader...
Plug in device now... (will timeout in 60 seconds)
> Please plug in the device ...
> Press CTRL+C to terminate the program.
> Device search timed out
I am using a USB extention cable that is 4 inches long. I do not have any windows machines, but I do have (ubuntu, kali) linux (dual boot on osx)
-
Was anything ever resolved with this? I seem to be having a bunch of USB problems since El Cap came out (USB Serial, etc). My Digispark Original will not be detected by the mac laptop. I have verified it works on an older windows pc and even loaded the keyboard demo so It should connect and write two words as a usb keyboard hid. On the mac it always shows: 5/20/16 12:17:30.759 PM system_profiler[11509]: SPUSBDevice: IOServiceGetMatchingService did not return anything for location 0x00200000
repeated on the console with the location changing. Never is seen as a valid usb device and cannot be programmed either on the mac. I can program with a windows computer if I had to but I will still run into the problem where I was going to use the device to capture data from a caliper and send it as keyboard keypresses but since the device will not be recognized even as a usb hid that functionality will not work.
-
I have the Mini ATtiny85 USB, dev.board in my ports list in the Arduino IDE version 1.8.3 ports menu.
I have a MacBook Pro (Retina, Mid 2012) running macOS High Sierra version 10.13.5. I have the Digistump AVR Boards version 1.6.7 loaded.
Can anyone help?
-
It is a well known issue with MacBooks going back some ways. It is as many have surmised, simply that the port design does not allow for contact with the pins on the digispark usb. Simple enough to get around, just add a hub or extension cable.