That’s interesting. What do you use on Linux?
Just acr122_usb under libnfc, the direct USB driver/module.
acr122u_pcsc is supposed to be deprecated I believe, so I haven’t tried anything other than acr122_usb. I’m using libnfc 1.8.0, the latest in the arch repos.
I can read my implants fine, but 0 LED activity at all. The LED works fine in Windows 
I don’t really care regardless, I usually just look at my xSIID’s LED to confirm read activity anyway, but it is indeed odd.
Oh ok, libnfc. Yes it does leave the LED off until you open the device.
I don’t know if pcsclite is supposed to be deprecated. It sure looks alive and well, and supported though 
pcsc turns the red LED on as soon as the pcscd daemon runs.
For me, the light never activates, even when I’m actively reading a card or implant via the nfc-poll command for example. Not red, not green, off. Wouldn’t even know it was plugged in.
It’s not pcsclite that’s deprecated, it’s the specific acr122_pcsc driver that’s included with libnfc.
Sorry I misunderstood.
I’ll have to recheck what libnfc does exactly with the LEDs. I seem to remember that’s what it does, but I actually never use it, because it’s very bare-bone and there aren’t very many applications that use it. I played with it because I wanted to talk to the PN533 directly - something pcsclite doesn’t let you do. But it’s too low-level for what I need it to do.
It’s not really a driver by the way, it’s just a library.
Eh, that feels more like semantics than anything else. I wasn’t using the linux definition of driver, but the more general definition.
Yes, libnfc itself is a library, but IMO, the device components of it, acr122_usb and acr122_pcsc, could arguably be considered drivers in a generic way. Libnfc doesn’t use any of the nfc-related kernel modules for the pn533, instead relying on either its own direct USB support, using libnfc, or using pcsclite & the pcsc daemon as a backbone for communication (also using libusb I believe). It’s still acting as an abstraction layer to allow higher level software, in this case libnfc itself even, to communicate with hardware.
Yes, it’s user-mode, not kernel-mode, but I would still consider that to be acting as a device driver.
Despite being a library, I would say that libusb acts as the equivalent of a user-mode driver as well. Same with FUSE.
I never pass up a chance to be pedantic
1 Like
Seems like this is on the road to working, and that’s pretty awesome. 
I kinda wish there was a way to avoid the scalpel, though… I’ve been doing a lot of research on just needles in general and nothing really goes past 4 gauge. If there was a way to install something like this without the goofus legal ramifications of all the scalpel shit, that would be fucking amazing. I’d get 2 or 3 different configurations for sure. I especially love the anti-rotation concept.
2 Likes
I would still be concerned about using too big a needle.
There’s a point where the needle would stretch the incision more than it would avoid damage… 
I do share your point about the legal ramifications, though…
now onto a slight derail:
I use nfcpy, bouncing off from libnfc, and my ACR122 keeps the lights off as well, but with nfcpy it goes red once I capture the device for use (such as waiting for a read), and goes green on read.
Returns to no off once I release it, though.
My PASoRI (Sony) reader is currently always off, though.
2 Likes
Sort of… the official driver for Windows enables the OS to see it as a PC/SC smart card reader, so even with no user space software running, placing a readable tag on the acr122u will result in a beep and a green light. Technically the OS itself is acting as “software” but the point is no actual application need be running to get the green light response on Windows.
1 Like
Did you see above that mine was a ISO15693 (SLIX) hence the Red light?
Which reminds me, I have to chase up my replacements. Thanks 
That comes straight from a factory default “middleware hardware” the ACR122 has… which is exactly what makes it a pain to deal with Type4 tags from Linux… 
(And why I use a pasori as my main reader)
You can talk directly to the PN533 in an ACR122U, either through PC/SC with pseudo APDUs or through the pn53x methods in libnfc. There is no inherent limitation other than the chip’s with the ACR122U if you’re willing to work at bare metal level.
Does a needle not cut what it needs to slide in? From my experience, they are not tapered.
I wouldn’t try a 4g In my pinky distal phalange though to add to your point.
1 Like
Both true and not true.
If I’m working in .cpp I can make use of libnfc to reach out. True.
But while I’m working with nfcpy, there are limitations caused by the approach the manufacturer took towards implementing PC/SC.
I won’t get into detail to avoid deepening the derail, but the link above goes straight into the official statement about that reader’s compatibility issues with nfcpy.
Good needles do cut. But they also stretch.
The surface of the needle’s blade is always going to be smaller than the volume it creates…
Won’t get into the math (because I’m just waking up, thus lazy)…
but basically:
The blade only cuts the skin roughly as much as the needle’s diameter (2xRadius), then the needle’s body stretches the skin to open to the length of the needle’s circumference (2xPixRadius).
The larger (as in “fatter”) the needle, the bigger the stretch effect.
Depending on your skin a stretch might be nothing. but depending on the person, might cause more harm than expected.
The issue for the comparison / my previous comment about needles stretching more than needed in some cases boils down to:
If I’m installing a hard-backed chip, like a flex EM, I do need to have an incision as wide as the chip itself. So…
With a scalpel approach I only need that, plus a half mm stretch to slide in the dermal separator tool (sort of a spatula).
The top portion of the skin will be only 1mm higher than the muscle layer.
If I were to get a needle large enough to slide in the disk part, it would stretch it to the point that the top portion of the skin would be {diameter}mm higher than the muscle layer.
(this editor cuts out if I use < instead of { )
This is acceptable when we’re talking about a wedge implant, especially if it has some sideways bend, but the bigger and less flexible you go (hello EM!), the less acceptable this approach becomes.
Therre are no “hard lines”, though!
1 Like