jmarsh's latest activity

I read it. It's gibberish nonsense to cover up the fact that it doesn't know how to proceed. You don't need a logic analyzer or a scope to know how I2C works. Regardless it's got nothing to do with how that state machine isn't anywhere near...

Seems pretty obvious that state machine is never going to work.

Indeed "Gemini is AI and can make mistakes." (used for tetris): {this is free - the other month it had a way to get a link to share the session} please provide the code for flexio i2c slave on imxrt1062 Gemini said To put it bluntly: there is...

At least it admits it has nothing to "copy-paste" from, I guess.

I'm saying ask it to implement I2C slave mode, using FlexIO. This is something that NXP themselves haven't bothered to write code for since it's considered too difficult. So the AI will have no existing code to borrow from and will have to write...

If you're using the onboard SD slot, there is no SPI bus being used.

It works with Teensy 3, so I have sent a PR to fix it for T4. It does appear that it's intended to work: template <class Vol, class Fmt> class SdBase : public Vol { public...

Call SD.begin(BUILTIN_SDCARD) as normal then use the SD.sdfs object.

Use the SD.sdfs object to access the underlying SdFat class. Use the end() method on that class to close the volume.

AI is garbage for embedded machine because it can only "learn" from existing code - if nobody has already written it (which is highly likely, even for boards like Teensy there are a lot of peripherals that do not have supporting libraries), it...

@defragster I think you are missing the part where OP is trying to talk to a sensor that has the CP2105 incorporated. There is no raw UART access, just a USB port.

Can you double-check the trace to J3? From the earlier picture it looks like it goes through a via, then a straight line down/left to another via, to the clk pin on the SD. But from that pic it passes directly over three other balls and it's not...

I'm not saying they're necessarily misaligned, I'm saying the wiring for SDIO is not compatible with SPI mode, so trying to tell the SD library to use SPI will never work. Keep it simple and just use SD.begin(BUILTIN_SDCARD);

Can you show the code you've tried using SDIO? It won't work as SPI2 since the pins don't match up in that configuration.

Let's backtrack a bit here.... are you intending to drive the SD chip using SPI2, or the native SDHC controller (USDHC1) ?

I actually meant changing ATTR_DST to 3, so that 8 bytes at a time are written to PSRAM (matching the burst length).

64 bits per transfer may be better since it's the burst size? I find using PSRAM for a framebuffer really depends on the refresh rate, it generally can't handle anything beyond 640x480 (25MHz pixel clock).

They are not the same product. That explanation (which I suspect is AI generated) is incorrect. Here is the correct explanation taken from the actual chip datasheet (and the reason why the -Q models will not work):

https://www.pjrc.com/store/ic_mkl02_t4.html Look under the "Supported Chips" heading:

What is the write size (set by ATTR_DST) ? The column headings don't really make the direction clear, "to mem", "from mem", "DMA to", "DMA from" are all ambiguous when you're copying between two different memory regions.

They're not going to be different between chips since the FlexSPI clockspeed sets the transfer rate, i.e. there's no non-deterministic behaviour involving polling.

There's not really much "hardware" to speak of: T4.1 is running a sketch based on edid-decode to interrogate the monitor's capabilities through the HDMI connection, then display them on the screen. I tried a photo of something else but...

Those photos don't really help much. Can you just wire it up powered by external 5V as described as not working in the first post and take a picture? It sounds like the easiest way to power it would be taking the 6V from the battery through a...

You are connecting ground as well, right?

It will be a miracle the day the Arduino IDE starts to clean up after itself.

Just use gettimeofday() instead, it will give you seconds and microseconds of the current RTC value.

Running an interrupt at that frequency for data collection is crazy - use a DMA channel instead.

Even at 528MHz the "5" chips still require higher voltage. Code to handle it is here. I cannot stress how important this is, unless you want to be chasing random software crashes without any obvious source.

Keep in mind the "5" in the model number means it is only rated up to 500MHz, not 600MHz like the MCU on the standard Teensy 4.x. It also requires higher internal voltage settings to run at the same speeds otherwise it's prone to suffer random...

Did you actually make this change? Because the original value of "44.1" is not the same as AUDIO_SAMPLE_RATE_EXACT.

There are some FTDI chips supported by USBHost_t36 that provide multiple ports, by claiming the individual interfaces rather than the whole device. The CP210x driver would need to be modified to do the same.

Windows can act very strangely when a previously used USB device changes its audio config. I would recommend getting USBDeview and fully removing any existing driver instances of the Teensy from the PC.

To be clear which model 1062 are you using?

Look at the source code of TeensyTimerTool to see how the interrupts are handled differently for PIT and GPT.

DAT3 and CLK both going to ball J2 doesn't seem right.

All IntervalTimers share a single interrupt.

Because the capacity doesn't play a part? There are both regular SD and SDHC cards that are 4GB...

You would if it was an older reader that didn't handle SDHC cards. The addressing mode was changed from bytes to sectors and an extra initialization command is required. This is all handled by the reader, not the PC.

You never set capturing back to true so presumably loop() is just continuously sending the same values endlessly.

I'm aware of how a flash translation layer works. But I can practically guarantee you the software is not "connecting to the card at a low level". If you plug a USB SD card reader into a PC, the PC doesn't even know that it's an SD card - it...

You haven't showed the complete code which makes it difficult to give specific advice... Try using one big buffer with a DMA interrupt at halfway as well as completion, that way it's always enabled. (Also don't Serial.write() from an ISR! Post a...

I really doubt the formatting program trims (the technical term for telling the card a sector is unused by the filesystem) all the sectors. Doing so requires low-level access to the SD card registers which most operating systems simply aren't...

In theory (not so much in practice), Teensy's SPI has a top limit of 30MHz so how exactly are you configuring it?

Quad timer 3 you say? The quad timers have an enable register with bits for each of their 4 timers, they default to enabled so they start running as soon as their mode is set. Try adding TMR3_ENBL = 0; to the beginning of your InitPWM function...

There's not really any "software" support needed for OTG. There's some IO pins to support detecting what sort of device is connected (and if it's a charger), but they're not routed anywhere. So the easiest way to handle it is let the default USB...

The micro USB port can use a storage device / be used as a host port.

Why go for something non-standard? Either go with USB-C or good old chunky USB B: https://www.digikey.com.au/en/products/detail/sparkfun-electronics/15463/10819707 Either way, the better option is a breakout cable that will be permanently...

If handling user input is part of the Teensy's "main job" then I wouldn't bother with interrupts, i.e. only if it was spending most of its time on some other task that would prevent it responding in a reasonable way.

The MCP23S17 supports activating internal pull-ups when a pin is configured for input.

It should just be a direct connection to the output pin on the toslink receiver. There's really not much that can go wrong unless you've wired it to the wrong pin, or the audio device isn't sending audio over the optical cable. You can always...