开云体育

I need smallest code to accommodate all of the features I've added.? And I can't use LTO because I need to prioritize heap.? LTO prioritizes stack.

I know of one oddity when you compile with "smallest code".? If you have a Serial.printf statement with a float, weird things will happen.? Integers are OK. There may be other oddities. -- 73 Greg

That's a possibility.? Stack is always tight but it's tighter when compiled with Faster with LTO than with Smallest Code and the latter is the one that crashes.? I assumed it crashed due to a memory

Maybe that pushed the stack over the edge?? Maybe there is some stuff you can temporarily remove from the stack to see if that is the case. -- 73 Greg KF5N

Exactly my thought.? So, I remove the DMAMEM keyword from the decimate state array in my version of the software and it crashed on PTT.

Or perhaps the Hilbert filter needs refined.? Here is an image from the setup mentioned in my previous post with the current v66-9 Hilbert filter: And here with the old Hilbert filter which is a tad

Eyeballing it, the noise hump is about 15 kHz wide.? So you should be able to open up the T41 filter all the way to 6kHz? and see more of the noise in the audio spectrum. I've never seen that

I wasn't able to hear a difference either, but if you put the lower sized decimation filter state variable in DMAMEM you'll sometimes see both sidebands raised as shown below.? Here my 4SQRP T41 is

I skimmed Hans' paper on the topic (what a great read!): https://qrp-labs.com/images/qmx/fdim/G0UPL.pdf It looks doable on the software side -- the T41 has plenty of processing power and the methods

But isn't the Hilbert transform a quite sharp bandpass filter?? Also you have a transmit equalizer, which can low-pass filter? I'm surprised more filtering is necessary.? Noise injection from a

I've added a lowpass IIR filter to the transmit audio chain. You can see the code modifications here: https://github.com/KI3P/SDTVer066-9/commit/69b0c5673da6ef05dcba047a719a6693359b1353 I've attached

After learning that Hans Summers was at Dayton Hamvention promoting his QMX which can now operate SSB using Polar Modulation, this got me thinking. Has anyone considered a T41 using Polar Modulation?

OK, I've pushed these changes to github. I tested my radio with both settings and can't say that I can hear a difference, but the changes are more correct. John W2TX pointed out in an email that the

Yes, I've used 279 for the size of FIR_dec3_EX_I_state and FIR_dec3_EX_Q_state.? Also, I set the size of FIR_int3_EX_I_state and FIR_int3_EX_Q_state to 151 in lines 1132 and 1133.

The size of the FIR filter's state vector should be numtaps+blocksize-1. For the decimation filter in question, this is 24+256-1 = 279. To be explicit, can you confirm that changing the sizes of the

I tracked down a problem I was having in SSB transmit to the size of the decimate state vectors, FIR_dec3_EX_I_state, FIR_dec3_EX_Q_state.? If these are allocated to DMAMEM, the RF out on the RF

Lynn, KU7Q, is using this in his CTR2-Dial Product. It's intended for use with remote operating apps that supports MIDI commands. It works great with Marcus Roskosch's Apple products apps.

Ken The suggestions from Jerry are good but they're not a permanent fix. The files in /tmp are exactly that: temporary. They will be recreated as and when needed the next time you boot the system. The

What are the audio reports like barefoot? Jack, W8TEE

By way of https://makezine.com/products/boards/m5stack-dial/ I was just looking at this https://shop.m5stack.com/products/m5stack-dial-v1-1 With my limited programming skills it would be a coin toss