Post Created date
Problem with pointer to struct field...
Looks like a cross-post here.   The same advice applies here. Post a short, self-contained, compilable, example that demonstrates the problem. I suspect the problem is...
Thursday, 9 July 2015 - 22:07
question about DDS synthesis RC filtering
joeymorin wrote: Exactly like I mentioned in post #36... including some pretty pictures! If I'm reading your post #36 correctly you're referring to amplitude quantization noise...
Tuesday, 7 July 2015 - 23:15
question about DDS synthesis RC filtering
I still believe the noise is caused by truncating the phase accumulator:   https://en.wikipedia.org/wiki/Numerically_controlled_oscillator#Phase_truncation_spurs   "...
Tuesday, 7 July 2015 - 22:20
UART RS232 Character Constant too long
clawson wrote:   Tuurbo46 wrote:        up to say 1020 from previous experience will fit in a byte.  But now not using a uart library or function and...
Tuesday, 7 July 2015 - 21:01
question about DDS synthesis RC filtering
Is your frequency sweep phase-coherent? That is, does each frequency start at the same phase that the previous frequency ended on? Also, DDS inherently has phase noise that is...
Saturday, 4 July 2015 - 22:05
question about DDS synthesis RC filtering
The sample rate of a DDS system can be very high, much higher than any other available audio source. The Commodore 64 DDS module (the SID chip), for example, had a 1 MHz sample...
Saturday, 4 July 2015 - 17:20
8 bit floating point arithmetic on atmega8535
You're in luck! The ATmega series all have hardware 8-bit multiply instructions (returning a 16-bit product) which makes it easier to implement floating-point multiplication. As...
Thursday, 2 July 2015 - 23:10
sending twos complement to a unsigned char.
joeymorin wrote: Just use bitfields: struct { int seven_bit_int : 7; } foo; . . . signed char bar; . . . bar = foo.seven_bit_int;The compiler will generate the appropriate...
Monday, 22 June 2015 - 14:12
sending twos complement to a unsigned char.
Ah, yes, I forgot the part where you need a positive value. Theusch's method should work.   I mentioned shifting and dividing because the shift puts the sign bit of the 7-bit...
Saturday, 20 June 2015 - 15:09
sending twos complement to a unsigned char.
Stuff the data in an int8_t, shift it left by one bit, and then divide by two. Or just copy the seventh bit to the eighth bit. Either method will give you a signed 8-bit value in...
Saturday, 20 June 2015 - 03:45
Have the progmem and 64kb ptr problems been fixed or shall I stay with BASCOM?
Foxcat385 wrote: Name one Linux phone that's not Android, but that's open source for GPL-less forking. Linux running underneath Android is still subject to the GPL. Android is...
Tuesday, 19 May 2015 - 23:50
avr-gcc never exits any functions
From your Makefile: CFLAGS += -mmcu=$(MCUARCH) -mmcu=$(MCU) -Os -Wall -std=gnu99Why do you specify -mmcu= twice with different values? I suspect that might cause gcc to build for...
Sunday, 10 May 2015 - 18:37

Pages