Cool, so what exactly is happening exactly when I did the following?
--I left arduino's atmega on the arduino just as it is, throughout the following steps...
--insert standalone atmega on breadboard
--connect up the circuit as shown in ArduinoToBreadboard tutorial (i.e., with SCK, MISO, and MOSI pins of arduino and of standalone atmega chip connected to each other)
--reset pin of standalone atmega chip (datasheet pin #1) with pull-up resistor, as well as a cable to pin #10 of Arduino (which is used in the ArduinoISP sketch for slave reset)
--arduino reset pin never handled by me (i.e., left open)
--I loaded ArduinoISP sketch to arduino
--burn bootloader using ArduinoIDE's Tools > Burn Bootloader command.
--then use AVRdude to load a hex file, e.g. avrdude -c avrisp -p m328p -P com3 -U flash:w:Blink.cpp.hex
The visible result is that everything worked fine... the Blink program is indeed uploaded, as I checked by placing an LED on standalone atmega chip.
But is there anything wrong with this approach of uploading code to the standalone chip? (i.e., with Arduino's atmega chip and reset pin left untouched, and shorting MOSI, MISO, and SCK pins between Arduino and standalone atmega chip)
I guess my question is...
What is the difference between that approach VERSUS the other approach discussed in this thread (i.e. where we first remove the Arduino Atmega chip from the board and ground Arduino's reset pin and then upload to standalone chip via ArduinoIDE/AVRdude with the RX and TX pins cross-connected?