[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [avrdude-dev] Calibration bytes on ATmega8
From: |
Theodore A. Roth |
Subject: |
Re: [avrdude-dev] Calibration bytes on ATmega8 |
Date: |
Tue, 17 Jun 2003 13:40:41 -0700 (PDT) |
On Tue, 17 Jun 2003, Brian Dean wrote:
> On Tue, May 06, 2003 at 05:40:13PM -0300, Francisco T. A. Silva wrote:
>
> > The Atmega8 has 4 calibration bytes. The atttached patch makes it
> > possible for avrdude to read them all.
>
> The data sheet speaks only of a single calibration byte. If you can
> show where this is documented somewhere in the datasheet or perhaps in
> an errata somewhere, I have no problem adding this to the
> configuration. However, I don't think it's wise to deviate from the
> documented data sheets too much. For example, even if this works for
> some parts, since it's not documented in the data sheet Atmel is free
> to change the behaviour at any time, and we probably should not rely
> on it being there.
Actually, it is in the datasheet (at least for the mega8) which I
just pulled off of Atmel's site:
Calibration Byte
The ATmega8 stores four different calibration values for the
internal RC Oscillator. These bytes resides in the signature row
High Byte of the addresses 0x000, 0x0001, 0x0002, and 0x0003 for 1,
2, 4, and 8 Mhz respectively. During Reset, the 1 MHz value is
automatically loaded into the OSCCAL Register. If other frequencies
are used, the calibration value has to be loaded manually, see
Oscillator Calibration Register OSCCAL on page 29 for details.
I haven't verified the other data sheets, but I suspect that any
device which can be fused to select more than one internal osc freq
will have a cal byte for each osc freq.
I don't remember the patch in question though so I can't comment on
it's correctness.
Ted Roth