Re: [avrdude-dev] ATtiny2313 and pony-stk200 problems?

[Jan-Hinnerk Reichert]

On Monday 28 June 2004 16:46, Bob Paddock wrote:
> > 1) The 1MHz mentioned above is the oscillator frequency of the
> > chip *not* the SPI clock.
>
> The Tiny2313 from the factory is running at 500 kHz (4 MHz/8).
> The Serial Programming Characteristics says that a minimum of two
> clock cycles
> are required which would give a max SPI clock of 250 kHz max.

I believe you are referring to table 80 on page 176.
It says that *both* "pulse width low" and "pulse width high" must be 
longer than 2*t_CLCL. Hence, the SPI-clock cycle must be longer than 
4*t_CLCL

> > 2) The above comment is from the current CVS. Bob is apparently
> > using 4.3.0
>
> I built the CVS code a few minutes ago.  I can now read the
> signature every time,

At least this shows that the problem was timing related.

> but I still can not verify the flash.  The error is always 0xbe !=
> 0x00 at address 0x0000.

Can you do a verify with the PPI-prommer after you have programmed 
with STK500?

> Placing the chip back in the STK500 to read it produces this:
[output deleted]

I think this error has nothing to do with timing.
It rather looks like Atmel has made a (copy&paste ;-) mistake when 
describing the "Load Program Memory Page" and "Write Program Memory 
Page"-instructions in table 79 ;-(

It should be:
Load Program Memory Page:
 0100 H000 000x xxxx xxxx bbbb iiii iiii
Write Program Memory Page:
 0100 1100 0000 00aa bbbb xxxx xxxx xxxx

If you adjust your config accordingly everything should be fine. Would 
you mind mailing Atmel if it turns out to be this way?

> > 1) Timing is a bit different with Linux (or BSD) and Windows.
[...]
> > Using usleep() is not an option here, because the granularity of
> > a sleep is greater than 1ms. Hence, programming will get *very*
> > slow.
>
> To me speed is not that much of an issue with small memory parts
> like the T2313.

I'm talking about something like 1s per byte here ;-)

Cheers
  Jan-Hinnerk