At 12:34 28-03-05 -0800, you wrote:
>Hi Robert.
>
>Re:
>>>>>
>>I presume, that by 9600 baud  you mean 9600 bps. So 1.04ms by 9600 
>>should give 1000. But it gives 10 000. Something is wrong, isn't it ?
>
>No. At 9600 bps,  _bits_  are sent out every 1/9600s, or every 104us. 
>But 1 start bit,  8 data bits and 1 stop bit is 10 bits total, 
>therefore each _byte_ happens every 104us x 10 = 1.04ms. Therefore 
>you get a Tx INT every 1.04ms.
>
>>
>>So I presume 0.104 ms, which gives 104us.
>
>
>1.04ms. See above.
>
>>With 115200 it is 12 times less, which is around 8us.
>>with MCLK 7372800 one clock tick i 0.13us
>>
>>Maybe I don't understand sth, but msp430 is almost close to it's limits!
>>Cheers
>>robert
>
>At 115200 bps, a Tx INT happens every 86us (1.04ms / 12). That is of 
>course much faster than the situation at 9600 bps. Most code that 
>I've seen that runs at these high rates sends data directly and not 
>via the ISR, because the overhead of getting into and out of the ISR 
>is too high. Plus, the characters leave very quickly, so the Tx buff 
>is usually ready for another very soon thereafter.

Just out of curiosity, has anyone tried to use the DMA controller to -sort
of- create a UART with a transmit and/or receive buffer?

Nico Coesel


Reply via email to