A while ago, I posted a question about the state of the RAM registers in the ADAU1450 after reset: DM0 and DM1 RAM content after RESET ADAU1450
Dave responded that RAM is initialized just before the core starts running. That's what I found later also in the datasheet (i overlooked it at first...). But I don't think that both are correct.
I use a microcontroller to program the ADAU1450 via an I2C bus. I use the output xml file, generated by SigmaStudio and write all registers mentioned in that xml file. But for DM0 and DM1 I do it a little different. Because I use a lot of RAM for delay, that registers are all filled with zero's. And, according to the answers I got, I do not need to write zero's, so I skipped doing that when programming the DSP.
But after releasing the hibernate state, the DSP is giving strange signals :-) So, to test my theory, I wrote all zero's to DM0 and DM1 in the range SigmaStudio wrote in the xml outputfile and then I write the non zero values. With that sequence, the DSP is working fine.... So my conclusion is that RAM isn't initialized, or at least not at the time I expect.
Maybe someone can shine some light here? That would be appreciated :-) My program is running fine now, but I could speed up the bootprocess when omitting writing the zero's to DM0 and DM1 at startup (I2C isn't very fast...). OK, I could switch to SPI, but that isn't an option, I use more I2C chips :-)
Addition: I found the following in the datasheet under paragraph "
Thanks again for all your input
With kind regards,