SigmaDSP вопросы

SigmaDSP questions

Активно 1 Ответ 130 Просмотры 2016-11-09 Работа с SigmaDSP

Customer will initial project to study SigmaDSP.

Could you reply some doubt:

As attached pictures,

1. Should  we adjustment each volume and EQ  or not ?  correspond to a register can be an external MCU with I2C ?
2. If you use the external MCU to adjust the value of the register, how about the delay time or latency?
3. When using the external MCU to adjust the register value, SigmaDSP can complete the performance between 23-85 changes (EQ and volume) or not?for example, from 23 to 85
4. Use external MCU to adjust the value of the register, for example, from 23 to 85, the process will not be crack(much distortion) or not?

For example, a tone on each level of a sonic boom.


Kevin Tsai

guest replied
3 Years

Hello Kevin,

These questions can have long answers so I will start with simple answers. Some of these questions have been answered before on the forum so I encourage you to search the forum for a quick answer.

SigmaStudio does two things. First, It allows you to create a DSP program graphically and it compiles and loads in the program. The second thing is that it allows you to control object in the DSP program using the GUI on the SigmaStudio screen. It also can gather information while the program is running so you can do things like a level meter or a readback of data from the program. This communication is via I2C or SPI and a microcontroller can communicate in the same exact way. There are features in SigmaStudio that allows you to see these communications and see the settings on each object. The Capture window shows you the communications between SigmaStudio and the program running in the DSP. Always keep in mind that the program is running in the DSP and not in SigmaStudio.

Then on bottom of the capture window you will see two tabs. One is the output and the other is where you can see the parameters of the DSP program in real time.

The delay time or latency using an MCU is no different from SigmaStudio. It is communicating directly with the DSP program. The latency is the time it takes for the I2S/SPI transaction and then the change will take place in the next sample period or sooner.

EQs are a little different. If you change the filters you have to load in new filter coefficients and that must be done using the Safeload feature. They slightly differ by the DSP and you did not specify which DSP you are looking at. There is a lot more to this subject but I will leave it here for now.

Your question 3 I am a little uncertain as to what you are asking about. If you mean that volume changes in SigmaStudio can be done in 23 -85 discrete changes? When using SigmaStudio you often grab" the control and move it. This produces a large number of individual moves but you do not have to do it that way. You can right-click on the volume control and input a new value directly and then it will be done in one step. Now there are different volume controls. Some with no slew where the change is a move in one step, or you can choose a slew type where the move is done in many small steps to make it have no zipper noise.

Your question 4 is where you use the safeload feature for updating filters. This will change the filter coefficients at one time when it is not being used. This will prevent clicks and sonic booms. Other controls that need to be changed often you can use the slew type to make it not click.

Dave T