No Sample Collected?

Natt,

There is a lot of information missing. Think of the things that you need to know to start your project and then to debug your project.

Did you get the sample program working that come with the DSK6713? At least one of those reads audio samples in and sends audio samples out. If you can get it working in the sample program, then you can get it working in your own program.

Regards,
RandyP

RandyP said:

Natt,

There is a lot of information missing. Think of the things that you need to know to start your project and then to debug your project.

Did you get the sample program working that come with the DSK6713? At least one of those reads audio samples in and sends audio samples out. If you can get it working in the sample program, then you can get it working in your own program.

Regards,
RandyP

Did you look at my project at all?

I fixed it.  I was returning a non-null value and was only collecting one sample and then exiting interrupts.  Ah, the joys of coding!

As for RandyP, I don't understand why your reply to my problem was so abstract.  Before, when I posted here, you asked me to move over to BIOS, which is fully supported by TI.  I didn't.  I translated this project from scratch using the C code and header files and ASM from the Chassaing book.  Maybe for a future project I will migrate to BIOS but it is important to my university that I understand how to do this so that they can possibly upgrade their CCS version.  If I don't know how to do this, they probably won't ever upgrade.  Even if the best path forward is to use BIOS. 

In half a day, you have solved your problem. Impressive.

What is your role at your university? Staff, professor, student, grad student? You certainly have skills that are valuable in the engineering field.

My advice has not been useful to you and my words have been offensive. I am sorry for that, and I hope it will not happen again. There are much better people than me on this forum, and some of them may be in a better position to take your code and tell you what to change.

I am curious about your translation project. If you are not changing from RTS to BIOS (or to a new version of BIOS) and you are not changing from one DSP to another, what needed to be translated from the Chassaing code example to make it run in CCSv5? Were there application code or library differences that were needed, or was the translation primarily in the CCSv5 project construction? The project construction is considerably different in CCSv4/CCSv5 compared to CCS 3.3, and that caused many of us to avoid moving to CCSv4 when it first came out. There are advantages with using CCSv5, but many of those are coupled with SYS/BIOS and even some features of the latest DSP cores and devices.

Regards,
RandyP

Well, it's been much longer than a half day to get to this point.  I made this post in a bout of frustration and quickly found my problem this morning.

Your advice has been good.  I understand your perspective.  Often, people ask for help before looking at the documentation.  They do that because the documentation is overwhelming and it's very, very difficult to pick a starting point when you are already knee-deep in your project.  I am not offended but I appreciate your candor and honesty.  That is a valuable trait.

I am an undergrad senior.  I took a course last semester with the c6713 board by Spectrum Digital using CCS v3.3.  When I asked for my own copy from TI, I received CCS v5.  Now, you and I both know that there is a vast improvement between the two IDEs.  With that improvement comes many more options.  I won't go through all of the details now but I plan on documenting what I had to do after this semester to save/help others that go through the same idea.  Meanwhile, I have three branches up on github.  The master branch is where I started.

My quest to translate the project using the Chassaing codes stems from the fact that our university uses the Chassaing book and has the c6713 Spectrum Digital boards and uses CCS v3.3.  Now, what benefit is it for future students to use this platform for development only to find a completely different IDE and approach used in industry/higher-ed?  Very little.  I wanted to go through the process and document it so that most, if not all, of the projects in the Chassaing book could be implemented in CCS v5--from scratch.  It may be done in vain, unless TI is willing to donate the IDEs for us to develop on or if I could secure a grant.

Who knows, what will happen from here but I do appreciate your kind reply.

Best,

Natt

Natt,

Natt Day said:

I took a course last semester with the c6713 board by Spectrum Digital using CCS v3.3.  When I asked for my own copy from TI, I received CCS v5.

TI does not support CCS 3.3 anymore. Getting copies of it are difficult, although it may still ship with some of the older boards, like yours.

Natt Day said:

I wanted to go through the process and document it ...

If you search the E2E forum for Chassaing, you will find a lot of people who have used Chassaing as a guide. And I do not remember any of them going through the positive effort to upgrade the flow to CCSv5. Once you have done this, your documentation on this forum or the TI Wiki Pages will be a huge help for a lot of your peers and those to follow.

Natt Day said:

unless TI is willing to donate the IDEs for us to develop on or if I could secure a grant.

I have been trying to figure out some really clever way to respond to this. But I guess it is easiest to keep it simple:

CCSv5 comes with a free license to use the included simulators or to use it with embedded emulation on TI supported EVMs. This should include the DSK6713, but I do not have one to try it on. In addition, there are low-cost JTAG emulators (less than $100 US) in the XDS100vN family (v1, v2, v3, so far) that can also be used with any custom board with no license fee.

So, it should satisfy your request to "donate the IDEs for us to develop on" since you can download and use CCSv5 for free.

If you want the boards at a discounted price, you would need to go through your faculty and the TI University Program. You may be able to apply on the TI University Program, also - go to TI.com and look for TI University Program. They may point you to more recent DSPs and development boards. For example, for the same $400 as the DSK 6713 board you can buy the TMDSEVM6678L which has a superset of the C6713 instruction set (supposed to be 100% code compatible other than peripheral drivers and reset vectors), eight (8) independent DSP cores in one device (like 8 x C6713) and each core runs at 1GHz or higher (compare with 300MHz for the C6713). Or for a lower cost DSP that is still 150% of the performance of the C6713, you could look at the C6748. The peripheral and I/O mixes are different on the EVMs and the devices, and the prices range from $14.00 to $200.00 for the devices, and the performance ranges from 150MHz to 1.2GHzx8. We have a very full menu.

Regards,
RandyP

Thanks again.  My main reaction is that there aren't any thorough text's like the Chassaing book to go along with these high-powered DSPs...that I've found anyway.  I haven't looked that hard to be honest.

I haven't seen the other folks that have used Chassaing's code but I doubt they've followed the canonical format for C programming...Chassaing doesn't either.  My project does, to the best of my ability.

So, that is the limiting factor.  If there was a text that could be used as a basis for a course, other than the Chassaing book then my efforts are in vain....