Blog - All Comments

RE: And the winner is... Monte Chan with project "Code Name"

ssilberhorn — Sun, 22 Jul 2012 07:02:06 GMT

Checkout the post on the TI E2E page...

http://e2e.ti.com/blogs_/b/mobile_momentum/archive/2012/07/19/simplelink-wifi-contest-_2d00_-ultimate-roadtest.aspx

RE: Project "code name" PostScript 011

COMPACT — Mon, 16 Jul 2012 22:07:40 GMT

Hi DAB,

The buffer issues are only related to the CC3000 Buffer allocated and within the MSP430 used in the Basic Wi-Fi Demo and not the additional FRAM Buffer.

What I have worked out thus far is the buffer needs to larger than the original value 5 and no greater than 256.

I've been getting everything else working and have not had the opportunity to look further into the reasons why.

I just looked up TI e2e forum and it mentions that the minimum buffer size for the current driver is 119 bytes.

With a bit of effort I think I will be able to find the cause and expand the buffer as necessary up to Wi-FI's MTU of 7981 bytes.

(I didn't realise Wi-Fi datagrams could be so big!)

Cheers

Very Compact

RE: Project "code name" PostScript 011

DAB — Mon, 16 Jul 2012 20:33:07 GMT

If you get some time, I think you can decouple these events to get a better control and a smoother transition between the various tasks.

The MSP430 provides a great set of independently operable timers and event triggered ISR options. By using the full device capabilities, you can decouple the events into a cleaner structure, which might also alleviate some of your buffer issues.

Just a thought,

DAB

RE: Project "code name" PostScript 010

DAB — Sun, 15 Jul 2012 22:01:33 GMT

Getting buffer size correct is a common problem I have observed in most distributed architectures.

It usually takes a while to assess the message traffic to see if your are sending too much data or too rapidly updating the status readouts.

Most projects I reviewed end up adjusting buffer size and message timing to work around the undocumented limitations of some of the components.

Overall, you have done a good job keeping the readers updated on your progress and issues. I think we all learned a lot about the MSP430 line during the road test.

Good job,

DAB

RE: Project "code name" PostScript 006

DAB — Sun, 08 Jul 2012 20:04:58 GMT

I agree.

If you are switching any device on the main power lines, you can get some very interesting line reflections. I have lost more than one device to an EM spike and not enough filtering.

I do like the optical isolators with the built in voltage isolation. I need to take a closer look that the TI product line for these.

Thanks for the update,

DAB

RE: Project "code name" PostScript 002

DAB — Thu, 05 Jul 2012 19:11:32 GMT

If I read the diagram correctly, it looks like you could use this circuit as a buffer between 3.3V and 5v devices.

If true, this would be a great way to interface legacy hardware to newer components within a design.

It would be a much cleaner approach then some solutions I have seen on the web.

Good post.

Thanks

DAB

RE: Project "code name" Part 040

DAB — Sun, 01 Jul 2012 21:02:33 GMT

I agree, it is better to take the time and do a good job in finnishing the project and documentation.

Either way, I think you did a very good job on the project, but you might have over estimated how much you could get done.

I am looking forward to reviewing the completed documentation, so do a good job on it, I might want to borrow some of your ideas for one of my own projects. [View:/resized-image/__size/16x16/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-33d02175-c0f9-4b51-b821-696dd0472667/contentimage_5F00_2516.png:16:16]

Good luck in the final voting.

DAB

RE: The Bread & Butter : 11

DAB — Thu, 28 Jun 2012 20:38:51 GMT

Good overview and demo.

I liked the way you repurposed exisiting devices to include them into your network. It helped show the versatility of your approach. It also showed how you can include legacy items into your system with minor changes.

Overall, I think your video presented your architecture, the components, and their connectivity, which is a very good achievement in the length of time you had. I was struggling with your end goal until I watched the video, now everything makes sense and I see where you were going.

All in all, I think you did a good job.

Good luck,

DAB

RE: The Nice Asgard Project Part 11 - Adding the Icing

DAB — Thu, 28 Jun 2012 20:26:21 GMT

Good overview presentation.

I would suggest you slow the video a little, some of us do not read as fast as we used to. [View:/resized-image/__size/16x16/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-bd7553db-e958-4f08-a941-635609077c25/contentimage_5F00_938.png:16:16]

I like your simple approach and keeping to off the shelf tools TI provided. From what I have seen, their libraries are very easy to use and the code works well right out of the box.

Are you going to have a video showing the lights and the blinds moving? That way we could see the code in operation as it uses your architecture to implement the commands.

Just a thought,

DAB

RE: Project "code name" Part 033

DAB — Thu, 28 Jun 2012 20:12:14 GMT

I had read about both the sensor and the booster pack, but I had no idea the actual device was that small.

I guess I should have once I read it was MEMS. Those things are awesome, but too small for my big hands. [View:/resized-image/__size/16x16/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-0e5160fb-2957-4212-8a74-29106bf0b932/contentimage_5F00_1.png:16:16]

Looks like you are making good progress going down to the wire.

Good luck,

DAB

RE: Project "code name" Part 032

DAB — Thu, 28 Jun 2012 20:05:41 GMT

I wouldn't call it the ancient art, but I suggest you only wire on the long ends of the posts.

The short ends were supposed to be soldered into the board.

Just a thought,

DAB

RE: Project "code name" Part 031

DAB — Thu, 28 Jun 2012 20:03:39 GMT

Good clean simple approach.

You can always expand the channels as you need them and then implement the new devices via auxiliary controllers as needed.

I like the way you kept the animation simple. You just run two canned scripts down to the toy so it looks like the animal is doing the talking while the rest of the system does its own thing. Simple but effective. I suspect you could also add additional scripts to have different performances based upon which sensors detect problems.

I like it,

DAB

RE: Project "code name" Part 030

DAB — Thu, 28 Jun 2012 19:57:40 GMT

Makes sense, the common serial interface keeps your options open and the C2000 takes care of implementing your lighting commands based upon the parameters you send.

Cool,

DAB

RE: Project "code name" Part 029

DAB — Thu, 28 Jun 2012 19:53:34 GMT

I did not know you could generate voice files using the VBA with microsoft applications.

You have made my day. I have been thinking about that subject for one of my projects, now I know where to go look.

I knew that the Adobe reader would talk and I kind of knew that you could set up the documents to talk to you so I guess it was under my nose the whole time.

Still, you have given me something new to play with.

Thanks

DAB

RE: Project "code name" Part 028

DAB — Tue, 26 Jun 2012 20:17:27 GMT

Good detailed update.

I agree that you can get by with just sending the commands as long as you can verify their implementation without repeating the commands.

Also building upon the existing code is a good place to start. It helps eliminate unknowns.

Good luck,

DAB