My energy is being devoted to making money by designing the P1 into products for customers. I have not yet found its limits!
-Phil
You wrote a P1 object based on Goertzel's algorithm that could identify four spoken words. I fully expect that a P2 version will be able to identify phrases.
My energy is being devoted to making money by designing the P1 into products for customers. I have not yet found its limits!
-Phil
You wrote a P1 object based on Goertzel's algorithm that could identify four spoken words. I fully expect that a P2 version will be able to identify phrases.
Stuff like that seems to be driven by big data models and machine learning these days. An algorithm addressing the specific problem could achieve great results with 1/10,000 the data, probably. And no need for big scaling/summing networks.
Also appreciate the insightful recommended reading, http://holometer.fnal.gov/GH_FFT.pdf. Pulling signals out of the noise for gravity wave detectors has got to be cutting edge!
It's amazing what you guys program and build! I have been waiting for years for the p2 to come out and I have been trying to follow along as best as I can.! It should make a hell of a flight controller, even if its programmed in pure spin!
I'm hoping to get this thread back on track since it is about "showing off" what you can do with your P2 rather than talking about instructions and tools etc. With 100 eval boards out there you would expect some more feedback. For instance it would be good to see some video of microPython running some demos on VGA and keyboard etc.
Perhaps the GETCT posts can be moved into a new thread as I know that this information would otherwise be lost in a few short weeks.
These posts are "nice" but this thread is about the results of all this knowledge and how we put it to use at an application level , something that many have not posted about despite all the eval boards out there, whereas these GETCT discussions are the exact opposite.
While I love all these discussions, they should be in their threads where they can be found again. MODs, could you please move these GETCT posts into a Using GETCT thread or something please.
P.S. Guys, don't be afraid to reply to a post by starting a new thread when the discussion is begging for a new thread.
I dug out a couple of old 4d systems 128-g2 devices. They are a pain to work with. That is why I gave up years ago.
Decided to play with P2. Took hours as you have to talk to them in hex.
Tried to deconstruct Beau Scwabe's driver for the P1. Can't make it out.
If anyone has a suggestion please let me know of how Beau worked this out.
Thanks
Hey Peter I a took a break from P2 assembly and did some spin stuff as you will see above.
Question I am going back to working in the smart pins. I want to write routines for each of the "counters" like the P1. Do you have some basic
assembly code in P2 that is well commented so that I can understand it?
Comments
You wrote a P1 object based on Goertzel's algorithm that could identify four spoken words. I fully expect that a P2 version will be able to identify phrases.
Stuff like that seems to be driven by big data models and machine learning these days. An algorithm addressing the specific problem could achieve great results with 1/10,000 the data, probably. And no need for big scaling/summing networks.
We should have P2 Eval boards with the new chip, probably in mid August, to sell.
Peter has some ES left and wants to run a batch of P2D2 soon, but he is sick.
Mike
I'm up and about again, but catching up on other stuff, but maybe I will get a chance this week.
https://github.com/MarkTillotson/Prop2_FFT.
Also appreciate the insightful recommended reading, http://holometer.fnal.gov/GH_FFT.pdf. Pulling signals out of the noise for gravity wave detectors has got to be cutting edge!
Nice Job!
Shawn A.
Perhaps the GETCT posts can be moved into a new thread as I know that this information would otherwise be lost in a few short weeks.
These posts are "nice" but this thread is about the results of all this knowledge and how we put it to use at an application level , something that many have not posted about despite all the eval boards out there, whereas these GETCT discussions are the exact opposite.
While I love all these discussions, they should be in their threads where they can be found again. MODs, could you please move these GETCT posts into a Using GETCT thread or something please.
P.S. Guys, don't be afraid to reply to a post by starting a new thread when the discussion is begging for a new thread.
The GETCT stuff can be found here: http://forums.parallax.com/discussion/170316/changes-and-getct/p1
Decided to play with P2. Took hours as you have to talk to them in hex.
Tried to deconstruct Beau Scwabe's driver for the P1. Can't make it out.
If anyone has a suggestion please let me know of how Beau worked this out.
Thanks
Hey Peter I a took a break from P2 assembly and did some spin stuff as you will see above.
Question I am going back to working in the smart pins. I want to write routines for each of the "counters" like the P1. Do you have some basic
assembly code in P2 that is well commented so that I can understand it?
Thanks