Quoi de neuf 2015

From Eric

(Difference between revisions)
Jump to: navigation, search
m
m
Line 13: Line 13:
A few things, anyway:
A few things, anyway:
-
- I have bought a new RPI 2, installed opencv on it and done a few experiments about line following using Canny and Hough. The idea is to be able to run some benchmark to compare performances of the PI with respect to a Zynq.  
+
* I have bought a new RPI 2, installed opencv on it and done a few experiments about line following using Canny and Hough. The idea is to be able to run some benchmark to compare performances of the PI with respect to a Zynq.  
-
- About the Zynq... I have also bought a Zynk board at MyIR. A nice little thingie comes with 1GiO of SDRAM, a Zynq XC7Z020 and many devices (Ethernet, Usb, a buzzer, a CAN network, etc.). The board comes with Ubuntu pre-installed. The idea is to run some hard/soft combination at bare metal level. Hopefully, I have found a nice tutorial [https://wiki.hackerspace.pl/projects:zturn-hackers:helloworld#in_vivado here].  
+
* About the Zynq... I have also bought a Zynk board at MyIR. A nice little thingie comes with 1GiO of SDRAM, a Zynq XC7Z020 and many devices (Ethernet, Usb, a buzzer, a CAN network, etc.). The board comes with Ubuntu pre-installed. The idea is to run some hard/soft combination at bare metal level. Hopefully, I have found a nice tutorial [https://wiki.hackerspace.pl/projects:zturn-hackers:helloworld#in_vivado here].  
Building an application on the Zynq is quite complicated since you have to generate the configuration of the FPGA (the bitstream), a FSBL, a specific BSP, your application... to end up with the "boot.bin" file that is loaded by the board. Hopefully, Xilinx' Vivado hide most of the details. It is nice as long as it works...
Building an application on the Zynq is quite complicated since you have to generate the configuration of the FPGA (the bitstream), a FSBL, a specific BSP, your application... to end up with the "boot.bin" file that is loaded by the board. Hopefully, Xilinx' Vivado hide most of the details. It is nice as long as it works...
-
 
+
* I have done a few experiments using SystemC and SystemC AMS (the analog part). You would certainly not replace Spice by SystemC AMS, but SystemC TLM + SystemC + SystemC AMS makes it possible to model and simulate a complete system at different levels of abstraction.   
-
- I have done a few experiments using SystemC and SystemC AMS (the analog part). You would certainly not replace Spice by SystemC AMS, but SystemC TLM + SystemC + SystemC AMS makes it possible to model and simulate a complete system at different levels of abstraction.   
+
   
   
== March 2015 ==
== March 2015 ==

Revision as of 14:08, 3 May 2015

Experiments carried out in the past (from 2011) are described hereafter:

"Quoi de neuf" means "What's new"...

For 2015, everything is described below...

Contents

April 2015

No time to do anything about "real electronics".

A few things, anyway:

  • I have bought a new RPI 2, installed opencv on it and done a few experiments about line following using Canny and Hough. The idea is to be able to run some benchmark to compare performances of the PI with respect to a Zynq.
  • About the Zynq... I have also bought a Zynk board at MyIR. A nice little thingie comes with 1GiO of SDRAM, a Zynq XC7Z020 and many devices (Ethernet, Usb, a buzzer, a CAN network, etc.). The board comes with Ubuntu pre-installed. The idea is to run some hard/soft combination at bare metal level. Hopefully, I have found a nice tutorial here.

Building an application on the Zynq is quite complicated since you have to generate the configuration of the FPGA (the bitstream), a FSBL, a specific BSP, your application... to end up with the "boot.bin" file that is loaded by the board. Hopefully, Xilinx' Vivado hide most of the details. It is nice as long as it works...

  • I have done a few experiments using SystemC and SystemC AMS (the analog part). You would certainly not replace Spice by SystemC AMS, but SystemC TLM + SystemC + SystemC AMS makes it possible to model and simulate a complete system at different levels of abstraction.

March 2015

(see above)

February 2015

(see above)

January 2015

  • 10/01/2015
    • I have started a small experiment with sound synthesis chips. Right now, I have hooked a YM2612 to an Atmega. My objective is to be able to read and play VGM files.
    • To familiarize myself with VGM, I am writing a small VGM parser in Python.
    • There are quite a large number of sound synthesis chips: SN76482, the Yamaha family (YM2608, YM2612, YM2149, YMF262, etc.), the AY-3-8912, etc. Wikipedia has a very good set of articles about those sound chips.
Personal tools