Sunday, March 17, 2013
Thank You, bunnie
It's a quadruple-post kind of day. I want to give a shout-out to bunnie and let all of my non-existent followers know about him. Last week, he released his book Hacking the Xbox: An Introduction to Reverse Engineering as a free PDF in honor of Aaron Swartz. I'm already about half-way through the book, and even though the concepts of the original Xbox that are described are not exactly cutting-edge, I think it's written very well and have already learned a lot about the basics of embedded digital security. I happen to own an old Xbox as well, so one of the first things that I plan to do this summer is to step through this book again while carrying out the hacks myself. It's not like I'm in desperate need of another old Linux box, but it should still be a lot of fun. And what other excuse does a hacker need besides that?
Thank You, Electronic Surplus
I want to dedicate this post to Electronic Surplus, because they just sent me a 40-character, 2-line display from them as a give-away. It was advertised on my favorite podcast, the Amp Hour, a couple months back, and I had nearly forgotten about it by the time this beauty showed up in the mail.
I'm not sure what I'm going to ultimately use it for yet. I actually haven't even gotten it up and running. It's a tad too big to fit in my power supply, so I guess I'll just have to wait for another project to come along before I put it to use. In the meantime, though, one of my buddies challenged me to run the countdown sequence from Predator on it, so that's another thing on the to-do list. I'll be sure to post results as soon as I have that up and running.
RPRAC: RTL Power Reduction Approach Comparator
Normally, I try to not publish schoolwork on this site, mostly because the ideas for assigned projects come from the professor and not me. However, the final project for my Advanced VLSI Techniques class allows for students to present an original idea, so I feel that sharing this work is appropriate.
My idea is not exactly groundbreaking. Rather, I aim to provide a convenient tool for students studying different methods for decreasing power consumption in integrated devices. I call it RPRAC, which stands for RTL (Register Transfer Level) Power Reduction Approach Comparator. It's a program that applies one or several power reduction techniques to a provided logic design separately and measures and compares the effects that each technique has on power consumption, chip area, and signal timing.
Essentially, my plan for this program is to simply write a script that employs other VLSI design tools to do the actual heavy-lifting. As a student of a graduate VLSI course, I have access to machines with actual professional software suites, so I'm using ModelSim by Mentor Graphics for the Verilog testing and simulation, Synopsys Formality for functional verification, and Synopsys Design Compiler for netlist synthesis and measurement. A simple chart describing the general flow of the program is shown below:
I plan on implementing this in Python, which I've only used for mall experiments so far. Check out this GitHub page to check on the source. I might eventually post some wiki pages on there as well.
My idea is not exactly groundbreaking. Rather, I aim to provide a convenient tool for students studying different methods for decreasing power consumption in integrated devices. I call it RPRAC, which stands for RTL (Register Transfer Level) Power Reduction Approach Comparator. It's a program that applies one or several power reduction techniques to a provided logic design separately and measures and compares the effects that each technique has on power consumption, chip area, and signal timing.
Essentially, my plan for this program is to simply write a script that employs other VLSI design tools to do the actual heavy-lifting. As a student of a graduate VLSI course, I have access to machines with actual professional software suites, so I'm using ModelSim by Mentor Graphics for the Verilog testing and simulation, Synopsys Formality for functional verification, and Synopsys Design Compiler for netlist synthesis and measurement. A simple chart describing the general flow of the program is shown below:
I plan on implementing this in Python, which I've only used for mall experiments so far. Check out this GitHub page to check on the source. I might eventually post some wiki pages on there as well.
Automate 2012 Overview
It's been quite a while since the last (and first) post, but Spring Break is upon me, which means I finally have enough time off from classes to come up for air and publish a few new posts. The first of these is some footage from Automate 2013, a conference of industrial automation providers. Floor passes were free, so I dropped by for a couple of hours in between classes and acquired the following photos and videos.
The first of these is a short clip of the Staubli TP80 picking and placing Tic-Tac boxes. This picker machine pivots its arms at mind-blowing speeds, topping out at 200 picks-per-minute. I heard that the one shown above was operating slightly slower than that due to the lack of reinforcements to secure the body. They even had to cut out a square of the carpet underneath the enclosure it to ensure that it wouldn't tip itself over. More details about this machine can be found here.
Next up is the Fanuc M-2000iA, "the world's largest and strongest six-axis, modular construction, electric servo-driven family of robots designed for a variety of manufacturing and systems processes" (more info here). I'll believe them, as this was easily the biggest robot present on the conference floor. The one pictured above spent its time continuously flexing and rotating a locomotive axle above the heads of the surrounding crowd.
Again from Fanuc, we have the M-1iA, which is a 6-axis gonkotsu (fist) robot. It's another picker robot, and while it's not as fast nor strong as the Staubli shown above (0.5 kg maximum load versus 1 kg maximum load), it supposedly has better repeatability (+/- 0.02 mm versus +/- 0.05 mm). It also seems to employ an integrated vision system to improve accuracy and reliability. More info here.
Last but certainly not least is this gem from Industrial Perception, Inc. Lacking from their website is a lineup of standard models, so they might only offer custom solutions. The robotic arm shown above uses a camera near the gripper to scan a mixed bin of items for one that matches a given model. It looks over the entire bin, highlights the item with the closest match, and then dips down to grab it. Due to the overhead video feed showing the robot's view and object tagging process, this display was my favorite from the entire conference.
Next up is the Fanuc M-2000iA, "the world's largest and strongest six-axis, modular construction, electric servo-driven family of robots designed for a variety of manufacturing and systems processes" (more info here). I'll believe them, as this was easily the biggest robot present on the conference floor. The one pictured above spent its time continuously flexing and rotating a locomotive axle above the heads of the surrounding crowd.
Subscribe to:
Posts (Atom)