Entries : Category [ Computers and Technology ]
[Miscellaneous]  [Computers and Technology]  [Travel]  [Education]  [Hacks]  [Robotics]  [Science]  [Programming and Software]  [iPhone]  [Digital TV and Video]  [Intellectual Property & Copyright]  [Personal] 

06 April

Event for startup company founders and possible investors

I attended LaunchFestival in San Francisco. It was a startup event with pitches from, services to small companies and talks on topics from fund raising to hiring. It was another interesting example of the hyper-energetic startup scene here and attracted a large range of founders and service companies. Apparently as many as 12,000 founders and funders purchased a ticket to got a free pass.

On the other hand the noise level in the startup scene is very high with a plethora of silly ideas, wannabes and short-attention-syndrome. That's the nature of the startup scene and by-and-large the organizers did a great job of selecting interesting participants, especially since it's almost impossible to accurately predict which zany ideas will eventually get traction.

I got to meet a couple of different representatives from Chinese investment houses who seemed pretty interested in some of the robotics ventures that I also liked. Overall Launch Festival was great and provided an interesting aspect of the scene here at an exceptionally large scale. I also made a couple of useful contacts and saw a couple of cool demos.

San Francisco Launch Festival presentation on scaling up
Sponsored by SeedInvest (Click to expand)

San Francisco Launch Festival demo pit
(Click to expand)

By Gregory Dudek at | Leave a comment |    
18 June

I have been looking at the competitive Esports world recently, that is the world of video game spectatorship where people watch top players display their prowess at various computer-based games. This follows the model of professional non-video sports like tennis, soccer and football, but based on modern games that arguably have a broader appeal in today's world. This has been an ongoing commercial activity for years, but it is only gradually coming to the attention of the general public (i.e. people over the age of 30). Many people are surprised to find that top video game tournaments draw thousands of spectators, both in person as we as on-line and have prize pools that can reach up to into the tens of millions of dollars (for example the Dota 2 International 2016 had a prize pool of over $20,000,000 USD). My first exposure to this world was when I heard one of my graduate students and his girlfriend were traveling to Toronto to watch a World of Warcraft live stadium event with tens of thousands of spectators. Today's hottest games include Dota 2, Counterstrike and Overwatch.

Some big investors and companies are seeking to get into the Esports craze either to promote their games, build their audience, or use it as an advertising medium for other products. The scene includes huge corporate players in the gaming scene like Blizzard Entertainment, traditional media organizations like ESPN or Turner Broadcasting that are dabbling with this new market and numerous smaller companies looking to establish themselves. Some pro Esports teams have also partnered with traditional sports teams like the Detroit Renegades.

Image by Sam Churchill

The fact that Esports is based on the Internet, especially around streaming on Twitch, means that a panoply of small players can participate and try to find their place. This includes not just the companies that run and sponsor the big tournaments, but announcers, players, media advisors, managers and game play analysts.

Having watched a few tournaments of different kinds, I am sometimes surprised by some of the ways the hard core aficionado community fails to reach out to the more casual viewer. The broadcasters core audience is the hard-core fans who play themselves and are immersed in the game, but the much larger community of casual viewers represents a huge untapped resource, just as tennis broadcasts make most of their money from people who can barely hit a tennis ball.

With that in mind, I have a few suggestions for twitch streamers looking to reach out to the non-expert audience.

  • Note that viewing audiences take a while to build up, and non-experts need to be curated and hand-held much more than passionate experts who are willing to actively search out information.

  • Advertise streaming broadcasts repeatedly, and update web sites actively to reflect current information, schedule changes, and news. This means hourly or more when tournaments are taking place.

  • Streaming broadcasts, such as on Twitch, need to start on time and, in fact start early even if it is mere filler content.

  • Pre-game "countdown" footage is often a still image with some music. It would be better to have live discussion or commentary, not mater how vacuous. In fact, to engage non-experts this can include basic tips and a synopsis of game play and what to watch for.

  • Amusing nicknames are a classic part of Internet culture, but when a game features a faceoff between "Toiletman" and "Dr. PeePee" it's a lot harder for viewer to leave with the feeling that it was a serious game between professional players. (Dr. Peepee selected his name back in 7th grade, gradually became well-known with that handle

By Gregory Dudek at | Leave a comment |    
07 December

Fall is always a busy time in our lab and this year is definitely no exception. We are several different kinds of projects coming to fruition some of them getting ready for testing in the ocean in January. These include some small embedded devices that report data in various ways, as well as some larger oceanborn crafts that will work in collaboration with other vehicles. In addition we have some new theoretical work, a couple of big grants in play, and some exciting you industrial collaborations.

By Gregory Dudek at | Leave a comment |    
24 June


The International Conference on Robotics and Automation took place in Brisbane, Australia this year. Despite some concerns that the regional robotics community in Australia was smaller than in other regions, the attendance at the conference exceeded any prior year. This is, no doubt, a reflection of how robotics (as well as AI and vision) are growing in significance, impact and relevance.

The industrial trade show that accompanies the academic tracks of the conference was only a small event that had only limited appeal. It has not grown to a major phenomenon with many companies of all kinds participating. This include companies that address leading edge robotics technologies, as well as organizations that use relatively small-scale (yet important) industrial automation. Some of the attendees seeks to advertise and sell technologies, while others focus on attracting attention to better hire new employees or maybe even build partnerships.

With over 1,000 research presentations, it's hard to summarize the diverse kinds of work being done. Field robotics got a fair bit of attention, which aligns well with some of the challenges in the Australian environment, as well as Canada. Rod Brooks gave a great keynote talk which commented on the importance of considered human needs and interactions, while at the same time delivering a stinging critique of classical HRI research. In fact, his critique related to much of the academic community which he felt was overly focussed on obscure or unrealistic quantitative research benchmarks, as opposed to having a greater impact on real-world needs and performance challenges. Rod has defined much of his career by railing against contemporary research trends, which is a great way to get attention. I feel his critique had some truth in it, yet also was a bit too hyperbolic since conducting scientifically valid quantifiable research is critical to ongoing progress, especially as the field matures. Doing stuff that really works in practice is critical, which was part of Rod's message, but it's also important not to use hype and public relations as the arbiters of progress. Anybody who doubts that should just reflect on the Theranos scandal.

On aspect of the conference that provoked a mixed emotional reaction from me was the fact that only a very small fraction of papers could get full-length oral presentations. While reviewing standards have increased and papers were fully-reviewed and generally of pretty good quality, there was simply not enough space or time to allow them to be presented orally and poster-style presentations were the order of the day. This seems like a development that is not likely to be reversed and we are planning something similar for ICRA 2019.

By Gregory Dudek at | Leave a comment |    
15 July

Why embedded systems like the raspberry pi should not allow swapping to an SDcard

SDcard and Flash memory errors

SD card corruption and eventual failure is common on embedded systems like the raspberry pi. I've realize there is a lack of consistent and reliable information on this. This post describes the problem and provides solutions.

This post has 3 parts: description of the problem, practical implications, and tricks to disable swap on the raspberry pi and related linux devices. People who just want a solution can skip to part 3.


In our lab we have a range of systems from expensive underwater robotics to small embedded IoT devices all of which use SD card or flash memory. There are modes of operation that are acceptable on a normal desktop or laptop that will gradually degrade and eventually destroy an embedded device after a few thousand cycles of writing to a given cell. Since the damage is gradual, it can easy to overlook the cause during setup and testing, and just blame it on "bad memory."

The root issue is that solid state memory wears out with successive changes to the contents, and this happens on a per-memory-cell basis. Modern SD and flash memory have sophisticated internal algorithms for "wear levelling" that try to reduce the impact and extend the life of the chips, but these cannot eliminate the problem totally, since it's intrinsic to the physics of the devices. Hard disks, by the way, can also wear out, but not generally as a function of the read-write behaviour. The key idea behind wear levelling it to change the physical location used for frequently-changing data, to spread the use.
[ Technical aside: As of 2011, 25nm MLC NAND lasted for about 3,000 write cycles. As of 2018 some sources suggest that single level cell NAND (SLC) can last for 10,000 cycles ]

For normal use, only a small amount of the device is written at a time and if this is moved around the device it can be many years before the flash memory storage wears out. Without wear levelling it's easy to write a malicious program that "burns through" a specific cell in just a few minutes (although OS cacheing policies also try to preclude this).

Note that there are differences between expensive SD cards and cheap ones. These differences include both the robustness of the memory elements themselves, their speed and also the nature of the wear levelling (e.g static vs dynamic).

Practical implications

Normal file IO will take many years to damage a modern (wear levelled) flash device. Swapping, however, can be much more destructive. The issue is that swapping is a mechanism used by the operating system to compensate for insufficient RAM (memory) and is depends on writing large amounts of data to "disk", and sometimes doing it a lot. This heavy usage could damage a flash device like an SD card in just a few months.

Note that swapping to a "slow device" will also make the system run slow, but on the other hand in can preclude a hard crash that could occur if the system runs out of memory.

The commands below require root (or prefix each command with sudo).

Turning off swapping

dphys-swapfile swapoff
dphys-swapfile uninstall

and also (to make it permanent):

update-rc.d dphys-swapfile remove

Turn off swapping temporarily

The following two lines work (or npt) depending on the kind of linxu distribution you have (i.e does it use systemd):

swapoff -a

systemctl disable dphys-swapfile

Totally disable the swapping mechanism for keeps

apt-get remove dphys-swapfile

Aside from swapping, you can also make the entire filesystem read-only that preventing any writes (but also limiting the kinds of things you can do).

By Gregory Dudek at | Leave a comment |    
28 April


Well, it's been a long time since my last update. I'm working at/with Samsung these days and it's been filled with new opportunities and challenges. They are building a new AI lab in Montreal and staffing it up.

By Gregory Dudek at | Read (1) or Leave a comment |    
Prev  1   2   3   4   5   6   7   8   9   10   11   12   13   [14]