posted by Bryon Moyer
In a discussion with Teledyne DALSA about their MIDIS MEMS process, we spent a few moments discussing how the ASIC die and the MEMS die are mated together. With this technology, the MEMS die has landing pads and the ASIC die gets micro-bumped and flipped and mated to the landing pads.
The question was whether this was done wafer-to-wafer or using known-good dice. The answer was wafer-to-wafer, since yield allows it and the costs are much lower. All pretty much reasonable reasoning.
But then we turned into somewhat more surprising territory. The reason it’s cheaper is that it’s a whole lot easier for a robot to take a wafer, invert it, align it, and stick it onto the receiving MEMS wafer. If you take a known-good-dice approach, then you first have to test the ASIC wafer to figure out which ones are good, then saw the thing up, and then pick out the good dice. You then have to place them on the waiting MEMS dice (which would presumably still be in full wafer form), placing them only on MEMS dice that have been shown to work by whatever testing could be done at the wafer level.
This is a lot of work and requires much more worker intervention than the robotic wafer-to-wafer process. More specifically, it requires more workers. Which costs more. We’re used to casting aside jobs with technology because, in the emotion-and-ethic-free world of finance, the dollar (or your favorite currency) is king and is all that matters. If jobs suffer while I make more money, it’s not my problem (because it’s not my job suffering).
It was as if they wanted to address this potential conscience twinge that they went one step further to justify the fewer-workers approach, and it went like this: These things are assembled in Southeast Asia. Southeast Asia has a bad reputation for employing child laborers. So by eliminating the jobs, we reduce the problem of child labor.
Bet you didn’t see that one coming! Nice to know we’re doing something good for the world…
posted by Bryon Moyer
We all know that if we want to be able to… well… transgress someone else’s private computer and internet stuffs, there’s a subterranean culture with a dress code involving black hats where, for the right price, you can get all kinds of tools that will open up all kinds of unsavory possibilities. These are the guys our computer security systems are trying to protect us from. They’re the guys your mother warned you about.
If we keep them out of our computers, then we’re ok. Right?
Oh yeah… there’s this NSA thing going around. Scooping up vast quantities of data (the exact amounts of which seem unclear, but all of which estimates seem to qualify as “vast”). Hmmm… and they’re not getting it from our computers, but rather from folks we pay for computer services (or, in some cases, from folks that offer the services for free). We can protect our data on our systems (or so we believe), but once it leaves us and starts traversing the net, we’ve lost control.
OK, not great, but at least we can encrypt our data and password-protect our files. Someone may intercept the transmission, but at least they won’t be able to read the payload, right? Assuming they’re not consorting with the black hats, anyway…
In order for… I’m not sure what to call the NSA types, since they don’t quite seem like white hats. Gray hats perhaps? In order for the gray hats to break into our actual messages, they’d need to figure out the key or some password or something. And that’s hard to crack – intentionally hard, or else it wouldn’t be secure. So we’re OK. Right?
It’s certainly hard to crack passwords and keys, but, given enough computing power, it’s doable. Of course, software takes time to execute, even when using GPUs; something that’s accelerated in hardware would be just the ticket!
And, voilà! Pico Computing has just announced an FPGA-based acceleration system for cracking passwords. Oops! Wait, sorry – “cracking” is an ugly word. “Recovering” is the preferred euphemism. As in, “Bob left the company and didn’t give us his password. How are we going to open his files?” Why, recover the passwords, of course. One obvious corporate use model. How often is that needed? Hard to say. Probably a lot less often than gray hats might want to recover a password, however.
This is where it’s easy to slip into the Land of Evil. Let’s be clear here: I’m not saying Pico Computing or their technology is being evil. (I know, I know: “Technology isn’t evil, People are evil.”) In fact, Pico Computing isn’t really doing the cracking; they’re accelerating tools from a company called Elcomsoft. Elcomsoft focuses specifically on locked documents that require a password to open, so it’s not so much about decrypting encrypted traffic.
Nonetheless, amidst a sea of technology announcements promising security, I think this is the first announcement I’ve seen that gleefully promises to help compromise security. Although they don’t really say it that way, of course… You can see what they do say in Pico Computing’s release.
posted by Bryon Moyer
We’ve talked before about indoor and pedestrian navigation and the challenges they pose. As part of the ongoing industry effort to crack that nut, Movea recently announced a demonstration of their indoor navigation skills in France and South Korea. I was trying to parse their announcement carefully to catch the nuances of what they were claiming.
First of all, they claim that this is a “first,” but I think the key qualifier is that this is the first time their capability has been proven in a manner optimized for cellphones.
Second, at times it sounds like this was strictly a dead-reckoning solution, at times not. To be clear, this was a dead-reckoning-plus-map-matching (data fusion) solution. Specifically missing was the use of WiFi or other signals to be used as beacons. It was sensors and maps only.
Third, they mentioned collaboration with SCNF in France (the local rail operator) and a subsidiary of SK Telecom in South Korea. The latter in particular made me wonder whether this solution somehow required the participation of third parties (something I’ll address more fully in a piece tomorrow). The answer is that, fundamentally, no, no third-party involvement is required. What they got from these guys was maps of the train stations. This was the data for the data fusion component. At some point, all such facilities will have been mapped (and Google will probably have the maps), but that’s not the case yet.
So, rounding it out then, Movea demonstrated their platform, optimized for cellphones, which combines dead reckoning and map matching, by having people successfully navigate through busy train stations using their solution. Sounds like a pretty good result. You can get more info in their release.