DIY Computer Video Microscopy For Under $50 117
cybrpnk writes: "The QX3 Video Microscope may have been an obscure commercial failure as an educational toy, but it is widely available (for now, at least) as a fantastic tool/toy for any geek. The QX3 hooks up to a USB port and delivers live color 10X, 60X or 200X microphotos at 512x384 pixel resolution. Its kid-friendly software even makes time lapse videos a snap, like this one of TNT synthesis - a whole new way to blow up the lab, do not try this at home! Educators are doing amazing things with the QX3 in their classrooms.
Sourceforge even has documentation on the software command structure used by the QX3, so it may be considered an open source microscope. Get yours today for under $50 at surplus closeout or EBay before they're all gone!" The Toys-R-Us nearest to me has one QX3+ left (now with my name on it) at $30, so I hope it really does work under Linux. And it's a lot less complicated than building a Scanning-Tunneling microscope.
Linux support [mandatory] (Score:5, Informative)
First, Intel's official position [intel.com] on Linux support: CPiA chipset inforrmation [sourceforge.net] which the QX3 uses for its imaging. Note that there are still some problems getting everything working right. Back up your kernel.
If none of the above works, try some more generic hacks with the CPiA driver. It seems the biggest problem is getting the lights turned on...
Proscope by Scalar (Score:4, Informative)
I am thinking about getting one, but want to see if higher (500x) magnification is going to be available. See www.bodelin.com for details.
Photo Gallery (Score:4, Informative)
Intel's Developer notes (Score:3, Informative)
http://developer.intel.com/technology/itj/q4200
More QX3 Links and Info (Score:5, Informative)
Re:Proscope by Scalar (Score:2, Informative)
The Proscope accepts C-mount lenses, which are readily available cheaply. But the product does not have an adjustable stage, and you do not want to handhold at x500.
The Qx3 has only the three supplied magnifications, but DOES have a (Z-axis) adjustable stage, which is more practical.
Works fine with Linux. (Score:5, Informative)
I'm using a mostly stock SuSE 7.3 distro with the 2.4.10 kernel, the camera built into the QX3 is the same CPiA chipset that many other webcams use. I haven't done the necessary tweaking run the lamps, I just an external light. The "gqcam" program works fine for viewing/grabbing the images.
Re:I thought they were just toys... (Score:1, Informative)
Driver Software for OS X (Score:3, Informative)
His CVS repository is a little outdated, so download the drivers from the web site.
http://webcam-osx.sourceforge.net/
Re:low power microscopes (Score:4, Informative)
I recall long hours of drooling over everything in the Edmund catalog when I was a kid. Heck, their catalog (and web site) is still worth some time drooling over. So many toys, so little money, sigh...
How to get lights to work with Linux (Score:3, Informative)
The CPIA driver works fine with it for Linux, for viewing. Unfortunately, the code to turn the lights off and on has been commented out of the driver due to a buffer overflow.
I've got it running with the RedHat 7.3. I had to merge the driver from http://webcam.sourceforge.net with the drivers in the kernel source and recompile. I tried compiling the driver outside the tree but got bit by a bug in USB link ordering. Once you apply the updates you'll be able to use simple command-line statements to turn the lights off and on.
If you're going to try these patches on RedHat 7.2 or on some other Linux distribution, you'll have to merge the driver on http://webcam.sourceforge.net with your distributions's driver yourself.
For my code and images, see http://graflex.org/klotz/qx3 [graflex.org].
Of course, the best thing would be for the webcam.sourceforge.net people and the kernel people to resolve their differences and get the write code for /proc/cpia enabled. Until then, turning the lights on under Linux will be a DIY project.
"Pentium 4 Optimized" Ugh. (Score:2, Informative)
In the literature accompanying mine, it actually specifies that with a Pentium 4 processor, you get better resolution at higher magnifications, additional effects, and faster screen refreshes. I'm sure it's a deliberate screw up of the software if it doesn't detect a Pentium 4, (or other Intel processor, I suppose) and that pisses me off.
In fact, I can see the artifacts that are artificially added on the screen. There's a spattering of pixels that appear "dead" on screen: they're always some obviously wrong color that's slightly "off" what you'd otherwise expect to see at that point, and the "dead" spots don't move at all when the underlying target images are moved.
Artificially screwing up the quality of your products for AMD customers won't bring customers back, Intel! In fact, it does quite the opposite for me. Assholes.