Greig Marshall

Penny Arcade - The MS Surface Pro

It seems the Surface Pro is a good substitute for the Wacom Cintiq 12WX. Similar size, similar price, both using pressure-sensitive Wacom systems, but the Surface is a fully contained computer into the bargain.

Until recently the Surface Pro hasn’t been viable since a bug prevented Photoshop from registering the pressure sensitivity. This has recently been fixed.



SpaceX rocket takes off, hovers for a bit, then lands. Show-offs.

Johnny Cash soundtrack.




About to start exploring d3.js and GeoJSON.js- JavaScript libraries for interactive online charts. I think this image is a glitch.

http://bl.ocks.org/mbostock/3763867



For his 1929 film The Woman in the Moon, Fritz Lang hired Hermann Oberth as a scientific consultant. Oberth had an assistant. The assistant was Wernher Von Braun. After a brief stint of trying to wipe London off the planet, Von Braun went on to lead up NASA’s Saturn V programme.

The film will be shown tomorrow night (16/02/13) at the Glasgow Arches with a score by Jeff Mills.

Fritz Lang & Jeff Mills: Back to the Future (BBC Arts & Culture)

Sonic Cineplex at The Arches, Glasgow



My alternative reading of Life of Pi.

Font is Potrzebie. Can be downloaded for free.



Continuing with the engineering theme, here we have Pinokio. Pinokio is a desk lamp familiar to anyone who has seen a Pixar movie, with life breathed into him by an Arduino controller and a few servos. The lamp head has a camera embedded and can recognise faces.

pinokio-lamp.com




Festo SmartInversion. It floats with helium cells in its six four-sided segments in a circular arrangement, turning itself inside out to move. It’s not CGI, this is footage of an actual test flight. There’s no practical use for it- it can’t drop bombs on people or fly at mach 10, but it is beautiful to watch.

More info: Flying object propels itself by flipping inside out (New Scientist, 20 April 2012)




Torus knot in a fabrication bay at my work. Test scene ahead of a bid proposal showing actual hardware in the bay.

The backplane is a 360° environment photograph produced from three photos taken with a Nikon D700 with an 8mm fisheye lens. The panorama was stitched in Panoweaver. The torus knot was rendered in 3ds Max, with a ground plane laid down and aligned to produce accurate reflections and collect ambient occlusion shadows giving the knot ‘contact’ with the environment map. The reflections are accurate due to the backplane being a wrapped-round environment map.



  • Data lifted from Google Trends.
  • Above is a graph showing the search frequency of four English terms associated with the Olympics- ‘cycling,’ ‘fencing,’ ‘rowing,’ ‘dressage.’
  • The popularity of each term is shown relative to the most popular ‘cycling.’
  • gym’ has been added as a seasonal reference. As expected, the term has a massive boost in popularity early on in the year.
  • Four area graphs are shown in front of each other. This is NOT a stacked-area graph.
  • The popularity of these terms was expected to spike during the Olympics.
  • All terms get a boost from the Olympics.
  • All terms except ‘dressage’ have annual cycles in popularity according to the summer in the northern hemisphere.
  • I only found out about the Pan-Arab Games when trying to explain the late-2007 cross-sport spike. The Commonwealth Games appear to have no impact on search behaviour.
  • It is possible to approximate absolute numbers from this graph, knowing that Google report the number of searches per term per region for their AdWords service.
  • ‘fencing’ may be an ambiguous term, however it seems to be used mainly to refer to the sport due to its Olympic spike being proportional to that of the other terms.
  • The popularity of English terms in reference to the Pan-Arab Games is hard to explain. It is the only non-seasonal spike for ‘gym’.
  • ‘fencing’ is more popular than ‘rowing’. This is surprising. Ambiguity of the term ‘fencing’ is ruled out as an explanation, as explained above.
  • Over time, all terms show a gradual reduction in searches. ‘gym’ is the exception to this.


Lytro Light-Field Camera

After a long wait and endless hype, the Lytro light-field camera is now available. It diverges from existing camera technology quite a bit, by recording the direction of light as well as its more conventionally captured qualities. This allows a picture to be refocused after it is taken. To achieve this, however, many pixels are required to record each final pixel of a Lytro image. A 10MP sensor is used to produce 1MP images, and that is the main limitation of the technology now and for the foreseeable future.

In the following picture, click on the cat you wish to have in focus.Note that this was not done using Photoshop gradient masks or similar traditional techniques.

If you have a look at the sample images, and have accepted focusing as a fundamental necessity of photography, you will quickly find yourself asking ‘WHAT IS THIS SORCERY?!’ The Wikipedia page on light-field cameras will go some way to clarifying how the technology works, possible applications, and current limitations.

Since the images are necessarily aware of the relative distance of each pixel recorded from the camera sensor, every picture has limited 3D data. If you watch this clip from Minority Report, you will realise that such 3D video could be captured by a light-field camera. Current cameras do not yet have sufficient bandwidth for video at the moment, but there is a lot of potential.

On a less ambitious scale, i.e. something which could be done today, the z-depth map from the image could be used in Photoshop to isolate parts of it for editing. A picture of someone on a busy street could have a light fog applied to remove distracting background information. This would be far easier than it currently is in image editing software, as it wouldn’t be necessary to fudge complicated gradient masks for perspective or curved surfaces. All that information would already be captured.

If you’re impatient to start experimenting with this, then fire up your 3D application of choice and create some images with a z-depth channel. Render a scene and save the image as an .EXR or similar multi-channel format and bring it in to After Effects. 3D artists have been doing this for years and the possibility of easily achieving the same image manipulation in photography is very encouraging. It is unfortunate that so much potential must come at the price of 90% of the sensor’s resolution, however.


14
To Tumblr, Love PixelUnion