I started preparations today for an exhibition at the Nieuwe Vide gallery in Haarlem. I’ll be creating a 256×192 pixel image – a screenshot from a ’80s Sinclair ZX Spectrum game, all made by fluorescent bacteria. The image is going to be 2m x 1.4m in size – kind of luxurious for a Spectrum screen
But this takes a lot of effort in the lab. Currently I can work with 12x12cm plates, which can display an icon of 16×16 pixels. I need 16 of these in each row to get 256 pixels horizontally, and 12 rows to have the vertical 192 pixel resolution – all in all, 192 plates have to be made.
My plan is to create 50 plates per day, so that I even have a little room for failure – as 5 days x 50 plates = 250 plates. I started to scale up the quantities that have to be used – 6 liters of LB agar are used per day, in each liter:
- 250µl of carbenicillin (100µg/ml concentration -> 25µg in total)
- 55ml of overnight liquid culture
thus I need to create about 350-400ml of liquid culture as well every day.
Today’s results show that out of 1 liter of LB agar, I can pour about 13 plates. I made a total of 70 plates from 6 liters of LB agar.
A lot of preparations have already been done, thanks to Olga Crapels from the Arts & Genomics Center, and to Maarten de Smit, who has prepared the first batch of liquid culture, by growing out the bacteria permanently stored in deep freeze. Trudie Brouwer and her colleague, Arjan have established all the materials and supplies I need.
Meanwhile, Roland Vos from the Nieuwe Vide gallery is building the glass housing, that will act as a frame for the whole image. The original plan was to hang it from above, so people can just lie down below it comfortably on pillows, and enjoy the view – but, unfortunately this is not feasible at the moment, as it would take too long to get the hardened glass that is sure to hold up the weight of all the bacteria. Thus, the display will be set at 45 degrees – but, still, we’ll have pillows around.
Meanwhile, a new bug has been introduced in the Linux kernel, which prevents the ‘supposed’ cdc-acm driver from working with the Fab@Home controller firmware. So I have to use the usbserial driver instead. Unfortunately Oliver Neukum could not reproduce the issue, so there’s little chance of him fixing it.