In the end i didn’t talk as they had a full timtable (which over-ran) but i managed to talk to several people there about my project and the problems i was having with it. The same solution was offered several times (including the first person i spoke to) – change the software i’m using to Magics STL. However this is easier said than done. The full version costs £3000. More on that later – here’s a summary of the talkers relevent to my project and the day.
Fistly the facilities at the Centre for Fine Art Print Research are amazing – they work closely with both industry, bristol university and art projects (including their own) – a third on each. They charge for use of various machines providing tecnical support – and research all types of esoteric 2D Printing (and more recently 3D) in alliance with people like Hewlett packard & Z-Corps. They are cutting edge – and support about 20 researchers – and expanding.
In 3D printing they have two z-corps machine (310 & 510) some small CNC milling machines and also several Laser Cutters. They showed us several full color 3D models of toy like designs which i thought were fantastic – however i got the impression that the CFPRers weren’t satisfied with the texture of the sculptures and the z-corps powder. A lot of their research is focussed into how to improve the material quality of the surfaces. Several researchers/artists throughout the day showed their experimental techniques often which involved exchanging the z-corp powder for their own formula – something which could then be prototyped and then on top of that fired (or glazed and fired) – so combining traditional pottery techniques with advanced 3D modelling techniques. Another technique which cropped up a few times was enamel – and how it could be coated onto any sort of protoyped model (eg steel jewelry templates) and this produced a durable and pleasant surface.
So the emphasis of the day was definitely on surface quality. Which is a natural print/sculptural/craft concern but is not my highest priority at the moment. The objects created were often ceramic type objects cups/tureens/jewelry so the physical aspect was naturally a major concern.
However one specific problem which they discovered was that using the 510 Z-corp machine, there is a difference in color saturation depending on whether a printed surface is horizontal or vertical. So printing a solid red cube would look like this.
But they also showed some rounded objects (stupidly i forgot to take photos) which had an all over strong color density – so maybe it’s just the exactly vertical which works out poorly. Some of the sculptures had had photos projected onto them (in software) and printed out showing that the 510 can handle color gradations pretty well – there was a sort of smudginess to the images which was acceptible given the possibilities of the process – but not perfect by any means. With my sculpture – and it looks as if the likely steps may be
1) generate watertight computer models (groan)
2) produce miniature 3d sculptures using the Prodigy (about 5cm high)
3) choose a model – and generate a color version using 510 (about 15 cm high)
4) upgrade to life size using CNC milling
5) work out whether and how to color this
so the 510 would be a maquette. as the subject matter is a clothed? human the smudgy coloring will probably be ok. see next post for ‘artistic’ reasons why.
They had also ‘cured’ the 510 sculptures in various different ways – and the superglue one looked fine to me. The colors were good the surface ever so slightly glossy and smoother to the touch. Several people commented that perhaps the brittleness/powderiness of the models could be used as a positive with art works referencing their own fragility.
A second theme which emerged from several artists was the ability of Rapid prototyping to generate ‘impossible’ mathematical figures – and in particular the Klein Bottle came up several times. This bottle with a continuous interior/exterior excited me when i was a kid – so it was funny to see it pop up. Also there were balls within balls and a few polgonal monsters – i guess some of these shapes come as default test shapes in the printing process (i’ve seen them before at St. martins).
The point is that computer modelling allows you to make shapes not realisable through traditional sculpting methods (or at least if not impossible then bloody hard work). Bruce gernand had some interesting models which involved the models cross section mimicking the models overall shape. For example take a donut shape. From above the donut is a ring, and this is also the cross-section. he’d developed this further creating his own shapes – i forgot to ask him what he called it using a teardrop cross section. this was generated using CAD-CAM software and then made using various 3d techniques – laser cutting i think. also he showed some later work in which architecture models had been isometrically stretched – so referring to they way in which they were generated which were clever uses of rapid prototypings ability to produce impossible objects.
The third thread which was least represented on the day – an most important to me – was the possibiity of using computer modelling (including 3d scanning) to generate representational objects which again would be painstaking to do by traditional techniques. Andrew Folan showed a charming piece – he was comissioned to make a mural for a hospital with access being given to the full body medical scanners – the catch being that the imagery couldn’t be offensive to patients or staff – so nothing living involved. hmmm? he decided to scan a clarinet! he did this with and without the metal keys (as these were dangerous to put into the machine) and out came a series of ghostly scans of the instrument. then he put these into 3d modelling packages and with some difficulty reproduced a 3d model of the naked clarinet. i wasn’t clear what ended up in the hospital but the fly bys on the 3d models were ethereal and the fragile almost shabby 3d model itself was loaded with meaning – as a shadow of its former self.
at various points in the day techniacal difficulties were expressed regarding how to generate prototypable models from sets of scanned data points – or in my case pre-existing 3d meshes. i think the general consensus was that these kind of technical problems were the price you had to pay for using this technology at this point in time – still early days in terms of non-commercial applications. several artists were using Rhino as part of their process – and some people seemed to have no problems generating watertight soup bowls. the difficulty was pointed out that there was a world of difference between tyring to watertight something representational as opposed to something geometric. Other people had used technicians/bureaus to tidy up their models for them – at a cost.
Barbara Rauch (who runs SCIRIA), Brendan Reid (a UWE researcher) and I all have similar problems watertighting these scans/meshes and no doubt will try and help each other sort this out.
On the software it was suggested i should be using Magics STL and i’m looking into how to get hold of this on a trial basis. For the scanners they should be using GeoMagics (confusingly a different product by a different company) which has the ability to throw a shrink-wrap surface over a point cloud – i’m also looking into trying to get this software for trial.
All in all it was a great day – and inspiring. I hope to work more closely with CFPR in the future, especially the ones trying to fill all the holes.