Opens profile photo
Follow
Click to Follow andymac3d
Andy McNamara
@andymac3d
Head of CG @ Cinesite - Into cool stuff re: science, art & the creative use of tech - CGI/VFX/VR/AR/AI/Experiential. ex. BBC, IBM, Rushes, Publicis etc. Welsh.
London, EnglandJoined December 2015

Andy McNamara’s Tweets

Absolutely shocking 😳
Quote Tweet
Microsoft has laid off entire teams behind Virtual, Mixed Reality, and HoloLens #vr #ar #mr windowscentral.com/microsoft/micr
2
2
Second lawsuit of the week filed for alleged copyright infringement by an AI company for using ‘scraped’ training data from the web. This time by Getty - who have more clout and resources than most. Will be interesting to see how this pans out..
2
1
Show this thread
Replying to
Here's a very long-winded reply. Happy to go more in depth on tech as needed:
Quote Tweet
This thread is a response to @jonathanhook’s questions about volumetric capture, as always lots of caveats, and apparently, I have to post a long thread in two parts? 1st note: I work for Microsoft, in this field. This may/should/will affect how you view my responses
Show this thread
Six different squares that show a different product, each of which uses volumetric content, which is three dimensional. the first square shows David Attenborough, the second an image from a VR documentary, the third an AR campaign for Coke, the fourth an AR campaign for Charlotte Tillbury, the fifth a piece by Accenture using Altspace, and the sixth an AR ad campaign for Wembley.
2
Great summary of the developments in Stable Diffusion this year - much more fun than Midjourney for tinkering under the bonnet especially when using UIs like Automatic1111 and the new InvokeAI to run SD locally. Interested to see where this goes next year.
Quote Tweet
As the year is coming to an end, let's take the time to highlight many of the amazing community contributions and projects over the year... Stable Diffusion has only been around for 6 months but a full thread is still warranted given the extremely fast pace of progress! ↓
Show this thread
2
More literary shenanigans with #ChatGPT - this time #ai driven poetry with a disparate set of themes : monkeys, baked beans and engine oil 😀. A tough challenge, but surprisingly coherent.
Image
1
Interesting expo at #180thestrand by the talented folk - lots of generative art and some nice interactive exhibits. Pretty much a Houdini and Unity-fest - but great to see this projected in an exhibition space. Exhibit 14 with the bouncing critters was 👌 😀
Image
Embedded video
0:08
43 views
Embedded video
0:18
15 views
Image
1
2
One of my first ever #CGI projects when I was a trainee in R&D at the #IBM Scientific Centre in Winchester (circa 1987). Raytraced logo for their 20th anniversary created on an IBM 3090 mainframe at 1k using Winsom, Esme and Rexx proprietary tools - render time was 2 days!
Image
5
Stinks of desperation for libraries such as shutterstock where AI has the potential to affect their bottom line and need to control this space. How can they tell if a well crafted, heavily Photoshopped image is AI anyway? The genies out the bottle… 🧞‍♂️
1
Great way of blocking out compositions, shapes and forms in 3d for SD with interactive feedback (for those that don’t like sketching!). Maybe this could be integrated with a more artist friendly/sophisticated voxel editor like 🤔
Quote Tweet
~ Presenting: DiffusionCraft AI ~ Stable Diffusion running live with Minecraft. Here is my latest experiment with the little help of some python scripts. #diffusioncraft #minecraft #stablediffusion #python #img2img #aigaming
Show this thread
Embedded video
0:44
38.4K views
1
Bored of the usual #AI concept art so tried something different with #stablediffusion to create a retro comic strip. A Marvel-esque Clint as his superhero alter ego takes on a refrigerator! Frames all AI guided compositions and retouched in PS. Just a test, but lots of potential!
Image
1