santaclaus
santaclaus on Hacker News
> its a cool Visual FX studio not a technology company..
Weta develops metric tones of tech internally, including employing some of the top graphics researchers at various points in their careers. There is tons of tech talent in VFX and animation shops. Hell, I work with libraries open sourced by Pixar as part of my day job at a FAANG.
I used to follow your Rust work and had no idea you are yinzer! Mineo’s or Aiello’s? :)
It’s certainly not East Coast, but it’s very North East - the rust belt spans well beyond the Midwest. Even culturally Pittsburgh feels like it has more in common with, say, Buffalo or other North Eastern rust belt cities than Wisconsin or Minnesota.
> There are plenty FAANG engineers minted from the top Midwestern schools - think CMU and other state schools - even if they rank behind the top schools on the coasts.
CMU is a private school, ranked number 1 in CS, and is not in a Midwestern state! That said, there are lots of top CS programs in the Midwest.
University of Illinois Urbana Champagne, University of Michigan, university of Wisconsin Madison, and Purdue are all top 20 in CS.
Very cool! Just the other day I was trying to set Blender’s camera based on a standard 3x4 computer vision KRT matrix, and it is surprisingly a pain in the ass —- I wish more of these graphics CAD packages (Blender, Houdini, Maya) made it easier to deal with vision data.
Posted by
santaclaus on Thursday August 20, 2020 @08:01PM
from the salon.com dept.
Very cool! Glad to see Nucleus still getting some love, Jos would be happy — I was talking to a Pixar effects fellow at SIGGRAPH and was under the impression that Houdini’s constraint solver thing had spread like wildfire in that department.
Maya is still used in Pixar’s characters department for modeling, I believe, and maybe a few shading-related (UV layout?) tasks here or there.
Autodesk is doubling down on CAD and engineering (their leadership has even said so publicly) — VFX and animation is just a much smaller market. I personally think this is very short sighted.
Autodesk laid off the entire R&D staff of Maya (outside of Bifrost, which was an acquisition a few years ago and is now a one person show, mainly) and transferred ownership to their maintenance engineering division, if that isn’t life support I don’t know what is.
> Autodesk tools (especially Maya) are way too expensive in what they deliver. I'm glad many big studios realized this in that they either build their own tools (Pixar/Dreamworks) or they adapt Open Source (Blender) and help improving it.
Cost isn’t the issue, it’s mainly the fact that Autodesk effectively put Maya on life support. Pixar is paying hundreds of software engineers in the Bay to develop their proprietary animation tools, which is most definitely not cheaper than a few hundred Maya seats per show.
> I was interested to find out which steel mill, since they are almost non-existent in Pittsburgh these days
As evidenced by the smell of sulphur yesterday morning the coke plants down in Clairton are still alive and well!
Posted by
santaclaus on Sunday August 04, 2019 @03:36PM
from the nytimes.com dept.
I don’t think Pixar’s Maya-like software ‘man v’ runs on macOS, at least it didn’t a couple of years ago (my ex was a TD at Pixar and used to complain about the old version of Linux they were on). I’m pretty sure the demos they showed were of their file format viewer which is a separate thing.
Film 3D is mostly Linux at this point. All the major VFX shops (like Weta) and animation studios (Disney, Pixar, etc) are on Linux for artist workstations. Apple stuff gets lots of use, but mainly iPads for story boarding and such.
> Curious if this is my bubble or consistent.
Depends where you live! I’m in Pittsburgh and there is still a culture here of bailing to the well heeled suburbs (cough cough Fox Chapel) when having kids. You couldn’t pay me to leave the city core, but the Midwest still clings to that suburban status symbol life more than the rest of the nation.
> I don't want to be that kind of stuck-in-the-mud, so I'm trying to just relax and accept linguistic progress in whatever form it takes.
Does this mean you will be dropping the 'u' from colour?
I love these types of breakdowns! It would be really cool to see breakdowns in other portions of a game engine, too: physics, character animation, managing the massive assets, lots of potential goodies.
It would be really cool if someone web-assemblied the best of class game system emulators and exposed them all through a uniform web interface. All vintage game consoles accessible through a single site!
I never said San Francisco was safer than NYC, but even looking at those statistics they are mostly comparable in terms of violent crimes (notice that I did say your car is much more likely to get burgled in San Francisco), and the rate of aggravated assault is higher in that study in NYC. My point is that that both San Francisco and NYC are by and large safe cities by American city standards.
Because ads and/or iPhones make waaaaaay more money
Sure, and the same anecdotes hold in Manhattan. I've had friends robbed at knifepoint in the Upper West Side, I've seen phone jackings at 1pm in front of Gristedes in the Upper East Side. Hell, my brother's roommate in Harlem was clocked with a baseball bat in the head while leaving his apartment and subsequently robbed blind.
By and large NYC and San Francisco are both very safe cites in the grand scheme of things. (Unless you are a car window in San Francisco. Then you are screwed).
Most of the construction in DTLA was previously parking lots... I find it hard to believe that mostly vacant parking lots adjacent to a mostly vacant sports stadium is somehow a superior proposition to a sustainable urban fabric than much needed housing.
> without all the graphics cruft
What cruft are you referring to here? I doubt that removing the rasterization logic from the GPU is going to make them signifcantly faster at matrix-vector multiplies or whatever.
This seems similar, but somewhat different than UV auto-layout in terms of the restrictions on the map. For this mapping to produce a buildable object, each flattened face must have the same area as the 3D face, and the mapping from 2D to 3D must be globally bijective. In UV-mapping, given a set of artist-specified cuts, I do not think it is always possible to obtain an area-preserving and globally bijective map. Most industrial-strength UV tools also allow artists to rescale and deform certain faces (to make it easier to paint under fine geometry, for example), which would be wouldn't make sense given the end goal of this algorithm.
I've met tenured computer science professors at UC Santa Barbara who couldn't afford to live in Santa Barbara. That place is nuts. Similar situation with Stanford and Palo Alto.
I remember reading a while back about a link between competitive bridge (the stereotypically old person card game) playing and a lack of cognitive decline in aging adults. I'm curious if competitive video games have a similar anti-brain-aging effect.