andygreenwell

Living on Cape Cod these days, I read “Burt Dow Deep-water Man” to my son fairly regularly when it’s time for bed. I think this guy just picked up a new nickname for himself.

Interesting development in this space this week for WiTricity (the company) is that China is standardizing nationally on magnetic resonance technology for wireless vehicle charging based on WiTricity’s designs and standards work. I’m sure this was an exciting week for my former colleagues there, and congratulations are certainly in order for them.

https://www.greencarcongress.com/2020/05/20200508-witricity....

There are academic connections between the group at Stanford publishing this work and the group at MIT that founded WiTricity.

You have to make due with 32-bit floating point accuracy for any vertices defined in Blender for the geometry to be passed along to any external libraries. In a prior gig where Blender was the primary development environment, I needed to do all kinds of gymnastics (converting floating point numbers provided by the user in the gui to strings then to Python floats or sometimes 32-bit or 64-bit numpy objects, then to the external package and back) in order to maintain double precision accuracy. Anything created or stored as a Blender FloatProperty is a 32-bit value on the C-side. But if you operate on that FloatProperty object within the embedded Python process you are operating on the 64-bit representation of the 32-bit pattern, since Python floats are all 64-bit values.

I heard more than once during that gig that “Blender is not CAD” even though we were doing everything possible to bring it closer to being so.

But with vertex values, those are created directly in Blender’s C layer, so no possibility to make those values have 64-bit accuracy if that is important to your application or external library.

I am in agreement that Blender would make a fantastic environment to pull together many open source scientific and engineering libraries, but the vertex accuracy is the one major drawback there.

Worked for Crane Currency a year ago...left after only a year there just after the acquisition by Crane Co.

The company has facilities in Tumba, Sweden as well as their new production facility in Malta.

There are some very talented engineers, scientists, designers and artists on staff there.

Hopefully Crane Co will be good stewards of what they have purchased, because Crane Currency provides an essential service for many countries around the world, but I’ll just say that I was not inspired nor impressed by the new acquirers and I hope the new ownership doesn’t drive too many more people away.

Maybe, but having previously worked for a company that makes physical money, it's a healthy business they are in. There is more cash being printed now than ever before, the second derivative on the the amount of cash over time (d2Cash/dt2) is definitely positive.

My personal opinion is that there will always be a need for physical analog money. The power goes out in Puerto Rico (hurricanes) and Manhattan (ConEdison) sometimes (just recapping the stories covered on NPR during my drive to work this morning).

Expect to see an implementation included in subsequent releases of Code V and Zemax.

In my time in MA, I worked for three separate MIT startups, one top software company and one IP law firm. The law firm was the only one that didn’t require a noncompete. One of those companies was backed by one of the loudest voices in the Boston/Cambridge VC community campaigning against such agreements. The company had me sign an updated non-compete upon closing a funding round at the same time their VC was giving public speeches denouncing noncompete agreements in MA. Until the garden leave requirement covers 100% salary, the updated provisions to MA noncompete law are toothless.

Crane Currency uses Blender in the design of the micro-optical animations of their Motion and Surface security threads that are integrated into banknotes around the world. Check out a US $100 bill for an example of the results. In my previous role there, leveraging Blender's flexibility for integrating functionality via the Python API into various parts of the UI was great.

I wish there was an open source CAD/CAE application that would copy a few of Blender's workflow ideas (integrated REPL, 3d view, node editor, text editor, high level language API).

All Blender is missing is 64-bit vertex storage and 2D data plotting and it would also be the perfect dashboard bridge between the physical engineering/design and data science worlds in addition to its primary audience of 3D content creation, but it's hackable enough to work around those few limitations.

And then you have the mayors of North Fulton (for example Mayor Mike Bodker of Johns Creek) that are fundamentally opposed to expanding various forms of rapid transit into their supposedly low density housing suburban communities:

https://www.johnscreekga.gov/NewsAndEvents/News-Archive/2018...

Having moved to Johns Creek from Boston last year, I think the Red Line should go all the way up 400 to Cumming and they should add more express lanes along 400.

But there are two messages I hear consistently around town when I ask various people about this issue:

1. People like their low density housing, and you are never going to be able to put in enough rail access (like Boston's T or NYC's Subway) close enough to where people live or where they need to be on the other end to eliminate cars off the road through most of the area.

2. They don't want particular groups from other parts of Atlanta having easy access to where they live...essentially to keep possible criminal elements away.

Point 1 is a legitimate concern, but commuter rail systems do exist in other areas of the country, and parking garages can be placed next to rail stations. Would definitely ease the nightmare that is rush hour on 400.

Point 2 is just a terrible legacy of racism that still exists in the South (and I'm a Southern boy born and raised until living in Boston for 10 years). There is a terrible acronym for MARTA that I have heard multiple times since moving here that just needs to disappear.

> They would be so much better if they had even a few more millimeters of leeway in their positioning.

Well if resonant based charging can ultimately win the standards battle versus Qi (standard magnetic induction), then you could have some spatial freedom to move around and still get efficient charge. Qi has had first mover advantage in consumer electronics, while resonant charging seems to be taking hold in the wireless vehicle charging sector.

Disclaimer: I previously worked at WiTricity on their internal computational electromagnetic modeling tools.

They are trade secret, so can't describe them here, but currently using algorithms created by my skip level manager [1] to make actual physical money.

[1]: https://www.cranecurrency.com/rnd-banknote-security-thread-f...

An additional note for posterity here...one of the main reasons Microsoft was interested in acquiring Interactive Supercomputing was that ISC developed it's own M-language (Matlab) compiler M# built on Mono/.NET...basically a faster Matlab replacement than Octave, using features like multiple dispatch. One of the primary developers of the M# compiler at ISC was Jeff Bezanson, co-creator of Julia.

Exie used a bit of code from M#. When Exie was cancelled, the libraries were given a C#/F# facelift as Cloud Numerics, which was eventually cancelled, and then given a GUI facelift as part of the initial version of AzureML.

Disclaimer: I previously worked at Interactive Supercomputing, was part of the team acquired into Technical Computing/HPC at Microsoft, and also worked at Julia Computing.

Glad to hear that the Excel team is finally listening to your requests on this one. It's been a long time coming. Too bad they weren't interested during the TC days. If it does happen, it would be great if you could get the Cloud Numerics effort rebooted as well. There might still be a few people from our old team at NERD capable of resurrecting it.