With literally just hours to go before the 2021 Hackaday Remoticon kicks off, editors Tom Nardi and Elliot Williams still managed to find time to talk about some of the must-see stories from the last week. There’s fairly heavyweight topics on the docket this time around, from alternate methods of multiplying large numbers to the incredible engineering that goes into producing high purity silicon. But we’ll also talk about the movie making magic of Stan Winston and some Pokemon-themed environmental sensors, so it should all balance out nicely. So long as the Russian’s haven’t kicked off the Kessler effect by the time you tune in, we should be good.
Take a look at the links below if you want to follow along, and as always, tell us what you think about this episode in the comments!
Direct download (52 MB)
Episode 145 Show Notes:
What’s that Sound?
Tell us your answer for this week’s “What’s that sound?”. Next week on the show we’ll randomly draw one name from the correct answers to win a rare Hackaday Podcast T-shirt.
News This Week:
Interesting Hacks of the Week:
- OpenGL Machine Learning Runs On Low-End Hardware
- Hacking Multiplication With Karatsuba’s Algorithm
- Exploring The Healing Power Of Cold Plasma
- Designing Electronics That Work
- Those Bullet Effects In Terminator 2 Weren’t CGI
- For All Their Expense, Electric Cars Are Still The Cheapest
Quick Hacks:
- Tom’s Picks:
- Elliot’s Picks:
On the EV discussion and not seeing expired batteries: it’s hard to make the observation because the vast majority of EVs on the market are less than 10 years old.
The Nissan Leaf for example made it to market in late 2010 and sold only 19 units that year. Of all the Leafs sold to this month in the US (163,190) only 7,218 are aged ten years old older. That’s just 4.4% of all the cars sold. It’s too early to say the predictions didn’t pan out when you most likely don’t know anyone with a really old EV.
And, if you read some of the comments from the previous articles, at least one person with an early model Leaf did report having their battery go down significantly around year 10. The argument was never that the battery would suddenly die overnight; it’s not the long gradual fade predicted by fanboys either.
We know how lithium batteries behave at their end of life and it is an accelerating loss of capacity over some months and a few dozen charging cycles until it just won’t go any longer. That happens because you’re cycling it faster and faster to get the same work out of it, so it wears out faster and faster. When you drive the same number of miles to work every morning, the amount of damage that does to the battery gets proportionally greater as time goes by, which makes each charging cycle lose more capacity than the last. You have a system with a tipping point.
Though of course there is always the dodge to say “Well modern batteries aren’t like the ones they made ten years ago, the new ones last forever”.
Well, keep buying the cars and we’ll see eleven years down the road how it was.
And secondly, speaking of people with cars older than 10 years as being special cases to be ignored is ignoring the fact that 70-80% of the market for cars is second hand. Very few actually buy new cars.
Here’s some interesting data based analysis.
https://towardsdatascience.com/figuring-out-a-fair-price-of-a-used-car-in-a-data-science-way-11450b3b252b
The mean and median prices for used vehicles are $7,275 and $4,900. That’s the sort of money the majority of people would pay for a car, typically between1996-2009 vintage. There are no EVs in this bracket, new or used, by price or by age. They are simply irrelevant for the vast majority of people.