dllu

If I remember correctly, the Valeo Scala that's in the Audi cars uses a discrete 16 element 1D array (940 nm diodes + APDs) plus a rotating mirror to scan.

The article talks about eye safety a bit in section 4.

> a stuck mirror

This is one of the advantages of using an array of low power lasers rather than steering a single high power laser. The array physically doesn't have a failure mode where the power gets concentrated in a single direction. Anyway, theoretically, you would hope that class 1 eye-safe lidars should be eye safe even at point blank range, meaning that even if the beam gets stuck pointing into your eye, it would still be more or less safe.

> 20 cars at an intersection = 20 overlapping scanners, meaning even if each meets single-device Class 1, linear addition could offer your retina a 20x dose enough to push into Class 3B territory.

In the article, I point out a small nuance: If you have many lidars around, the beams from each 905 nm lidar will be focused to a different spot on your retina, and you are no worse off than if there was a single lidar. But if there are many 1550 nm lidars around, their beams will have a cumulative effect at heating up your cornea, potentially exceeding the safety threshold.

Also, if a lidar is eye-safe at point blank range, when you have multiple cars tens of meters away, laser beam divergence already starts to reduce the intensity, not to mention that when the lidars are scanning properly, the probability of all of them pointing in the same spot is almost impossible.

By the way, the Waymo Laser Bear Honeycomb is the bumper lidar (940 nm iirc) and not the big 1550 nm unit that was on the Chrysler Pacificas. The newer Jaguar I-Pace cars don't have the 1550 nm lidar at all but have a much bigger and higher performance spinning lidar.

A flash lidar is simply a 2D array of detectors plus a light source that's not imaged. It's mentioned super briefly at the start of section 3 but you're right, I should have gone into more detail given how common and important they are.

For pulsed emitters, indeed adding random jitter in the timing would avoid the problem of multiple lidars being synced up and firing at the same time. For some SPAD sensors, it's common to emit a train of multiple pulses to make a single measurement. Adding random jitter between them is a known and useful trick to mitigate interference. But in fact it isn't super accurate to say that interference is a problem for continuous-wave emitters either. Coherent FMCW lidar are typically quite robust against interference by, say, using randomized chirp patterns.

Mine is a DAS connected to my main workstation [1] that indeed has 96 GB of RAM.

[1] https://daniel.lawrence.lu/blog/2024-04-26-small-form-factor...

Very sad that HDDs, SSDs, and RAM are all increasing in price now, but I just made a 4 x 24 TB ZFS pool with Seagate Barracudas on sale at $10 / TB [1]. This seems like a pretty decent price even though the Barracudas are rated for 2400 hours per year [2] but this is the same spec that the refurbished Exos drives are rated for.

By the way, interesting to see that OP has no qualms about buying cheap Chinese motherboards, but splurged for an expensive Noctua fan when the cheaper Thermalright TL-B12 perform just as well for a lot cheaper (although the Thermalright could be slightly louder and perhaps be a slightly more annoying spectrum).

Also, it is mildly sad that there aren't many cheap low power (< 500 W) power supplies for SFX form factor. The SilverStone Technology SX500-G 500W SFX that was mentioned retails for the same price as 750 W and 850 W SFX PSUs on Amazon! I heard good things about getting Delta flex 400 W PSUs from Chinese websites --- some companies (e.g. YTC) mod them to be fully modular, and they are supposedly quite efficient (80 Plus Gold/Platinum) and quiet, but I haven't tested them out yet. On Taobao, those are like $30.

[1] https://www.newegg.com/seagate-barracuda-st24000dm001-24tb-f...

[2] https://www.seagate.com/content/dam/seagate/en/content-fragm...

For those who are interested, the Will Evans course on graph drawing [1] covers a lot of cool graph drawing algorithms. Graph drawing is very interesting and many applications get it wrong. I once contributed [2] some small bugfixes to the Dot lexer for the Open Graph Drawing Framework, which has fast implementations of some amazing graph drawing algorithms, and my experience is that the OGDF draws graphs vastly better than the various algorithms in GraphViz (fewer crossings, faster, etc).

[1] https://www.cs.ubc.ca/~will/536E/

[2] https://github.com/ogdf/ogdf/pulls?q=is%3Apr%20author%3Adllu...

One day I'll muster up the motivation to bring my setup to Roaring Camp to scan those Shay geared locomotives but those moving parts will indeed appear weird and distorted.

Yeah they have a whopping 300mm f/2.0 lens for photo finish! I have been using various primes including a Samyang 135mm f/2, Voigtländer Apo Lanthar 125mm f/2.5, Voigtländer Nokton 58mm f/1.4, Voigtländer Ultron 35mm f/1.7, Myutron 50mm f/2.6, etc. The problem with a really large aperture is that it's hard to nail focus.

I've taken a couple of pics from a moving train...

Nankai 6000 series, Osaka:

https://i.dllu.net/nankai_19b8df3e827215a2.jpg

Scenery in France:

https://i.dllu.net/preview_l_b01915cc69f35644.png

Marseille, France:

https://i.dllu.net/preview_raw_7292be4e58de5cd0.png

California:

https://i.dllu.net/preview_raw_d5ec50534991d1a4.png

https://i.dllu.net/preview_raw_e06b551444359536.png

Sorry for the purple trees. The camera is sensitive to near infrared, in which trees are highly reflective, and I haven't taken any trains since buying an IR cut filter. Some of these also have dropped frames and other artifacts.

Yeah I actually have it disabled by default since it makes the horizontal stripes more obvious and it's also extremely slow. Also, I found that my vertical stripe correction doesn't work in all cases and sometimes introduces more stripes. Lots more work to do.

As for RCD demosaicing, that's my next step. The color fringing is due to the naive linear interpolation for the red and blue channels. But, with the RCD strategy, if we consider that the green channel has full coverage of the image, we could use it as a guide to make interpolation better.

Thanks, I added a section called "Principle of operation" to explain how it works.

Thanks for reminding me of this cool building --- I just updated the Wikipedia article with an infobox and a photo that I took in 2019.

It was at CVPR 2019, a computer vision conference. I may be biased since I used to work at Ouster, but cost notwithstanding, I would definitely pick the OS1 again for its unparalleled number of points per second combined with low weight and decent accuracy.

I once put an Ouster OS1 on a hat and walked around with it. Pic of me here: [1]

[1] https://x.com/ddetone/status/1141785696224477184?s=46

Anecdotally, using Darktable, I could never get as good of a demosaicing result as using the straight-out-of-camera JPEGs from my Fujifilm GFX 100S. In challenging scenarios such as fine diagonal lines, Darktable's algorithms such as LMMSE would add a lot of false colour to the image.

However, modern deep learning-based joint demosaicing and denoising algorithms handily outperform Darktable's classical algorithms.

Fujifilm lossy compressed raw still isn't supported after many years [1].

[1] https://github.com/darktable-org/rawspeed/issues/366

And in my experience there has been lots of bugs with Fujifilm raws in darktable:

[2] https://github.com/darktable-org/rawspeed/issues/354

[3] https://github.com/darktable-org/darktable/issues/18073

However, Fujifilm lossless compressed raw actually does a decent job keeping the file sizes down (about 50% to 60% the file size of uncompressed) while maintaining decent write speed during burst shooting.

Same in German, die Fledermaus. In fact, in English, the word "flittermouse" also means bat [1], although that word is rarely used now.

[1] https://en.wiktionary.org/wiki/flittermouse

Yes basically. Bird lungs are relatively rigid, open at both ends like a tube, and have a one-way flow of air, so they are less prone to pressure-related injuries. Bat lungs are mammalian lungs that expand and contract as they breathe just like us, so they are particularly vulnerable to barotrauma near wind turbines.

After writing a bunch of MATLAB code to find the bats, I handed it off and haven't heard back about whether they actually built the wind turbines or not.

Here's the report [1], written when I was a second year undergrad in 2010.

It's very basic. The species identification is based on matching contours of the spectrogram against some template contour. The multilateration was, embarrassingly, done by brute force by generating a dense 3D grid. At the time, I didn't have any knowledge of Kalman filters or anything that could have been helpful for actually tracking the bats.

[1] https://daniel.lawrence.lu/public/bat-report.pdf

I once did a project to do multilateration of bats (the flying mammal) using an array of 4 microphones arranged in a big Y shape on the ground. Using the time difference of arrival at the four microphones, we could find the positions of each bat that flew over the array, as well as identify the species. It was used for an environmental study to determine the impact of installing wind turbines. Fun times.

> There's another, more subtle critique of this system: it lacks path independence. This means that if you start and end a drag with your mouse at a particular location, the rotation will depend on the path that your mouse took.

Actually, when I accidentally tumble models with that kind of UI, I just drag it in a circle until it's right side up.

3D printing technology is amazing now. I used to struggle with my ABS prints warping 12 years ago with a PP3DP --- I couldn't even print a giant 3x3x3 rubik's cube that worked. Now there are lots of 3D printers that are essentially just zero configuration and everything works out of the box. I even printed a lens mount for my camera and it came out quite well aligned. So it is very nice to see some regular consumer 3D printers being good enough for a functional 34 x 34 x 34 cube.

Fortunately, modern TVs can disable the sharpening and contrast/saturation enhancements. Since I do a lot of photography and image processing, I am also extremely sensitive to oversharpened halos, so I was a bit worried about that at first --- but fortunately, that is 100% nonexistent once I applied the appropriate settings on my Samsung QN800A. See pics: [1] [2]

I also detest the miniled HDR that they have going on, which can cause bright things to glow, so I disabled that. Unfortunately my QN800A still have a bit of "full screen HDR", namely, that the whole screen may uniformly dim if it detects that the scene is dark. This means that sometimes when you have a black screen with a single cursor on it, it gets dark, and it becomes hard to see the cursor. This doesn't affect normal usage though, when the screen is at a constant brightness.

[1] https://i.dllu.net/2024-10-30-09-36-11_DSCF2616_11f836ca2100...

[2] https://i.dllu.net/2024-10-30-09-37-18_DSCF2617_eeab771f780f...

Anecdotally my house draws 0.4 kW when idle and 0.6-0.7 kW when both my 8K screen and my computer are on. Since my computer draws 0.1-0.2 kW, I surmise that the QN800A doesn't draw 300-400 W total --- maybe 100-200 W.

I run my screen on a brightness setting of 21 (out of 50) which is still quite legible during the day next to a window.

Also, I have solar panels for my house (which is why I'm able to see the total power usage of my house).

Yeah it does emit a bit of heat. I think around one or two hundred watts? I haven't measured it directly. I have a mini split air conditioner in my home office.

I wish Dell would come out with a refreshed version of the UP3218K that's cheaper and fixes its various little glitches.

Oh I just emailed them, and their rep Jeremy Postma is very nice and responsive.

If you want a cheaper option you could buy an IKEA Karlby Countertop that's 74" x 42" [1] and mount it on the legs yourself.

[1] https://www.ikea.com/us/en/p/karlby-countertop-for-kitchen-i...

I was previously working at a lidar company and now I am working at a robotics company providing calibration and localization software to customers using a combination of lidars, cameras, and other sensors.

I have the Uplift 4 leg standing desk [1].

I got the black laminate desktop in a custom 75" x 42" dimension so the whole thing cost me almost $2000.

[1] https://www.upliftdesk.com/uplift-4-leg-standing-desk-v2-v2-...

I have fairly bad eyesight with both myopia and astigmatism (-5 sph, -2 cyl) and I wear glasses. I got glasses with 1.71 index lenses, which I greatly prefer over the more common 1.74 index lenses due to the higher Abbe number, resulting in less chromatic aberration.

Anyway, I use browsers at 150% scaling usually, although the text is finer on my terminals. I don't use any scaling for UI elements and terminals. Using the i3 tiling window manager, I put more commonly used terminals on the bottom half of the screen since I find that the top half does require more neck craning.

I'm 184 cm tall.