As you’ve pointed out fork() isn’t ideal for a number of reasons and in general it’s preferred to use torch tensors directly instead of numpy arrays so that you are not forced into using fork()

There’s also this write up which I found to be quite useful for details: https://ppwwyyxx.com/blog/2022/Demystify-RAM-Usage-in-Multip...

This dataset is clearly targeted for research on AR/XR/VR applications.

I’ll “yes and” here…beyond AR/VR a more powerful use case is multi-modal learning (with RL) which is what Meta is probably the leader in IMO.

Example paper here: “ Towards Continual Egocentric Activity Recognition: A Multi-modal Egocentric Activity Dataset for Continual Learning”

https://arxiv.org/abs/2301.10931

This IMO is the pathway to AGI, as it combines all sense-plan-do data into a time coordinated stream and mimics how humans transfer learning to children via demonstration recording and behavior authoring.

If we can create robotics with locomotion and dexterous manipulation, egocentric exploration, and a behavior authoring loop that uses human behavior demonstration and trajectory reinforcement - well, we’ll have the AI we’ve been all talking about.

Probably the most exciting area of research that most people don’t know or care about.

That’s why head mounted all day ego centric AR is so important - it gives eyes ears and sense perception to our learning systems with human directed egocentric behaviors, guiding the whole thing. Just like pushing your kid down the street in the stroller.

That is some B.F. Skinner level future we're aiming for--only this time around, humans become the fully surveilled 'teaching machine'.

If we want to make machines with equivalent or better capacity as humans we have to transfer the process for scientific discovery, including the sum of our cognitive capacity and knowledge to them.

If you quantize human adult-infant interactions, then it boils down to Human adults introducing learning trajectories, labeling input data and biasing weights with reinforcing behaviors for new reinforcement agents. If we can re-build the infrastructure to do precisely that, where the agent is in the place of the infant and society is in the place of the "Human Adult" then we will have re-built at scale the process for human development.

The best way we know how to do this today is implementing transfer learning approaches from the basic human developmental research. I started down this road back in 2010 trying to follow the work of Frank Guerin out of the University of Aberdeen [1] [2].

[1]https://www.surrey.ac.uk/people/frank-guerin

[2] https://scholar.google.co.uk/citations?view_op=view_citation...

Even when this barrier can be overcome (i.e. people become accustomed to wearing these devices), I worry about the opt-in nature of it. We've yet to see a disruptive technology adhering to this principle through-and-through, and if current learning efforts are anything to go by, training data is not something companies want to willingly let go or lose out on.

Taken both, this path has the potential to be quite coercive if no strong guarantees or safeties can be upheld, especially if early exciting trials generate an interest-boom similar to the one we're seeing right now in the LM-space.

That still leaves my first question regarding observer effects and how people would respond to such a technology on an individual level. It would have the capacity to reshape behaviour towards preferential and/or optimal interactions, would it not? Seeing how we do not want reinforce models with 'erroneous' interactions?

You are persistently surveilled in London and Shanghai and New York City - yet people act just as unhinged in ways they did before cameras were installed.

I'm not sure what other data acquisition/technology arc is possible though, and open to ideas.

Unhinged people do, but ordinary people? I'd be willing to bet that normal people who are in areas where they are aware they're on camera don't behave as their normal selves. It's hard to see how it could be otherwise.

It’s my understanding (though I haven’t looked at the primary sources myself) that one of the facts that inspired Chomsky’s language theories and work for instance, was that when you quantify the information communicated by parents to language learning children, there’s actually not very much of it. Not nearly enough to support that what’s going on is anything like the kind of learning embodied by machine learning models.

If that’s true, and there is something of how to act intelligently / humanly already encoded in children (maybe genetically?) and not communicated by this sort of training, wouldn’t ignoring that and trying to get to it purely in this machine learning way be.. at least not at all informed by evidence / examples of it working in nature?

I certainly acknowledge my own bias with this however, with respect to what Chomsky discusses, I make the distinction that most of the “code/data/information” that you need in order for the language capacity to develop is actually embedded in our biological mechanical systems. That is to say, if you were to take a human infant and never expose it to another human with respect to generating sounds for language, the infant would still develop some sort of sound based communication system. We see this with feral children, mute children, deaf children. They still have a verbal function, even if it’s not connected to any semblance of coherency.

So in that sense it’s like you’re given all of the building blocks for language out of the gate biologically and then the people who are around you tell you how to assemble them into some thing that is functional. This is why different languages have different rules yet language acquisition is consistent across cultures.

This is why I am insistent on holistically understanding the computing infrastructure and systems because the sensors processors, etc. are the equivalent to our cells, genes muscles, bones, etc. Most people don’t think about computing systems and generally intelligent systems this way.

If you go back and look at the work of wiener and early Cybernetics it does discuss a lot of this, however, after Cybernetics was absorbed into artificial intelligence, which was an absorbed into computer science, it doesn’t really look holistically at systems of systems, unfortunately, in the general case.

And I would argue that all of machine learning currently is very much moving in to the direction that I am describing where is exposure to frequency of correlated data that gives you your effective understanding of the world, and being able to predict the future state. That’s what I mean when I say multi-modal is “sequential and consistent in time” with respect to causal action.

If used correctly (if is doing lots of heavy lifting here) this type of system, eye gaze, imu & microphones would provide much much better hearing aids than the current state of the art, at a much cheaper price (go look up the price of hearing aids, its _extortion_ )

Using gate analysis, it would be possible predict when someone is prone to falls, allowing much longer independence for older people.

Assuming that its possible to understand who you are talking to and what they said, you could mitigate and support dementia much more than we can now.

However.

You also have a vast network of headsets with highly accurate always on location, able to see what you are looking at, who you talk to, what you say, and in somecases what you feel about things.

Add in some basic object/facial recognition and you have an authoritarian's wet dream.

now is the time to regulate, but alas, that wont happen.

Applications of embodied AI very interesting. Additionally a lot of hard problems are increasingly being solved in simulation like this. See Wayve's GAIA world model

A behaviour that authors loops?

In order to do this you can use a source/stem separation model like spleeter (https://github.com/deezer/spleeter) and then run the basic pitch model (or any other midi transcription model). There's other you can try which may yield better results, for example: (https://github.com/Music-and-Culture-Technology-Lab/omnizart)

Either way the key words you want to be looking for are "midi transcription" and "stem separation", should help you find more models to try for both steps. Good luck! :)

EDIT: Oh it looks like there's even a stem separation leaderboard on papers with code, neat: https://paperswithcode.com/task/music-source-separation

It is not an open question and I hate it when writers frame it this way. Camera only (specifically, monocular camera only) systems literally cannot be safer than ones with sensor fusion right now. This may change in the future at some point, but it's not a question right now it is a fact.

Setting aside comparisons to humans for a second (will get back to this), monocular cameras can only provide relative depth. You can guess the absolute depth with your neural net but the estimates are pretty garbage. Unfortunately, robots can't work/plan with this input. The way any typical robotics stack works is that it relies on an absolute/measured understanding of the world in order to make its plans.

That isn't to say that one day with sufficiently powerful ML and better representations we would be totally unable to use mono (relative) depth. People argue that humans don't really use our stereoscopic depth past ~10m or so and that's a fair point. But we also don't plan the way robots do. We don't require accurate measurements of distance and size. When you're squeezing your car into a parking spot you don't measure your car and then measure the spot to know if it'll fit. You just know. You just do it. And it's a guesstimate (so sometimes humans make mistakes and we hit stuff). Robots don't work this way (for now), so their sensors cannot work this way either (for now).

From how humans drive, its pretty clear that there exists some latent space representation of immediate surroundings inside our brains that doesn't require a lot of data. If you had a driving sim wheel and 4 monitors for each direction + 3 smaller ones for rear view mirror, connected to a real world car with sufficiently high definition cameras, you could probably drive the car remotely as well as you could in real life, all because the images would map to the same latent space.

But the advantage that humans have is that we have an innate understanding of basic physics from experience in interacting with the world, which we can deduce from something simple as a 2d representation, and that is very much a big part of that latent space. You wouldn't be able to drive a car if you didn't have some "understanding" of things like velocity, acceleration, object collision, e.t.c

So my bet is that just like with LLMs, there will be research published at some point that given certain frames in a video, it will be able to extrapolate the physical interactions that will occur, including things like collision, relative distances, and so on. Once that is in place, self driving systems will get MASSIVELY better.

Self-driving is still a robotics problem, and robots are probablistic operators with many component dependencies. If you have 3 99% reliable systems strung together running 24 hours a day, that's 43 minutes a day that it will be unreliable ((1 - .99^3)*1440). Multi-modality allows your systems to provide redundancy for one another and reduce the accumulating correlated errors.

Eh, kind of...

https://youtu.be/HU6LfXNeQM4?t=1987

Check out this NOVA video on how limited your acute vision actually is. It is only by rapidly moving our eyes around that we have high quality vision. In the places you are not looking your brain is computing what it thinks is happening, not actually watching it.

Tangential: event cameras work more like our eyes but aren't ready for AVs yet.

Whether it's fine to be less safe in certain situations because it's safer overall is a different question.

The existence of peripheral vision disputes that pretty definitively, though.

You could drive fine at 30fps on a regular monitor (SDR). More fps would help with aggressive/sporty driving of course.

What? This is preposterous.

Have you tried playing a shooter video game at 30 FPS? It's atrocious, you get rekt. There is a reason all gamers are getting 120 FPS and up.

30 FPS means 33 ms of latency. Driving on a highway, car moves over a meter before the camera even detects an obstacle. The display has it's own input lag, so does the operating system. Your total latency is going to be over 100ms, so the car will have travelled several meters. If a motorcyclist in front of you falls, you will feel the car crashing into his body before the image even appears on the screen.

Also, if you truly are only a few meters behind a motorcyclist when driving at highway speeds, by definition you are being unsafe. The rule I learned in driving school was roughly 1 car length per 10mph of space, so you should be ~90 feet (~30 meters) away.

Finally, the average reaction time for people driving in real life is something like 3/4 of a second. 750ms to transition from accelerating to braking. A self-driving car being able to make decisions in the 100ms time frame is FAR superior.

You’d also lose depth perception entirely, which can’t be good for your driving.

That being said, this idea of a latent space representation of the world is the right tree to be barking up (imo). The problem with "scale it like an LLM" right now is that 3D scene understanding (currently) requires labels. And LLMs scale the way they do because they don't require labels. They structure the problem as next token prediction and can scale up unsupervised (their state space/vocabulary is also much smaller). And without going into too much detail, myself (and others I know in this field) are actively doing research to resolve these issues so perhaps we really will get there someday.

Until then however. Sensors are king, and anyone selling you "self-driving" without them is lying to you :)

https://media.mbusa.com/releases/mercedes-benz-worlds-first-...

However Waymo, Cruise and others do exist. If you haven't already, check out JJRicks videos on YouTube. I think you might be changing the number of years in your estimation ;)

I disagree. When in a car, we are using more than our eyes. We have sound as well, of course, something that we provide feedback even in the quietest cars. We also have the ability to feel vibration, gravity and acceleration. Sitting in a sim without at least some of these additional forms of feedback would be a different skill.

There was another even where they took another sim driver and put him in a real drift car, and he was able to drift very well.

Both vids are on youtube. Yes, real world driving has more variables, and yes, the racing drivers had force feedback wheels, but in general, if a person is able to control a car so well as to put the virtual wheel in the right square foot of the virtual track to take a corner optimally, its probably likely that most people could drive very well solely from visual feedback. Sound and IMUs can provide additional correctional information, but the key point still remains, is that whatever software runs has to deduce physics from visual images.

Self driving cars obviously have microphones and accelerometers too.

Even if parity with human vision can be attained, humans kill 42,000 other American humans each year on the roads. If human driven cars were invented today, and pitched as killing only 42,000 people per year, the inventor would get thrown into a special prison for supervillains.

Taking things to the extreme, perhaps it’s actually a networking problem.

Cars should have the ability to send signals and hear signals from the cars around it.

Imagine if Car A could improve its own understanding of the environment using inputs/sensor data from nearby Car B.

The most basic example of how this could be helpful is if the car ahead of you turns a sharp corner and crashes into a truck stopped in the road. Without car-to-car networking, you won't brake until the crash is in your line of sight.

Have you ever seen those youtube videos of massive car pile ups on highways caused by a crash, and then a cascade of additional crashes afterwards? E.g. icy conditions or dense fog. What if the original crash could communicate to cars behind it, wouldn't that be helpful if the crash isn't yet in the driver's (or car's) line of sight?

I agree "not much would change" overnight. It's just another input for the car's software to have at its disposal.

With the current hardware on the roads, I don't think it's technically possible for autos to achieve legitimate self-driving (if that's even the goal anymore?) - there are way too many edge cases that are way too difficult to solve for with just software.

Even if every car is required to be part of the network, there may be badly maintained cars that don’t work properly, or even malicious cars, that send wrong data on purpose.

At this point it's all about edge cases. Certain edge cases are impossible to overcome with just software + cameras alone.

We already know for sure that we can’t solve (1), which means we have to solve (2). Therefore, car-to-car communication is, at best, a value add, not the enabling technology.

You can't rely on this in real time because urban canyons make it hard to get consistent cell signal (for one thing), but you can definitely improve your models on this data once the data's been uploaded to your offline systems, and some SDC companies do this.

It involves a situation with networked self driving cars.

Binocular cameras provide absolute depth information, and are an order of magnitude cheaper sensors than the other options.

Since this technology is clearly computationally limited, you should subtract the budget for the sensors from the budget for the computation.

According to the article, the non-camera sensors are in the $1000’s per car range, so the question becomes whether a camera system with an extra $2000 of custom asic / gpu / tpu compute is safer than a computationally-lighter system with a higher bandwidth sensor feed.

I’m guessing camera systems will be the safest economically-viable option, at least until the compute price drops to under a few hundred dollars.

So, assuming multi-camera setups really are first to market, the question then is whether the exotic sensors will ever be able to justify their cost (vs the safety win from adding more cameras and making the computer smarter).

As seen on their website [1], and confirmed numerous times, they have monocular vision around the car and, though having three front-facing cameras, they each have different focal lengths and are located next to each other and thus can not operate as binocular vision.

[1] https://www.tesla.com/en_AE/autopilot/

Don't forget some interesting scenarios like dust particles or mud on one of the cameras (or windshield if cameras are located behind the windshield) or rain beading and distorting the image thus breaking the feature matcher and resulting disparity estimates.

Next, to "see" further, a stereo rig needs to have a decent baseline. For example, in a classic KITTI dataset, the baseline is approximately 0.54m which is much larger than, for example, human eyes (0.065m). Such baseline, 54cm, together with focal length, which, if I remember correctly, is about 720px in case of KITTI vehicle cameras, would give about 388m in the ideal case of being able to detect 1 pixel disparity. But detecting 1px of D is very difficult in practice - don't forget you will be running your algo on a car with limited compute resources. Say, you can have around 5px of D, that means max depth of around 77m - comparable to older Velodyne LiDARs.

Some of the issues I mentioned are not specific to stereovision (e.g. you need to calibrate monocular cameras as well and so on), just wanted to point out that stereovision does not magically enable depth perception. The solution would likely be a combination of monocular and stereo cameras, combined with SfM (Structure from Motion) and depth-from-stereo algorithms.

They measured out to 18m & point out that the typical measured limits of angular resolution of the human eye mean that we could extract stereo image information out to 200m or more.

This paper claims to demonstrate stereopsis out to 250m, which is roughly the limit you’d expect from typical human visual acuity: “Stereoscopic perception of real depths at large distances” https://jov.arvojournals.org/article.aspx?articleid=2191614

This paper suggests that steropsis occurs out to somewhere between 20m & 65m before other cues dominate 3D depth perception: “The Role of Binocular Cues in Human Pilot Landing Control” https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.30...

It seems that the claim that stereo vision only occurs in the near field case is probably wrong? Human stereo vision is much more capable than that & if it reaches out to significantly > 20m is surely being used when driving?

Human life is worth about $10 million (in US), that's a bit more than the sensor does. If one in 10,000 of camera-only car causes deaths, then it's not economically viable.

A London bus costs about $300,000, it is economically viable. Why is $1,000 sensor a problem. It is definitely viable to be installed on busses and trucks. Maybe you need to get out of the mindset of personal cars. It is not a viable business model, and it is not a viable model of dealing with congestion either.

The $100,000 system would only ever make sense if the fleet was already entirely self driving, and money for other life saving stuff like the environment and health care also hit suitable diminishing returns. Of course, by then, the cheap systems will have improved.

This argument holds for any non-negative dollar value you place on human life.

It is also independent of who owns the vehicles. Money the bus fleet spends in expensive self driving pulls money away from bus stop upgrades, pollution controls, etc, etc.

But I agree that current software (Tesla?) is not able to do that in the same way. So it may need more sensors until the software gets better.

In theory cameras should also be able to see more than humans. They can have a wider angle, higher contrast, higher resolution and better low-light vision than the human eye.

That said, I think what you're referring to is more extreme perspectives that shift in ways the car cannot drive and you are correct that this would aid in reconstruction. This is how NERF models do their 3D reconstruction (https://nerfies.github.io/).

Can't cameras do this by just comparing frame 1 to frame 2?

> monocular cameras can only provide relative depth

While the environment awareness is nowhere near as good as two or more cameras would be, if you consider the output over time, you get valuable information about the change rate of the environment, i.e. how fast that big thing is getting bigger, which may indicate one should actuate the brakes.

Of course, I'm with the crowd that answers the question with a "how many can we have?" question. The more, the merrier. And the more types, the better - give me polarized light and lidar, sonar, radar, thermal, and whatever else that can be plugged in the car's brain to make it better aware of what happens (and correctly guess what's going to happen) outside it.

“Monocular” cameras > no “absolute” depth > less safe

The last leap is not well justified.

Also, cars with vision based driving have multiple cameras. Whats the difference between a “binocular” camera and two “monocular” cameras?

How does a “binocular” camera get better depth information?

Is using multiple cameras to drive sensor fusion?

Why is absolute depth a strict safety win? How do you know how the sensor details translate to the final safety of the full system?

If this is just a handwavey upper bound on safety, how do you know that such a system can’t be safe enough for its design goals?

If humans with only one eye are able to drive, why wouldn’t mono surround vision be at least as good as that?

> Whats the difference between a “binocular” camera and two “monocular” cameras?

For the camera itself, nothing. They are probably referring to the implementation. You can have two cameras side by side but unless you are using homography to estimate depth from the two images, then your setup is monocular.

> How does a “binocular” camera get better depth information?

A pixel in two images (with known separation) will have a geometric relationship that can be used to extract depth information. This is a lot faster than alternative methods with a single camera and multiple images.

> Is using multiple cameras to drive sensor fusion?

This is really just a question of semantics.

> Why is absolute depth a strict safety win?

Why is it better to have two eyes than one? You can be more certain about what you are seeing.

> If this is just a handwavey upper bound on safety, how do you know that such a system can’t be safe enough for its design goals?

If you had a system with infinite compute you could probably do enough math to calculate absolute depth with 100% certainty. I believe you can already extract absolute depth with something called bundle adjustment-- but it requires multiple images since you are relying on parallax effects. It is also computationally expensive.

> If humans with only one eye are able to drive, why wouldn’t mono surround vision be at least as good as that?

Computers are not humans.

If you can get reliable depth information, the algorithm needed to avoid hitting stationary and slow-moving objects is extremely simple.

Is the stationary object in our path, of nontrivial size, and about to enter our minimum stopping distance? If yes, do we have a swerve planned that will let us safely avoid it? If no, emergency stop.

Because this logic is simple and well defined you can audit the implementation to the high standards applied to things like aircraft autopilot systems.

And it'll work even if the stationary object is something that didn't appear in your training data - you know the algorithm will work the same even if that concrete barrier is painted with some cheery flowers, or if that fire truck is airport yellow instead of the normal red.

Of course, this relies on the assumption you can get reliable depth information. If your depth sensor gets confused by a cloud of dust while driving in the desert, or gets blinded by the light of the setting sun, or is unable to detect a barbed wire fence, things are no longer quite so simple....

> If this is just a handwavey upper bound on safety, how do you know that such a system can’t be safe enough for its design goals?

Personally I would say that in freeway driving, a self-driving car should be able to avoid 100% of collisions with clearly visible stationary objects in dry, well lit conditions when all system components are in normal working condition.

... and I want to grab the guy who says that by the collar and scream in their face "The whole point is to build something that can do better than a human."

The second paper I reference in this comment https://news.ycombinator.com/edit?id=36232198 claims that humans can maintain stereopsis out to 250m.

That’s a huge difference from 10m & if true suggests that human drivers might well use 3D vision when driving.

IMO it's better for us to focus on what the robots can and cannot do right now, and focus on solving those problems :)

Thanks for the sources though, those papers are definitely neat and I'll be taking a look when I get a chance.

It's pretty easy to see, even as someone with very little experience, the benefits of stereo vision over monocular. In addition to the depth stuff it's a lot easier/faster to create your point clouds from disparity maps.

Stereoscopic vision in humans only works for nearby objects. The divergence for far away objects is not sufficient for this. You may think you can tell something is 50 or 55 meters away through stereoscopic vision, but you can't. That's your brain estimating based on effectively a single image.

That said reality is not a single image, it's a moving image, a video. Monocular video can still be used to estimate object distance in motion.

Eventually AI will be good enough to work better than humans with just a camera. The problem is we're not there yet, and what Tesla is doing is irresponsible. They should've added LIDAR and used that to train their camera-only models, until they're ready to take over.

Honestly while ggml is super cool. It started as a hobby project and you probably shouldn’t use it in production. ONNX has been the defacto standard for ML inference for years. What it is missing (compared to ggml) is 2-6bit inference which is helpful for large scale transformers on edge devices (and is what helped ggml gain adoption so fast).

ONNX really is the universal format. If you can get your model exported to ONNX, running it on various platforms becomes much easier.*

*as long as every hardware platform supports the ops you use in your network and you're not doing anything too fancy/custom :P

Edit: often if you go through the ONNX intermediate format, be prepared to perform some 'network surgery' to clean up some conversion cruft, but also to remove training-only stuff left in the network...

Quick disclaimer that this doesn't reflect the views of my employer, nor does any of what I'm saying about self-driving software apply specifically to our system. Rather I am making broad generalizations about robotics systems in general, and about Tesla's system in particular based on their own Autonomy Day presentations.

When you drive on the road as a human, you rely a lot more on intuition and feel than exact measurements. This is exactly the opposite of how a self-driving car works. Modern robotics systems work by detecting every relevant actor in the scene (vehicles, cyclists, pedestrians etc.), measuring their exact size and velocity, predicting their future trajectories, and then making a centimeter level plan of where to move. And they do all of this 10s of times per second. It's this precision that we rely on when we make claims about how AVs are safer drivers than humans. To improve performance in a system like this, you need better more accurate measurements, better predictions and better plans. Every centimeter of accuracy is important.

By contrast, when you drive as a human it really is as simple as "images in, steering angle out". You just eyeball (pun intended) the rest. At no point in time can you look at the car in the lane next to you and tell its exact dimensions or velocity.

Now perhaps with millions of Nvidia A100s we could try to get to a system that's just "images in, steering angle out" but so far that has proven to be a pipe dream. The best research in the area doesn't even begin to approach the performance that we're able to get with our more classical robotics stack described above, and even Tesla isn't trying to end-to-end learn it all.

That isn't to say it's impossible (obviously, humans do it) but I think one could make a strong argument that "images in, steering angle out" is like epsilon close to just solving the problem of AGI, and perhaps even a million A100s wouldn't cut it ;)

The best human drivers do this not at centimeter, but at the millimeter level. Look as downhill (motor)bike racing, Formula 1, WRC, etc..., These drivers can execute millimeter level accuracy maneuveurs that are planned well in advance at over 100km/h.

Basically humans are really really good at guesstimating with great accuracy (but poor reproducibility) and since we don't use basic measurements in the first place, having better measurement accuracy wouldn't really help us be better drivers on average (it does help for certain scenarios like parking though, where knowing the # of inches remaining to an obstacle can be very useful).

But for everyday driving at speed, we wouldn't even be able to process measurements in real time even if someone was providing them to us. AVs are different and that's basically the gist of what I was trying to say. Because they actually do use, rely on, and process measurements in real time, improving their measurement accuracy (ie. switching from camera based approximate depth, to cm level accurate depth from a LiDAR) can have a meaningful impact on the final performance of the system.

So if I was a betting man, I'd say that you can expect Cruse, Waymo & others to scale a little bit now, just to show investors that they can but for them to really save the bulk of the scaling (to hit that targeted figure of 1B/yr of revenue) until after they've found a way to get the costs down. That's going to come in the form of more bespoke vehicles that are better vertically integrated with custom hardware and sensing solutions (like the Cruise Origin).

Tesla of course claims that the HW3 cars will still get FSD at some point, but unless they somehow figure out how to bend light, that blind spot will continue to be an issue on older cars.

The fact that some early HW4 units will ship with the updated camera suite but not the Radar only further adds to my point that some users are going to be left with inferior sensing systems despite having paid the same $10k for FSD as everyone else. The whole "radar will be used to train and improve the vision" argument is just nonsense made up by Elon & Tesla fans. A properly functioning radar camera sensor fusion system will be superior in every way to a camera only solution. And there will be 0 Tesla's that actually achieve "full self driving" (ie. you going to sleep in the back-seat and waking up at your destination) until Tesla adds things like a cleaning system to their existing camera solution for example. The hardware is simply inadequate.

I'm saying that, as we see, HW4 and radar are two distinct hardware configurations: saying that there will be a large sore spot because HW4 cars have radar is objectively false as not all of them do.

The fact that HW4 and radar are separate configurations by name is not important. HW3 is also included in the mix (you can buy FSD on it) and it has totally different camera placement. So radar aside there’s still a difference in the sensor suite.

People who paid $10k and were promised “FSD” and future hardware upgrades have every right to be pissed off about this.