For a split second, I visualized you walking into a brick and mortar Microsoft Store to get your security update done on your laptop, and somewhere in the middle of that split second I caught myself thinking „typical Microsoft“, and then I came to my senses.
My smart fortwo has two clocks, one on the radio, one in the driver’s dashboard;
Both go wrong over time, but both in two different directions, so the trick is to calculate the „average“ between both, and you end up with roughly the correct time.
My first car, a Hyundai i10 I bought back in 2010, was very basic and had few features. It only had a simple radio with an FM/AM tuner and a CD player, no navigation system.
What surprised me most was that the clock adjusted itself automatically whenever daylight saving time started or ended.
There was no calendar or date setting, so I assume it must have received the time either via RDS (I don't know whether RDS supports this) or through a built-in DCF77 receiver.
All my subsequent cars had far more features, yet none of them had this one.
I also wonder why cars with built-in GPS navigation don't synchronize their clocks using the time from the GPS signal. Heck, the car could even determine your time zone from your GPS coordinates, so there'd be no need to configure anything manually.
Checking wikipedia, there is optional support for transmitting time and date in RDS, with an explicit usecase of updating the clock in cars.
> I also wonder why cars with built-in GPS navigation don't synchronize their clocks using the time from the GPS signal.
GPS doesn't transmit timezone information, nor does it transmit leap seconds. It transmits a very, very accurate UTC minus 18 seconds, but without access to the internet for timezone database updates or (until recently) leap seconds, it's a bit useless for updating a clock.
I'm not sure if you guys read the article, but OP's car does use GPS for the time, applying a user-settable offset (or in OP's unfortunate situation, a fixed offset).
It did in the 1999 Mercedes-Benz S-Class I purchased after 15 years of insane depreciation.
I thought I broke the clock with a DIY repair because it too couldn’t be manually set (aside from enabling a +30 min offset). Luckily it fixed itself after driving out of the garage when it found a GPS signal.
My '97 Range Rover has an analogue clock with a quartz-controlled movement, and a button to set the time.
It gains anything up to an hour a day, particularly when my six-year-old is helping me clean the car and he pushes the button to watch the hands go round really fast.
It's fine. Time is time, we get there when we get there.
Who said anything about getting there faster? What if you are stuck in traffic and need to know if you will be on time or late? What if you see a fuel station and want to know if you have enough time to top up your tank? There are actually quite a few valid reasons that you need to know the time while travelling in a vehicle
If you’re stuck in traffic and you know you are goung to be late, you can phone ahead to let them know! This won’t be possible without knowing the current time, something which a reasonably accurate timepiece will inform you of
But I don't need to know the time when I'm going somewhere. I need to know the time before I set off, so I set off at the right time, and then I get there at the right time.
Okay, have you ever visited Earth before? Space and time are both surprisingly linear here.
If I want to be somewhere 250 miles away, I know it takes about six hours to drive there. So, in order to get there in time, I leave maybe six and a half hours before I need to be there. If I know the road is going to be busy - quite often where I live there are a lot of tourists doing 30mph with no idea there's anyone else on the road - I'll give it even more time.
MOE: How long has that been in the soup, froghead?
CURLY: About uh--- [Looks at the three watches on his wrists.]
MOE: Hey! What’s the idea of the three watches?
CURLY: That’s the way I tell the time.
MOE: How do you tell the time?
CURLY: This one runs ten minutes slow every two hours. This runs twenty minutes fast every four hours. The one in the middle is broken and stopped at two o’clock.
MOE: Well, how do you tell the time?
CURLY: I take the ten minutes on this one and subtract it by the twenty minutes on that one. Then I divide it by the two in the middle.
MOE: Well, what time is it now?
[Curly grabs a clock from the inside of his jacket pocket.]
CURLY: Uh, ten minutes to four.
MOE: Hmm so it is!
[Moe grabs the clock from Curly and hits Curly on the head with it. The clock breaks.]
It's not just cars. Our house alarm had two clocks, one in the alarm itself, one in the display/control panel, and the two didn't talk to each other, not just for that but everything else as well, for example motion sensor names.
It's now in an e-cycling waste bin somewhere, replaced by an ESP32 I/O board and Alarmo in Home Assistant, at 1/20th the cost of the original PoS alarm.
Well, the positive traits you listed are certainly true, but let’s face it: she was also very intelligent and had a self-deprecating sense of humor, so there’s that.
Oh I left super smart out, but the self deprecating I felt like I got with self-effacing ("It takes a lot of money to look this cheap" is Mae West level gag writing).
I'm a dev not a video or 3D guy, but I'm looking at people I used to know and also some youtube doom-scrolling, the blender MCP, cinema 4D MCP, and seeing that it's a big impact
If used correctly, I find the em dash one of the most elegant and reader-friendly ways to structure certain parts of a text — that it got such a bad reputation lately really is a shame!
Allow a question from someone who’s only got a very vague idea of how this kind of stuff works behind the scenes: say I rent usage of this model through one of the many LLM hosting providers out there, and let‘s assume I use it extensively through something like Pi or OpenCode and vibe code away all the time, keeping the hosted model occupied as much as I can, happily burning my credits.
Does that mean that there is a hardware cluster as described by you above that is crunching away just for me?
So at FP16, I alone keep a 1,664 GiB system occupied all the time?
No, a cluster can server multiple users at the same time, providers cap the tok/s so that one cluster can run inference on multiple inputs at the same time. OpenAI with their new ultrafast mode is probably reserving the whole cluster or prioritizing requests of ultrafast users above others with a higher tok/s hence the high price and high speed. There's many other knobs providers tweak that they don't show the users, for example I doubt many providers are hosting the full FP16 version.
The "expensive part" of generating the next token is streaming in the model weights from memory. The computations are relatively simple, which is called a "low arithmetic intensity" in industry jargon.
So what they do is batch multiple chats together and compute the neuron activations for all of them together.
This is vaguely similar to how some database engines work, where if multiple users need to run a "whole table scan" query, the additional users "join" the streaming workload of the first query mid-way, then loop back around to complete the first part that they missed. The AI accelerators don't do this looping, but the concept is the same: amortize the expensive I/O over multiple computations running in parallel.
The "turbo mode" token rate thing is almost certainly your query getting sent to slower or faster hardware, like B200 vs newer B300 kit.
It depends hugely on what "rent usage of this model through one of the many LLM hosting providers" means. If you're asking them to host the model privately then yes, all of that 1.6T of RAM is likely in use holding weights, activations and KV cache by an inference engine that's only getting/answering requests from you alone. When you aren't actively using the model the hosting process is still active and waiting with all of that memory still wired to it.
As background: For the most part VRAM oversubscription/paging/swapping isn't a thing in the same way that RAM for a VM often is. There are some approaches to it, but (to my knowledge) not at that sort of scale.
There are some systemic reasons for this, but very broadly speaking the GPU vendors are building toward the highest bandwidth and lowest latency possible, and the overhead/complexity of something like protected memory modes serves neither of those priorities.
The obvious book recommendation here is: Joseph Henrich: The WEIRDest People in the World: How the West Became Psychologically Peculiar and Particularly Prosperous.
The funny thing is that in spite of reinventing so much of it, the event still feels dangerous and leaves you to make your own decisions.
Burning Man still enables and allows dangerous art that you can fuck up your shit on. The organization has an interesting take on this: they allow dangerous art, so long as it looks as dangerous as it is. They don't want participants to be surprised that something was dangerous, because they want you to be able to do your own reasonable risk assessment.
(Though, I think the really impressive installations are the ones that look/feel more dangerous than they actually are.)
For what it's worth, I think this is a great exercise that should be undertaken now and again: you may not realize the ways in which a system is good or bad until you endeavor to tear it down and rebuild it from scratch.
In the case of Burning Man, there's structure that let's people be mostly safe by default, while supporting some reasonable risk taking. In the case of crypto, I think we've seen use cases where there's value in having a means of storing and transferring wealth outside of government control, which we saw the benefits of early on as governments tried to financially wall off groups like Wikileaks.
Does that mean we want all of the safe guards gone? No. Does that necessarily mean society and its structures are perfect and should never change? I think obviously, also no.
Interesting, I came to the same conclusion, making my dangerous robot look dangerous on purpose to keep people away from it. Are there actually written rules on this policy somewhere?
This is one reason I oppose rounded corners in GUIs. A computer is a power tool for information manipulation. Power tools are always dangerous. When you can cause massive data loss or worse with a single mistake, it shouldn't look "friendly", any more than a firearm or a table saw should look friendly.
I recall laughing with a friend as we realized the railing on the second story art car we were on was attached by staple and hot glue, but back when I went we assumed things unsafe
This is actually what I was thinking of when I made this comment. I witnessed someone in some position of authority (from either DPW/Artery/someone else clearly sent to check it out) on the first day talking to one of the artists. They were literally watching people climb it and were like "yep, that looks pretty dangerous" and then went on their way.
It felt safe-ish to climb at first, until about halfway up when you could feel it swaying back and forth in the wind. I am usually the "climb dumb shit at Burning Man" type, but had my stomach drop about 3/4 of the way up and had to abort. It was very easy to grip, nothing felt insecure, etc, so it was definitely "safe" if you knew what you were doing. But if you ate it off the top of that, you'd be fucked.
>so it was definitely "safe" if you knew what you were doing.
Those quotation marks doing a lot work...I guess those IBCs are 1m tall? so 8m at the top. A fall from 8m could result in severe injury if not fatality. Fucked indeed...
You're one unfortunate motor control fuckup away from killing yourself pretty much always. Most animals are good at not killing themselves and humans are decent at climbing to boot.
Yeah, if you fall you might hurt yourself, but when you're paying attention to what you're doing you're likely less at risk than when you absent-minded walk down some stairs.
We spent probably hundreds of hours climbing all manner of tall playground structures as children, and I cannot recall anyone ever falling. I do remember some instances of people tripping and falling down a flight of stairs though...
A while ago I was at a playground chatting with a neighborhood mom, and her 5 yr old son fell off the top of a spiral slide, and broke both his forearms. It's sobering how quickly and unpredictably a situation can turn from pleasant to horrible because somebody's grip on a railing was an inch off.
There are things you do 100,000 times in your lifetime, with a 99.999% success rate, where failure may kill you. You can't really avoid those.
well, no, because the individual doing the thing isn't the only factor. You may very well know what you're doing but that doesn't mean all aspects are under your control.
I think the implicit assumption you may be missing is that "know what you are doing sufficiently" includes knowing when not to do something because it involves unpredictable risks.
Like free solo rock climbing: it's actually fairly safe if you do it on a route you know inside out, that is well within your ability, and in good condition. But if it's not in good condition and holds might come loose, nobody in their right mind - and the knowledge to recognize the condition - would free solo it.
It's not just a truism. The point is that sufficient knowledge allows you to understand and avoid or mitigate risks much more specifically. With your example, if you have only freshly fallen snow, and a slope of less than 30 degrees with nothing above it, avalance danger is so ridiculously small that it counts as "safe" for practical purposes.
How the heck do you even climb this. I don't see any places to grab hold? There are gaps between the cubes I guess but those seem awfully smooth and slippery...
The metal cage that the plastic cubes sit in. On the corners where they bend, there's room to get fingers wrapped around. They're also rectangular and pretty chunky. It's not super comfortable, but it's secure.
Also, fun fact: they are literally held together with large metal zip ties.
humonkeys .. we loose something when we disallow going for the treetops.. yee who dwells in the safety of the savanna is condemned to inventiveness, fire, spears - tiger, lion and bears..
reply