Dyson Lightcycle Task Lamp
dyson.com
dyson.com
[0]: https://en.wikipedia.org/wiki/Semi-Automatic_Ground_Environm...
[1]: eg https://www.computerhistory.org/revolution/mainframe-compute...
I eventually found out if you use an AUX cable (not included) you can connect the two devices together without an app.
For a soundbar, that sits in my living room to produce sound from a television, maybe stream music to it via Bluetooth or listen to online radio why does it need to send analytical data home! I run my own local DNS server with blocklists with one domain being metrics.bose.com. No way am I buying the newer models with built-in microphones.
Likewise, I have the soundtouch on it's own IoT of crap VLAN which has very restrictive firewall rules. I had to jump through several hoops to get this to work so I can control it from my phone which is on a trusted VLAN. This should be straight forward enough using avahi mDNS reflector and igmpproxy however Bose says the Soundtouch does not support cross VLAN https://community.bose.com/t5/SoundTouch-Archive/SoundTouch-.... That's not completely true, by default it does support it as it is just standard networking with unroutable requests being past to the default gateway however Bose have gone out of there way and done extra work to explicitly block traffic from a different subnet the soundbar is on, probably as a cheap way of securing the device for people who leave their internet routers wide open. The proper solution would have been for Bose to provide some authentication mechanism between the app and soundbar. Luckily it's easy to work around with NAT / IP masquerading.
Even with this, I've had the app switch the soundbar from it's IoT wifi SSID to the trusted wifi SSID on one occasion. I'm not sure how or why but I'll be monitoring it.
I'm going to actively avoid IoT devices now on, the general implementation seems to be substandard, end-of-life will generally be the end of app updates instead of the device physically breaking. Sonos have already done this by essentially changing the terms and conditions for users with existing devices and users not agreeing have their devices essentially become bricks https://www.zdnet.com/article/sonos-accept-new-privacy-polic....
Its not too hard but you may want to dedicate an hour on the weekend to set it up.
It might work now, what if someone needs to reconnect it 3-4 years from now and the app no longer runs?
The app which now displays the "This app needs to be updated by the developer to work on this version of iOS" message.
That ain't gonna happen. (Not even sure if anybody still has the keys to that Apple Developer account or not. The end of that startup was not very pretty...)
Fortunately, this device exposes all its functionality via a local web server - the app was mostly a neat way of finding the device's local IP address then opening up a web view for that. So at least it still works and can be configured - you just need to know how to find it's IP address via other means...
Six decades? We didn't quite make it 6 years...
:sigh:
Dyson had one decent idea that everyone else adopted too - cyclones instead of bags. Everything else seems to require paying 10x what a sane premium and expensive cost might be - lamps, hair dryers, fan heaters.
The innovation is making it appear complex enough to justify the absurd price. I bet they are still using too much ABS plastic that'll be disintegrating after 5 years.
They updated those that were sold to disable the contra-rotating. No idea why.
In short, I can't imagine a more perfect hair dryer.
Sadly, I am bald and do not get to use it. I did spend a while gushing about how neat it was, though.
I'm sure there's one or two others selling emperor's new clothes at absurd prices like Dyson too -- that's very far from "often".
The LED component probably will last 60 years (8 hours a day). My guess is they are using Nichia LEDs, which tend to have an L70 (lifetime at which the brightness stays above 70% of the original output) of 150,000 to 300,000 hours. If you run them cool enough and at the right current/voltage, the LED itself will actually last that long.
However, the LED is probably the most reliable part of the whole lamp. You might see a few units last that long, but most of the other electronics will have failed by then. And even if they have not failed, both the LEDs and controls will be seriously out of date by then.
If you look at the market Dyson is in (very high-end lighting), $600 to $900 is in the same order of magnitude on price. Brands like Artemide, Marset, Flos, Roll & Hill, and Anglepoise all command prices most people would think are ridiculous. With Dyson, you get a lot more functionality than those other brands, but I would consider this purchase more like a luxury car (service life of maybe 20 years, original owner has it for 3-10 years) than a high-end watch (could last many decades with consistent service).
I've seen the previous task lamp from Dyson in person, and it does use a lot of metal (minimal ABS). But that doesn't change the argument around electronics. I feel like Dyson really didn't have to make the ridiculous 60 year claim. Enough people in the target market would still buy it.
One "advantage" of consumerism is cheap = changeable
The sad part of modern technology is that you can't fix it yourself, nor do you understand how it works. It doesn't make you any smarter.
That said, having a luxury product that last 60 years is a very niche market.
This. Technologies evolve/improve so fast, you don't want to be stuck with them forever anyway.
But then again, how many years could it be operational when depending on an app.
There's a whole subreddit devoted to it, /r/buyitforlife.[1] I guess that's not much of a claim, since there's a subreddit for everything, but this one is kinda neat. :)
That said, I think what a lot of people take issue with on Dyson devices is the extreme price premium which isn't really required most the time to achieve the advertised level of quality.
I do miss having items that when they break, I could pull apart and repair too.
Yet they only provide a 5 year warranty. I would expect a $600 desk lamp to have at a lifetime warranty.
And they couldn’t even be bothered to get the warranty details right: at the top of the page, it says parts and labor, but at the bottom it says parts only.
Or in many/most CPU and GPU heatsinks for the last decade or so, both Intel and AMD have heatpipe-based stock coolers in their range. There are heatpipe-based heatsinks available for $10~15.
If you're going for the best possible LED lamp, irrespective of cost, I can't think of a better cooling method.
Kinda? It already exists, but usually to better spread heat to the sink in high-power LED lamps (50W+, not equivalent, actual).
And my point is outlining that it's "satellite technology" is complete bullshit marketing copy.
I would expect more or less every laptop with fans[0] to use heatpipes, even for gaming / workstation laptops it hasn't made sense to stick a radiator on the CPU/GPU in a long time.
[0] apparently fanless devices don't, I guess there wouldn't be enough of a heat differential between the CPU and a hypothetical heatspreader for it to work?
Actually, I have a friend who works at a company which designs extreme heat pipes and heat wicks for satellite applications.
He has a satellite-spec heat pipe specimen which he shows around with heat-sensitive paint to demonstrate heat transfer speed, and it's way faster than conventional heat pipes. It's almost instant. Don't know the secret sauce though.
Clearly I'm missing something.
Even that is not.
...water cooling process works by condensation. As the processor heats up, the liquid essentially vaporizes, keeping the CPU cool. The vapor then travels to the opposite end of the of the heat pipe, where it condenses back into liquid when cooled off.
2016. Samsung Galaxy S7 https://www.howtogeek.com/365535/how-does-water-cooling-work...
But for an LED? Come on.
The cutting edge technology in LED lighting is to use power supplies that don't need electrolytic capacitors. The capacitor usually fails long before the LED.
> Even that is not.
Did you mean to claim that a heat pipe is not reasonable? (the rest of your comment doesn't make your point clear)
The daylight cycle matching (and the manual tuning) varies color temp between 2700K and 6500K; it doesn't (at least, in any of the description) vary light level. It's basically the same concept (but more localized) as f.lux for desktop screens.
The entire 'warm white' concept is based around the colour temperature of candles, those being traditionally associated with night illumination. But look at the actual colour temp of an overcast sky just after Sunset; it's 5000K+ blue-grey, not 2500K yellow. The latter has no place in nature other than fire. Yet somehow our ancestors managed to go to sleep...
I think you just mean colors, but tried to find three different ways to say it (and kind of failed, because lots of people “deal with” document all the time and aren't concerned with color.)
At any rate, the light came be fixed at any supported point, too, the matching is just an available function, not something you are locked into. So for uses for which it doesn't meet your needs, you would just not use it.
> But look at the actual colour temp of an overcast sky
Not sure about handling the overcast part, but that's explicitly what this does otherwise.
> The latter has no place in nature other than fire. Yet somehow our ancestors managed to go to sleep...
Not sure what your point is, since the color of natural light just after sunset is only relevant here if you think humans are optimized for sleep at twilight. Which...they aren't.