Comment by rpcope1
1 day ago
One of the things I never see in reports like this is what sensor is being used (is it a Bosch unit? Sensiron? Plain thermistor? Pixies and Fairy Dust? Some bodged together 1wire deal?) and what MCU (Nordic? Espresif? Microchip? Something else?) as those have a huge bearing on what the sensor is actually capable of, and would say a lot about precision/accuracy and battery life.
How is that relevant for the average buyer (as much as these things have an average buyer anyway)? Precision, accuracy, and battery life is helpful info but the inner components are an irrelevant detail. It's not like having a high quality part prevents them from using it wrong anyway.
The datasheet gives you much better fidelity data than an N=1 test. You'll never know what the tolerance is if you're just looking at one unit. Presumably the average buyer cares if they expect the temperature sensor to be 5% off on the unit they buy. The accuracy of all of these sensors have ratings, but they are often just copying the value on one component and not even considering a stackup. Devils lie in details.
Regarding accuracy, it's been about a decade, but I did some embedded projects related to accurate temperature measurements back in the day. From what I recall, anything sensor that is < $100 or so is going to be 5% margin of error at the very best, and regardless of price, you're going to have to calibrate it and figure out the curves in your ADC, it's not going to be linear. Additionally, just the act of measuring will heat your sensor slightly so that will need to be taken into account as well.
In short, if you need less than a couple degrees of accuracy, it's going to take a bit of tinkering. I wouldn't trust a $1 sensor for a tropical fish tank, for example, but probably good enough for a home thermostat.
2 replies →
The datasheet will gives you absolute best case scenario. Like current consumption in sleep mode, or what register to set for best accuracy. It does not mean the Firmware will do exactly that.
1 reply →
> How is that relevant for the average buyer
Avg buyer wont read a token of tests before buying whats marketed best.the whole point in test like these is to provide non-average buyer with info average buyer doesnt care for.
Average buyer googles "best cheap zigbee temperature sensor" and reads this article.
People who read ingredients lists of food are a tiny, parochial subset of consumers and yet they strongly influence what is on shelves.
The average person doesn’t read Hacker News, so why does it exist, and why are you here in the comments?
Similar reasons: the intended audience isn’t the average person.
There's no guarantee that the COOL SMARTHOME SENSOR V1 has the same sensor or MCU like the COOL SMARTHOME SENSOR V1 that I can buy. Sure, it's interesting to know, but it's not too usefuly
If they replace the senor then all properties which depend on the senor may change.
So because it could be incorrect, you'd rather not know at all?
Chipset models only form a weak lower bound on crappiness.
You can certainly wreck connectivity with bad and old firmware. Similarly, housing design, PCB thermal isolation, not putting the sensor next to the MCU, measurement averaging settings, etc. all are going to affect accuracy even with the same sensor (particularly dynamic accuracy).
Sometimes Smarthome Scene does do teardowns and checks which chip is used. I guess they didnt' this time due to the number of devices.
I think buyers care more about accuracy and resolution than which sensor. If it’s accurate, with hugh resolution and has good battery life, do you care if it’s Bosch or A Chinese name you’ve not encountered before?
I don't so much care if it's a Chinese knock off or a real deal Bosch BMP, but if you're interested in measurements that are better than vague vibes, the difference between a cheap poopoo glass thermistor hooked up to the noisiest 8-bit ADC (let's be real about what's actually in there), sets a real limit on anything that device will ever be capable of, especially if you have more than one. If a device has got a real deal BMP581 (yes that's a pressure+temp sensor, but let's roll with it) that's a whole other ballgame, and yeah they could have messed up the housing or it could brown out, have bad lower power characteristics, or bad firmware, but that BOM difference says a lot about what the engineer who designed it probably thought the device should be capable of.
Maybe none of the tested devices is available but another one with similar hardware is.
For most of these small sensors you'll need to crack open the device to find out exactly what hardware is really there. More than once I bought at different times apparently identical Zigbee sensors which behaved very differently and it turned out they were different inside despite having the same model name. Probably a hardware revision thing but with no clear markings. Manufacturer couldn't find some component anymore or not for a good price and instead redesigned the device.
Those who care about the internals are techies who maybe want to build their own or are very curious. Most buyers don't care about that, they care about how well it works in practice. Just like they don't care what RAM chips are used in their phone or computer, what modem, what screen panel, what camera sensor, etc.
If they change components the properties in this test would also change so not sure I get your point other than that any test is useless since manufacturers change stuff without telling the buyers pr changing model numbers.
Unfortunately a lot of these tests are useless because the manufacturers often change the insides without warning. It’s better when they mark different hardware revisions but still not great because reviews rarely get an update.
I have a bucket of devices which don’t do what I hoped they’d do because they were different from what the reviews said. And not just IoT unfortunately. Some do a better job at flagging these changes, some don’t.
But in the end taking the common case that the product is what it says on the box or that the changes have no impact on how it works, most people don’t care about the name of the chips. Not unless they are very keen on the tech and then there are more tech-oriented niche sites that might provide this.
1 reply →