Hacker News Reader

Comment by solardev | Hacker News Reader

Hmm, thanks for explaining the profiles. I'll try a few and see how they taste.

> The rice cooker doesn't know either, it will take whatever is needed to take. Short of putting an extensive sensor array and ask you for the kind of rice you put in it, there's probably no way to actually predict it.

Why doesn't it know when it's so consistent every time? How does the ninja know? How does this one know when it's closer to 10 minutes? Seems like you should be able to estimate it down to a within a few minutes at start, then refine that estimate if need be. It's helpful to know whether something will take 15, 20 min or 45 to an hour.

View on HN

> Why doesn't it know when it's so consistent every time?

Because rice isn't consistent.

This is most apparent with certain varieties - a cup of imported Basmati "Royal" brand will take longer and require more water than the same amount of "Three Elephants" Jasmine rice from Thailand.

Even the same varieties produced by different brands/countries will turn out differently in terms of time required. This relates to differences in climate, storage, processing, and other factors.

On the one hand, I can imagine some estimation solutions that all make me think about software installation progress bars.

But on the other hand, if this is a piece of equipment you use regularly, you'll learn pretty quickly how long each thing typically takes, and after that initial learning period, the fuzzy progress indicator is no longer very valuable, and doesn't seem worth investing the engineering and UX time (and resulting higher product cost).

As a software-oriented community, I understand the impulse to throw code at the situation, but this seems like a situation where the cost of doing so doesn't really help very much. And it might actually make things worse, since most users are conditioned to trust/believe the sensors/indicators they see, and a countdown that keeps dynamically changing/jumping/slowing will confuse quite a few people who don't understand why it's a hard problem to solve.

And I think it's worth noting that this is par for the course with most things related to food prep. I'm always looking up cooking times for various things until I memorize the things that I cook often, and this seems like just another case of doing a quick search for a timetable.

Imagine you are making rice in this for the first time and you have no idea how long its going to take. That is ridiculous when you think about it. How do you even plan the rest of the meal assuming you won’t just be eating rice? Take the loss the first time and take 90 minutes to make dinner since the manufacturer didn’t have actually cooking with the thing in mind I guess.

> Imagine you are making rice in this for the first time and you have no idea how long its going to take.

How many times do you use it for the first time? Seems like a bit of an edge case to spend engineering time/add extra parts for.

I don't think this is some insolvable problem. It's literally estimating the time it takes to cook rice, and many other rice cookers (including other Z models, apparently) do this just fine.

It's not some edge case, it's literally the biggest question after you push start.

Imagine you’re grilling steak (or making lasagna, or cooking chili, etc) for the first time. The same principles apply. It’s easy to look this up, and once you do, the problem is solved.

Expecting a cooking tool to hold your hand to this degree doesn’t seem reasonable to me.

I’m almost certain that the manual gives an approximate time range for each mode.

Well, because rice is inexpensive, you.... could do a test batch? Sure, it would add $0.25 or something to the cost of your $100.00 rice maker.

Or you could just put the rice in extra early. It will get done early. But it will still be hot and perfect when you're ready to eat?

It feels like you have a big misunderstanding of rice cookers.

If it's your first time using it, you'll probably read the manual which will contain a table of general cook times. I just used a rice cooker with no screen for the first time and I had no problem with its lack of UI.

> Why doesn't it know when it's so consistent every time?

They had a feature (that I believe it's just integrated into every model now) labeled as "neuro fuzzy", decades before the neutral networks boom. I believe it's really just describing hysteresis. So it's doing things based on feedback from whatever sensors they have, but there's no learning from run to run (based on how you actually use the thing).

I also hate its lack of estimated times.

Different rice varieties have very different cooking times, also ratio or rice to water also influences the time, so the machine can't just estimate based on weight.

What sensor do high-end rice cookers have that varies the cooking dependent on the rice variety? Steam?

Simple rice cookers have only a 100°C temperature sensor (or switch) that turns off the main element once the water is boiled.

Some rice absorb water very quickly, some rice very slowly, some rice you want more water, some very little, sometimes you want to add other ingredients too. That's more than enough to vary cooking times but not be able to estimate cooking times via the sensor.

Obviously.

My question is what sensor do they use to detect the difference?

I did some Googling but couldn't find any info.

IIRC, it's just temperature (and maybe time). Once the bottom of the pot goes above boiling, all the water has been absorbed/boiled off. Cheap rice cookers then just click over to warm, fancy ones might do something based on time, which can change based on other things like how log it took to come to a boil or how long between coming to a boil and the water finishing absorbing.

Usually the cooker senses the temperature and amount of unabsorbed water left (conductivity ? ai forgot the exact input but Technology Connections had an excelent video on youtube on rice cookers for those interested).

That's enough to manage a temperature curve and predict the last 10~15min of cooking (mine shows a time only 15min before the end), and doesn't require much input from the user.

It’s not a timer function but a sensor that detects signals the rices is done.

Depending on environmental conditions (temperature, etc) and amount being cooked, will vary but the results are repeatable quality rice, rather than a fixed cook time and variable quantity rice.

More expensive but better quality rice.

Reply on news.ycombinator.com