5 ms·
I'm curious, how do you know that you need 1) 3500mah battery or 2) 2.1Ghz processor or 3) 4GB Ram on a smartphone? These things are meaningless without 1) det
by mrtksn 9y ago
I'm curious, how do you know that you need 1) 3500mah battery or 2) 2.1Ghz processor or 3) 4GB Ram on a smartphone? These things are meaningless without
1) detailed energy consumption analysis.
2) SoC Architecture details
3) OS Architecture details
Even if you are knowledgeable enough to understand all these, that information is useful only if you are going to do something different than the original design purpose. If you are going to use the product for the purpose that it's designed for, all you need is the average use time between charging.
Maybe you are an expert in the area but what about washing machines or tables? How likely is that the engineering specs of these products will also be in your domain?
- TeMPOraL 9y agoFor a random, unique electronic device alone, the specs like you gave are difficult to make use of; but in reality, consumer electronics is not random - most competing devices are in the same ballpark of functionality and internal implementation, so battery capacity, processor speed and amount of RAM can be easily used to compare between models and against expected use cases. E.g. if last year's model with 1GB RAM choked on its own stock operating system, I can be sure that this year's model with 1GB RAM will be choking even more, and it's probably a safe bet to go for the 4GB one. I do not expect consumers to have expert knowledge in every possible domain. But some things can be estimated pretty easily if companies are providing relevant data. Which they go out of their way to avoid, to muddy the waters, so that people don't ballpark things. You said washing machines or tables - two areas which are as far away from my expertise as one can be; I still believe I'd be able to do a good job shopping if I had the following data: For washing machines - peak/average power/water/detergent use on "typical" and "highest cleaning" (worst efficiency) program. Also peak and average generated noise (in dB) for those programs. For tables - size, weight and supported weight (I'm flabbergasted by companies which don't provide the last; do people buying desks/tables not care about the safety of secondary use cases like "sitting on a desk" or "standing on a desk to change the lightbulb in the ceiling light"?).
- mrtksn 9y agoHow battery mAh is any better than Talk time, browsing time and video watch time that Apple provides? After all, all you'll do is to try to guess these things from the battery capacity but despite what you believe battery capacity and use time are not always correlated linearly as the radio, SoC efficiency and OS optimisations etc. will have dramatic effect. Giving engineering specs for consumer products is often very easy way to con to the consumer. All these knock off cheap devices have amazing specs that are advertised very loudly. An interesting observation is that you actually don't want engineering specs about washing machines or tables :) That's probably because you are not very knowledgable about these things so you seek information that will affect you as a user. You are interested if you can fit the table and how much it will cost you to operate the washing machine, which are not exactly technical specs of the product but impact on the user. If you step out of your techie shoes and step in the shoes of a washing machine nerd or a carpenter shoes you will find out that all you want to know bout a smartphone is how long the battery lasts and does it work nicely. Now, you probably want to know the type of the electric motor of the washing machine, it's power circuit and the material of and type of the pipes. As a carpenter you probably want to know what kind of process was applied to the wood and what's the brand of the glue used.
- TeMPOraL 9y agoTalk time, browsing time and video watch time are even worse than technical specs, because they're heavily dependent on specifics of particular use cases. Is it "time watching Netflix" or "time watching YouTube" as tested by C-suites' kids? Or "time watching a video on our optimized player under minimum brightness, no connection and all other measures of power optimization engaged"? The thing is, I don't trust the vendor. They have every incentive to stretch and lie in their marketing copy, and they do that. And you can put any bullshit into self-selected "real use" metrics like "talk time", "browsing time" or "number of songs / photos that fit on the device", by choosing unrealistic values in some parameters that make the result look favorable. But you can't cheat much on raw GHz, mAh or GB. In the atmosphere of (experience-based) lack of trust, I prefer raw facts and making the necessary ballpark calculations myself, thank you. ("Number of songs / photos / movies that fit on the device" is one of the more annoying/dumber metrics ever. It's as if car companies gave you "number of cities you can round-trip through on full tank" instead of MPG or l/km values.) Or, if the vendor insists on "real use" metrics, they should at least provide the math behind it, so that I know they're not pulled out of the asses of people in marketing department. The core problem is, still, that I have experience-based (and market-understanding-based) bias against trusting what a vendor says at face value. -- RE washing machines, don't confuse engineering numbers on the interface with use cases (the "technical specs" of a product) and the engineering numbers for internals. I don't give a damn what kind of motor you put inside, much like I don't give a damn what particular CPU you put in a mobile phone. The motor turns, and apparently has enough power and torque to drive a loaded machine to stated speeds; all I need to know now is how much power it (or rather, the whole machine) eats and how noisy it is. With phones, the CPU is ARM and runs Android; there's little user-facing difference between various SoC models, so all I care is how much cores it has and how fast are they. (And yeah, I actually might be interested in the materials used in the internals of my washing machine, as a proxy for its reliability; however, given how hard it is to find reliable white goods these days, I just don't bother - I assume they're made out of costs saving and super glue, and plan accordingly.)
- bryanrasmussen 9y agofor washing machines - how much is max load is also needed for ballparking.
- devish 9y ago>>'m curious, how do you know that you need 1) 3500mah battery or 2) 2.1Ghz processor or 3) 4GB Ram on a smartphone?... Your way of comparison is wrong, most of the user neither need a phd in electric or computer science or are going to compare that way. The way is simple 3500mah> 3000mah 2.1 Ghz<2.6 Ghz 4 GB Ram < 8 GB Ram See.
- mrtksn 9y agoBut why? What benefit this brings to the user? Practice their pre-school math? Why just don't write random numbers on the box so the users can compare these? Write, for example, 400 capacitors and let users compare the number of capacitors on the circuit board. Let them brag that their phone have more capacitors. Maybe write the number of diodes, so Huwei Hr67 users can enjoy the wonders of diode dominance over Mi 8jj that has 20 diodes less. ...or maybe you forgot the /s :)
- TeMPOraL 9y agoClock speed, number of cores, amount of RAM, battery capacity, etc. are not just "random numbers". They're very descriptive in real life, for devices grouped by similar purpose. They really are good enough they can be meaningfully used to compare mobile phones between all vendors using nothing but pre-school math, and they also have two additional benefits: 1/ They're more "natural" - which means it's easy for everyone to give those numbers, giving you something to compare in the first place. With "real-use" measurements, one vendor will give you "video time", the other will give you "browsing time", and you can't compare them. 2/ They're harder to fudge. Amount of RAM is amount of RAM, it does not depend on what testing procedure a vendor uses. "Video time" is not directly comparable between brands.
- lmm 9y ago> But why? What benefit this brings to the user? Practice their pre-school math? Because these numbers are harder to fake. Suppose phone A advertises "8 hours of typical use" and a 4Ah battery. Phone B advertises "10 hours of typical use" and a 3Ah battery. Which do you think would actually last longer in use?
- skate22 9y agoI was at the Sprint store a few months ago and wanted a cheap phone that had a gyroscope. I couldnt find it listed anywhere on the phone kiosk, so i asked the sales rep. He didnt know so he asked his manager who also didnt know. I had to go home and google it.