3 ms·

But in real terms, that might actually not be the case. The OS implementation, use case, and hardware setup will affect all of those things meaning direct compa

But in real terms, that might actually not be the case. The OS implementation, use case, and hardware setup will affect all of those things meaning direct comparisons aren't really all that valid. You can dive down the rabbit hole of comparisons; a 3500mah battery might be bigger than a 3000mah, but if the 2.6Ghz processor demolishes it faster than the 2.1Ghz one, it'll end up worse. This assumption that humans only make rational, logical decisions is inherently flawed. I know HN likes to decry marketing, but humans crave the way these things are framed. Granted that means it's open to manipulation, but the continual argument here is always "give me every single piece of objective information about absolutely everything and I'll make my decision" doesn't extend too far outside of this echo chamber.

TeMPOraL 9y ago
Direct comparisons are valid enough, and are definitely more valid than whatever comparisons you can try and make with manufacturer's cooked numbers like "hours of video time". The goal isn't "perfect", it's "good enough in practice". > Granted that means it's open to manipulation, but the continual argument here is always "give me every single piece of objective information about absolutely everything and I'll make my decision" doesn't extend too far outside of this echo chamber. Again, the argument is that objective information, even if mostly incomprehensible to people, is still better than whatever fake-metrics vendor's marketing department spits out, by the virtue of reality not actively trying to confuse you.