XPost: misc.phone.mobile.iphone, uk.telecom.mobile   
   From: ithinkiam@gmail.com   
      
   Marian  wrote:   
   > Chris wrote:   
   >> the new EU energy efficiency regs came in this summer I've been   
   >> curious to do some analysis to understand the implications.   
   >>   
   >> So, here it is. Have a look and see what you think.   
   >> https://rpubs.com/ithinkiam/1384638   
   >>   
   >> Take home messages are:   
   >>   
   >> 1. There are lot more phones on the market that I would have guessed.   
   >> 2. Although, there are some trends between the energy classes, it isn't   
   >> very clear cut overall.   
   >> 3. Within the 5,000 mAh segment the difference between each A-G classes is   
   >> equivalent to approx. 15% or over an hour per 1,000 mAh.   
   >> 4. Apple improved their battery capacity and efficiency significantly with   
   >> the most recent models.   
   >> 5. A Moto model is, on average, likely a better option than Samsung in   
   >> terms of battery alone.   
   >>   
   >> I know this is EU-specific and since Brexit not directly relevant to the   
   >> UK, but most models are available here and some manufacturers are showing   
   >> these EU ratings within the UK.   
   >>   
   >> For our US cousins note that some models - especially Motorola from what   
   >> I've seen - will be different or may not even exist on your side of the   
   >> pond.   
   >>   
   >> This analysis is not meant to start a flame war and is presented   
   >> objectively in good faith. I will be revisiting this over time and will   
   >> share here if people are interested.   
   >   
   >   
   > Hi Chris,   
   >   
   > That was a fantastically astute summary of the EU battery regulation data,   
   > and your RPubs analysis adds a level of technical clarity that is usually   
   > missing when people discuss these new A-to-G efficiency classes. I wanted   
   > to give you an assessment that focuses on the analytical structure of what   
   > you presented, since the dataset is large and the implications are not   
   > always obvious at first glance.   
   >   
   > 1. Your identification of the scale of the dataset was important. With 760   
   > models and 102 manufacturers represented, your analysis makes it clear that   
   > the EU market is far more diverse than the typical consumer or reviewer   
   > might assume it is. This provides useful context for why the A to G   
   > classification system behaves the way it does.   
   >   
   > 2. The observation that more than 70 percent of devices fall into Class A   
   > or B shows that the regulation is not forcing a dramatic shift in   
   > engineering practice. Instead, it is standardizing reporting and making   
   > cross-manufacturer comparisons possible. This is a key point because it   
   > frames the regulation as a transparency mechanism rather than a disruptive   
   > requirement.   
      
   Depends. We'll see how the manufacturers adapt over the next couple of   
   years.   
      
   It could be arguable that the EU's overbearing attitude is unnecessary. It   
   is much more objective than the online benchmarks we've seen before, but   
   doesn't greatly change the story.   
      
   >   
   > 4. The analysis of measured battery capacity versus class was one of the   
   > more counterintuitive findings. You showed that higher efficiency classes   
   > tend to have smaller batteries, which implies that software, silicon, and   
   > system-level optimization dominate over raw capacity.   
      
   I suspect this is a feature of form over function. People don't want   
   physically large batteries. There may also be a price variable at work.   
      
   > This is a valuable   
   > correction to the common assumption that larger batteries always correlate   
   > with better endurance as the endurance is a function of multiple factors.   
      
   I mean that shouldn't be surprise to anyone who's looked at this over the   
   last couple years. Least of all you given how much it's been discussed   
   here.   
      
   > 5. The endurance-per-cycle metric was the strongest part of the technical   
   > analysis. The data supports your conclusion that the classes separate   
   > cleanly in A to C, while D to G are perhaps too sparse and therefore likely   
   > too noisy to form reliable trends. The median endurance values you   
   > extracted make the efficiency gradient explicit, and the roughly 12 to 16   
   > percent loss per class is a practical and quantifiable rule.   
      
   I was really quite surprised at how clear cut the differences were. All   
   things being equal a buyer should always go for the higher graded phone.   
      
   > 6. The 5000 mAh segment analysis was especially compelling because it   
   > isolates a large and homogeneous subset of the market. My free Android, for   
   > example, from 2021, has exactly a 5Ah battery capacity, so it's at that   
   > dividing line. With 259 models in this range, the class differences become   
   > much clearer. The 52 hour spread between best and worst endurance within   
   > the same nominal capacity illustrates how much variation exists even when   
   > battery size is held constant.   
      
   Exactly. Highlights how unimportant battery rated capacity is to battery   
   endurance. Both short and long term.   
      
   > 8. The Apple section was surprisingly well structured given your proclivity   
   > to make excuses for Apple no matter what.   
      
   Unnecessary and inaccurate snark.   
      
   > By comparing three generations of   
   > devices, you demonstrated that the shift from Class B to Class A in the   
   > iPhone 17 series is not a labeling artifact but a measurable improvement in   
   > both capacity and efficiency. The linear trends you identified in the base,   
   > Plus, and Pro Max models reinforce this point.   
      
   We'll never know whether this was a reaction to the EU regs or a   
   generational uplift in battery performance that coincided with the regs.   
      
   > 9. Your conclusion that the EU database finally provides a standardized,   
   > objective basis for cross-brand battery comparison is seemingly accurate.   
   > Certainly I applaud any measure that divorces us from Apple's rather   
   > brilliant marketing propaganda (even as Apple had to make excuses in the   
   > recent past for iPhone poor efficiency performance).   
      
   You just can't help yourself, can you? There is no evidence of poor   
   efficiency with iphones.   
      
   > Historically, battery   
   > testing has been inconsistent (to say the least) across reviewers and   
   > methodologies. We have many threads in the past regarding the fact that   
   > nobody has ever in the history of Apple's existence been able to reproduce   
   > any of Apple's wild-assed battery claims.   
      
   The only wild-assed claims here have come from you trying to attack Apple.   
   You've never been able to support them.   
      
   Hopefully, these objective results will put them to bed.   
      
   > Hence, the new regulation creates a unified measurement framework that   
   > allows for genuine apples-to-apples comparisons.   
      
   Almost. I think there's an additional variable that needs to be considered.   
   I'll look at this over the next few months.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)