What protection glass. What gorilla glass used? Why this information is not given by the manufacturer / distributor?
When it comes to comparing devices on batteries and power GSMArena needs to compare stuff differently.
Most of people are interested in 1 category. Like if I buy Note series this tells that I careless for smaller screens or non QHD screens. And That Im interested in high end devices.
So in batteries (and we all know that screens play the biggest role in consumption) I would like to see big screens and QHD compared to each other. HD big screens. High end with smaller screens together. The same spplies to power. What are high end devices doing with mid rangers.
At least you can have a bottun to press which compares them together, and thise who want to see a full list can.
How many are really interested in all and every kind of cell phone???
Dont take benchmark in your life.. Just buy phone which you want to buy..
Anonymous, 22 May 2016Take Snapdragon vs Exynos in Samsung S7. The review could make you think that they are almost ... moreAFAIK, the SoC has little to nothing to do with RAM management - that's all on the firmware.
That aside, I say little benefit because most devices these days are already good enough at cold launch times and caching that any differences between them would be imperceptible unless held side by side. If there's a significant difference, enough to be noticeable as slower than the standard, then the reviewer would notice, and mention it in the review. Finding out what the difference is would not be worth the effort, at least not for me. Why? Let me start with benchmarks - nobody considers them to be any proper indication of real world performance/UX, which means that one test, under one set of conditions, is enough. However, if you're going to try and represent real world UX in precise figures, it can't be shallow; to live up to the claim, you have to do tests (like the one you suggested) in a variety of conditions: high and low battery levels, low, middling, and high temperature levels, and weekly readings of the same (so we know how it fares with varying states of storage usage). And that's just the things I would want measured. Yes, this is a lot of detail to go into, but if you're going to make an attempt to quantify real world experience, this is the minimum that I feel must be done. This would take time and effort, and I personally don't believe the tradeoff would be worth it, if the delay is too significant. If they were to add this after posting the main review, however, so that the main bits are made available as quickly as possible...
Anonymous, 23 May 2016why on earth would mi5's benchmarks be so much higher than other sd820 devices? its simple act... moreThey should have been investigated for weekzzz now and investigation would have headlined for dayzzz now if what you claim is true
why on earth would mi5's benchmarks be so much higher than other sd820 devices? its simple actually: xiaomi is a cheater. they do this to get customers who just base their smartphone buying guide to benchmarks
Anonymous, 22 May 2016where is the htc 10 ? who cross 150k in antutu...Bankrupted already - all 2 devices were sold and factory now closed.
ssbatman21, 22 May 2016Ya i knw, its called fanboy syndrome..... U shud try applying some logic on it....My opinion purely based on the benchmark. No fanboy here. Not even using LG. Stop making unnecessary assumption as you like. If I were to make such assumption, I assumed you are a diehard Zopo fanboy.
Anonymous, 22 May 2016where is the htc 10 ? who cross 150k in antutu...2 things;
1 - this is not a comprehensive list of all smartphones, some are added, some are missed. Simple. So stop asking where the HTC 10 is.
2 - I'd personally like to see a specific anti-cheat version of the benchmarks as HTC have a long history of benchmark manipulation. Seeing it score so much higher than the competition on the same hardware has to raise more eyebrows than just mine. Especially since Sense is far from the best skin out there - for performance purposes that is, for a user I can't really say.
Ill chip in here for an objective observation...ive had the exynos S7 edge, the Z5 premium and Z5 normal, the Nexus 6P and the Samsung A5 2016, i now also have the LG G5...the performance of the G5 after the update last week is simply miles better in day to day use than the others, The s7 edge is great but has issues with google now and power usage...dont use now and its fine...camera is not as good, battery is a little better, not much, screen is brighter, but lg g5 is improved now , fingerprint reader on the G5 is miles better than the S7 and the A5...the LG G5 is the fastest and smoothest of all of them, even better than my 6S plus on IOS
Julliard, 22 May 2016What I learnt from the benchmark? I learn that we should buy LG G5 instead. The rest of the li... moreYa i knw, its called fanboy syndrome.....
U shud try applying some logic on it....
As mentioned by some of the other commenters; peak performance is rarely used and shouldn't be used as a deciding factor when looking for a phone. It has no real validity on how the phone will work or run apps as - as mentioned below - the type/age/size/bug count all matter and some will conflict on, say, an LG but not on a Sony.
What we can take from this post is that Mediatek have released a chipset - the Speed 8 did come on the market in February by the way - that will deal with pretty much everyone's day-to-day activities - on paper. From reviews of the phone on more obscure phone review sites their seems to be no lagging issues or slow down even under heavy strain.
It's hard to get, most people have never heard of it and fewer will see one but if you have the chance/option/patience(to import) it'd match most of the bigger names punch for punch.
SirVer, 22 May 2016Valid point, but measuring what you propose won't work either, as app launch times and number ... moreTake Snapdragon vs Exynos in Samsung S7. The review could make you think that they are almost equal. Actual hands on with both devices will leave you with the clear sense that the Exynos version is faster (better IO performance, better ram management).
IMO that's the kind of metric can give you how the phones actually feel. Of course benchmarks should also happen mostly to measure the few occasion where peak performance is needed. They also give you a good sense of which platform is better for high end gaming.
The current tests however take too much time and give little actual information. So if they have to choose between measuring peak performance and real performance they should always choose real performance. I don't see how it gives "little benefit", it's the thing that matters to most users before bying a phone. What's the need of a review if it can't tell you which phone is actually faster. I literally buy phones based on reviews. So it's of great benefit IMO.
As for the test, it's very easy. Code a monkey runner (Google gives you the tools) that will open set apps (say chrome, Gmail, youtube, maps, Facebook, Twitter, temple run, and maybe another game). It would open them once while measuring using a set timer. That would measure launch times. And then do it again to measure "memory retention".
Many youtube channels already do that and actually show which phone is faster. Obviously both phones would be factory reset and freshly rebooted before running the monkey runner script.
It's not a hard test to run and make it reliable. Reviewers should finally stop treating phones as if they are PCs, we are clearly using them differently...
Anonymous, 22 May 2016Please stop doing that. Peak performance matters very little on mobile devices and that's all ... moreValid point, but measuring what you propose won't work either, as app launch times and number of cached apps vary with load, age, and a bunch of other factors; to get reasonable averages would take in-depth testing over an extended period of time, which is out of the scope of blog posts like this; these benchmarks simply serve as a teaser of sorts, before the actual review comes in, not as an accurate representation of performance. The testing you propose is something they could add to their main review, certainly, but that would require them to come up with a standard process for it, like they have for everything else, and perform it on every device, adding one more delay to the review process. If any particular thing is slower or does not work up to usual expectations, they do mention that in their reviews, going into detail if necessary. It's something that would add a whole lot of work, for not very much gain, IMO.
What I learnt from the benchmark? I learn that we should buy LG G5 instead. The rest of the list is horrible phone. Btw, I never heard of Zopo.
Anonymous, 22 May 2016Ok for a midrange chipset, but mediatek should consider Building high End chipsets. It is there high end chipset...
What a stupid remark... If you come 5th in a race, no one will tell you: "You should CONSIDER to come first next time" You cant just think to make thr beat chipset and make it just by thonking it...