A new HP Elite x3 Windows smartphone might be coming soon

5. fyah_king (unregistered)

The iphone has always done well in single-core. This new geekbench is kinda weird to me. 99% of Android phones scores are down but the iphone went up like crazy!!!lol

8. TechieXP1969 (Posts: 14967; Member since: 25 Sep 2013)

Yes the iPhone is great at things that don't matter. The thinsg that do matter is what its terrible at.

Example, taking pictures. The new iPhone 7 takes worse pictures than the 6S. Stated by several tech sites including PA, Amandtect, Tom's Hardware, CultofMac and others.

The iPhone is fast because it ha slimited tasks and capabilities, yet crashes more than any single Android phone. Even over Samsung's.

That's all I will say.

It is a great phone that I like, but it needs ot be better based on what it costs. I live in a world where a product with limited capabilities costs less than those that do not. Apple is the only company that gets way with being the except.

I dont mind them being an exception, if their products actually ended up being betetr then the competition. The facts show that the phone the iPhone mimic or are considered on the same caliper, are far better.

The Galaxy S has been better then the iPhone at "nearly" everything since the S4.

Speed is a very small metric of how good a phone is. Basically if you took all the heavy weight off a Galaxy, it would blow the doors off the iPhone.

10. kiko007 (Posts: 7337; Member since: 17 Feb 2016)

"Yes the iPhone is great at things that don't matter. The thinsg that do matter is what its terrible at."

So let me get this straight. UI performance, arguably the most important aspect of modern phones, which Apple DESTROYS Android in, doesn't matter. Okay then..... dumbass.

"Example, taking pictures. The new iPhone 7 takes worse pictures than the 6S....IN LOWLIT AREAS. Stated by several tech sites including PA, Amandtect, Tom's Hardware, CultofMac and others"

FTFY

"The iPhone is fast because it ha slimited tasks and capabilities, yet crashes more than any single Android phone. Even over Samsung's."

Um no:http://www.cultofmac.com/428792/android-phones-crash-at-almost-twice-the-rate-of-iphones/

"That's all I will say."

*Continues rambling for four more psuedo-paragraphs*

"It is a great phone that I like....."

Bulls**t. You know damn well you HATE Apple products, otherwise you wouldn't post entire thesis papers regarding how all Apple user "Should kill over already". STFU you idiot.

"I dont mind them being an exception, if their products actually ended up being betetr then the competition."

That's not how being an "exception" works.

"The Galaxy S has been better then the iPhone at "nearly" everything since the S4."

Sales figures seem to disagree with your assessment as usual.

"Speed is a very small metric of how good a phone is. Basically if you took all the heavy weight off a Galaxy, it would blow the doors off the iPhone."

It already BLOWS apparently. And there is nothing "heavy" about the Galaxy's need for processing power.... It's s**tty optimization that's the problem. Unless you're going to now admit that Android is coded in such a terrible fashion that it actually has a garbage compression mechanic to keep it from imploding upon itself. What am I saying.... you'd sooner die than admit that. Oh well....

17. sgodsell (Posts: 6485; Member since: 16 Mar 2013)

plain and simple GeekBench is a piece of crap. It should never be trusted.

18. mikehunta727 (unregistered)

It's actually the most reliable cross plat CPU benchmark we can use at the moment. Many high level engineers can attest to GB4 being accurate

Geekbench 3 wasn't as accurate and didn't support all CPU features

21. sgodsell (Posts: 6485; Member since: 16 Mar 2013)

I don't know what bs your pulling out of you pants, but I have some native code that runs on a SD 820, and it runs almost 2 times as fast as an iPhone 7. So please feed your BS to someone else that believe that crap.

7. TechieXP1969 (Posts: 14967; Member since: 25 Sep 2013)

Multi-core is all that matters. No OS or applications uses a single core...thus the scoring is moot.

But benchmarks are BS, inconsistent and have a small bearing on actual phone usage.

9. kiko007 (Posts: 7337; Member since: 17 Feb 2016)

"Multi-core is all that matters. No OS or applications uses a single core...thus the scoring is moot"

Are you sure you're an IT expert? Because that statement alone contradicts basic computing knowledge. The majority of ALL applications operate in a Single-Thread environment, with the exception of strenuous tasks. Placing meaningless tasks on multiple thread unnecessary amounts of power, and is generally frowned upon.

"But benchmarks are BS, inconsistent and have a small bearing on actual phone usage."

Geekbench is far from "BS" as it operates independent of OS, meaning it gives an accurate description of CPU power. And while I sorta agree with your second point, the fact that this was all QC could muster chip design wise is embarrassed.

13. mikehunta727 (unregistered)

Techie is talking out of his ass and has no clue what he is talking about.

Single core performance directly correlates to multi core performance, but he doesn't even get this. LMAO. Much easier for developers on iOS/Android to have their app max out 2 cores then 8 cores, apps on Android nor iOS will never max out 8 cores fully at 100% load besides benchmarking results, but apps without much coding and heavy parralelization can take advantage of fewer cores.

So basically, the A10 is able to hit its maximum peak performance a lot easier and better then a 8 core system does, much easier to saturate 2 cores to near 100% load then it is 8 cores. Most apps don't scale well beyond a few cores and you can't break down every single task for 8 cores to work on. 2 A10 cores at both 75% load is hitting higher performance then 8 cores all with load of 40-60% usage

In a world where only the most heaviest applications on Windows make use of all cores efficiently, how does one possibly think where development costs and manpower and experience are generally much lower on Android(most apps only coded by a few devs, not even)

14. mikehunta727 (unregistered)

If single core performance doesn't matter at all like how Techie is saying, we would be on 16 core A53 SoCs already but that isn't the case. OEM's like Qualcomm, Samsung, etc all very well know the aspects of better IPC in their SoC's, which is why the 835 is going to again be a 2+2 setup but with better IPC. Samsung is going with custom cores 4 powerful cores +4 weaker power efficient cores setup in their new Exynos that should step single core performance up quite a bit

Better IPC allows apps to run better without it needing better optimization through brute force, i.e if single core performance is great, it can brute force most of the code

It's no wonder why on most govenors on Android devices that control CPU usage and core usage in X scenario, the main powerful cores are generally always used while the more power efficient cores are on idle most of the time and handle light tasks, only in intense computing scenarios are they all active as well, A53's are considerably weaker then their A57/73 counterparts though

15. mikehunta727 (unregistered)

Take for example 8 core A53 Mediatek SoC's, they have considerably worse performance in many applications vs a SoC with less cores but stronger IPC/single core performance(takes longer to load app data, slower to load websites, etc), it is because A53 IPC is quite bad and is not meant for heavy tasks/intense compute scenarios compared to stronger more high performance cores.

Also devices like LG with their skins, have almost just as much "bloat" as their Galaxy counterparts yet LG devices perform noticeably better in real world vs their Samsung counterpart, Samsung's optimization is just not up to snuff yet and thats facts. XDA's article from last year on Note 7 S820 performance proves this also

Techie literally trying to make the argument that "who cares about better performance. who cares if its faster" lol then go back to your Galaxy S1 with Hummingbird SoC then if you don't care about better, faster performance

16. mikehunta727 (unregistered)

Think of it this way in most simple laymen terms, a single wheel barrel that can move 10 pounds of product delivered in one go is more efficient then having to have 5 single wheel barrels that can each only move 2 pounds of product each delivered,(pounds in this case being single core performance/IPC and the barrels each being a core each)

More cores almost never scales linearly in application performance so remember that too. After a certain point, more cores don't add linear performance gains

20. rocker91 (Posts: 86; Member since: 09 Dec 2016)

Yep my company also use Windows phone other than apps personally I love the platform over android.