shinjitsu, 06 May 2016Don't forget camera usage and video. Write performance allows the video to keep up the bitrate... moreWell you have a point but if the HTC 10 does not differ in terms of camera quality, I don't see the advantage as well. From what we know s7, g5 or the HTC 10 all have comparable quality in camera.
Anonymous, 07 May 2016It is already a turn off when I learnt that HTC 10 is using emmc. It is pretty much a junk bec... moreHere Samsung S7 and S7 Edge each costs approx $650 and $750 respectively.
While HTC 10 has yet to come up with an official price.
What feature does the HTC 10 have over the other two? Firstly, different people different needs. I appreciate your preference for bigger battery and storage speed (S7 Edge) and you could as well mention the IP68 for good measure.
OTOH, people who value audio more, for instance, would opt for HTC 10, or G5 but with extra B&O module.
There are also those who avoid glass back, thus prefering the 10 or G5.
And the iPhones memory speed is the fastest. Why you didn't test that?
Possible they chose this because it would work better with adoptable storage of SD card?
A Quality Junk for llot of money... totally waste G5 is far better option...
LLL, 07 May 2016Thanks for digging deeper and coming up with this fact. I smell something fishy here and this... moreIt is already a turn off when I learnt that HTC 10 is using emmc. It is pretty much a junk because for its price, it is not worth it. I not sure about your area but at my country, currently the HTC 10 cost a bit more than even the S7 Edge 32gb and way more than the LG G5. What does HTC 10 has over the two 2 phones? What does S7 Edge has over HTC 10? Faster performance in terms of storage speed and CPU speed if you have the Exynos, bigger battery and better screen (I love AMOLED but it is subjective). LG G5 has also better price, better storage speed and modular feature. There is just really no reason to get the HTC 10.
swirly, 07 May 2016http://www.anandtech.com/show/10252/htc-10-battery-storage-results
As you can see, the resu... moreThanks for digging deeper and coming up with this fact.
I smell something fishy here and this has already fueled HTC haters' agenda to trash the hell outta the 10 even more.
The 10 might not top all of performance charts but dismissing it as junk is downright ridiculous.
Couple of discrepancies in the article - copying photos from phone to PC is a read operation. hTC & LG support USB 3.0 whereas the GS7 supports the slower USB 2.0. Clearly the bottleneck is the USB interface. Dont understand how the user experience difference highlighted can be attributed to this. Are you sure there's no setup issue? Androbench results (in your pic) shows much higher SQLite results (3x to 20X) for hTC vs. the other two, any comment on these beyond seq/rand R/W? I'd imagine this will have an impact on UX.
Anonymous, 07 May 2016Assuming what you say is true (I'll wait for you to provide the source to back what you say), ... morehttp://www.anandtech.com/show/...rage-results
As you can see, the results there look a bit more sunny. Let me quote the author of the article:
"By default, AndroBench uses 8 IO threads for all of its tests. This behavior showed up with AndroBench 4 and continues in AndroBench 4.1.
eMMC is half-duplex, and designed for single-threaded IO tasks. It's not the greatest system, but it is the most common storage in use in Android phones, so applications are going to be designed for eMMC storage instead of the 5 or so phones that are shipping with UFS storage. Multi-threaded IO actually can negatively affect storage performance with eMMC because of resource contention issues, so in general it's rare to see multi-threaded IO in real apps."
Will this change the fact that the 10 is the slowest of the three? No. But Ricky shouldn't be misleading the readers. The 10 is nowhere near that slow.
swirly, 06 May 2016They're using AndroBench's default settings other than changing sequential to 256KB. The defau... moreAssuming what you say is true (I'll wait for you to provide the source to back what you say), they should change the settings so it gets better results? Why? That won't change the fact that the actual performance of file transfers takes longer on the HTC device and will result in benchmark results that are not in alignment with actual performance.
They're using AndroBench's default settings other than changing sequential to 256KB. The default settings are designed to give a huge advantage to UFS in ways that real apps generally do not.
No need to discuss. This is just one more nail in the coffin of unfortunate and late flagship of 2015.. this phone has nothing to do in 2016. Not only it is coming late to New year party, but as well dressed for Halloween ... Maybe next year HTC... or maybe newer (because you will be in same boat with Blackberry by then).
Anonymous, 05 May 2016"The time that it took to transfer sample photos from HTC 10 to our PC definitely felt much sl... moreThey only commented on the transfers from photos from phone out to PC (read), and not the other way around (write). Either way, people seem to ignore the final takeaway to pursue their own agenda, which is "We wouldnÂ’t want our readers to get the wrong message. The HTC 10 is still blazing fast in every day performance. UI, launching apps, and such are all still exceedingly snappy for day to day."
mitchytan92, 06 May 2016Yeah and to be exact I think random read is the most important than seq read. Write is only us... moreWrite performance is important for camera usage and video recording, not just for installing apps.
Lol, 05 May 2016Only when you are installing apps. When you are actually using the phone, read speeds matter f... moreDon't forget camera usage and video. Write performance allows the video to keep up the bitrate, and cameras to do burst shooting.