The OnePlus 6 Review: Among The Best Of 2018
by Andrei Frumusanu on July 27, 2018 8:30 AM EST- Posted in
- Mobile
- Smartphones
- OnePlus
- OnePlus 6
Video Recording
Video recording on the OnePlus 6 is done on 720P, 1080P 30 or 60fps, and 4K 30 or 60fps. The camera has OIS in all modes, but on top of that uses EIS in 720P and the other 30fps modes. Unfortunately there’s no software switch for the EIS so you’re getting what OnePlus is giving you- which comes at a cost of reduced field-of-view.
EIS stabilisation works extremely well and results in a very stable and detailed video. Unfortunately because you can’t turn it off you can’t independently compare the EIS in the same capture mode to see how the detail retention works.
One thing to note was that the 1080p60 footage was extremely disappointing as it was very blurry. Looking at the encoding the 1080p30 is done on AVC on High Profile 4.0 at 20Mbps while the 1080p60 footage is encoded at High 4.2 at 40Mbps, so there must be something wrong in terms of the ISP part of the video pipeline that results in the much reduced image quality. It’s also notable that I found it to focus slower in 60fps mode than in the 30fps modes.
The 4K footage doesn’t suffer from the same degradation when going to the higher frame-rate. Here in both modes there’s great amount of detail retention. The 4K30 footage is encoded in a High@5.1 profile at 42Mbps, while the 4K60 footage is encoded at High@5.2 at a dazzling 120Mbps which is actually the limit of the Snapdragon 845’s video encoder. Here the sample video of 1m29s length came in at a hefty 1.24GB. While I applaud OnePlus for the high quality encode settings, the fact that the device lacks HEVC encoding options means that you’ll have to really be careful in terms of storage management when using 4K60 footage.
Audio recording in the video was very good.
Speaker Evaluation
I’ve first introduced speaker evaluation back in our Xiaomi Mi MIX 2S review a couple of weeks ago and we’re going to apply the same methodology to the OnePlus 6.
We’re using a calibrated flat response measurement microphone for the measurements and are using REW as the supporting software suite.
For speaker loudness measurements, we’re using a pink noise source, with the measurement microphone at head level 40cm away from the phone. We measure once the perceived volume when the phone is held in portrait one-handed mode, and once while holding the phone in landscape mode with both hands, cupping the sides.
One-handed and in portrait mode, the phone gets reasonably loud at up to 77.6dB(A). Holding the phone two-handed and cupping it however results in redirecting a lot of the sound towards the listener again, attaining a very loud 86.3db(A). In an optimally designed phone, we actually don’t want this much of a difference as it means a lot of the audio is firing away from the phone. The OnePlus 6 is one of the less optimal phones in terms of its speaker design as the sound pressure is very directional and bottom-firing out of its main and only speaker. This means you’ll have to pay more attention on how you hold the phone as it can make a large difference to volume and audio quality, and also it’s very easy to muffle the sound when the speaker holes are covered up.
We use a logarithmic measurement sweep within REW to measure the frequency response of the phone’s speaker. To get a more accurate comparison between phones, the volume is calibrated via pink noise to 75dB(A) and the measurement is done in landscape two-handed mode. The graph is represented with a psychoacoustic smoothing filter for better representation.
Unfortunately the OnePlus 6’s frequency response represents what we can also clearly hear when playing back media: It lacks depth as well as clarity. In the low frequency and lower mid-ranges ranges the OP6 fared among the worst, and this results in a very hollow-ish sound.
The OP6 is actually the loudest in the mid-ranges in its sound profile.
While a lot of mobile phones have an evident peak in the treble frequencies, the OP6 actually remains quite flat, which will affect sounds such as ringtones and represent them quieter than other phones.
Finally, the OP6 also doesn’t fare well in the high frequencies and has a very steep drop-off after 12KHz. Here it’ll depend on the user’s age and his sound profile, but most will still be able to hear these frequencies and interpret them as the “brilliance” in music.
Overall, the OP6’s speaker is average at best and can’t really compete with some other flagship phones.
Facebook
Twitter
RSS
90 Comments
View All Comments
Total Meltdowner - Friday, July 27, 2018 - link
Art Photography? I'm not an unemployed liberal. I like phones to snap photos of family and friends doing things worth remembering.Cooe - Saturday, July 28, 2018 - link
Jesus Christ...Quantumz0d - Friday, July 27, 2018 - link
Starting with the wireless charging changing ergonomics and thickness, I refute. Look up at LG V30 - 7.3mm zero bump no notch. Has Qnovo on top of it, looks better made better with MIL spec. IP68 certified as well. OnePlus failed at Metal design with signal flaw with OP3/T and they cheaped out with glass just for marketing this time. About time when they ditch the jack. OPPO started that with their feeble built Find X.Too soft on Oneplus and other companies notch "there's really nothing controversial" its downright copied from Apple except that keeping the bottom bezel for no reason except to make it a marketing trash. Blacking it out for what ? Breaking the Android apps, but like all mainstream brainwashers Google also decided to add it with utter shameless act. Seems like even at AT no one cares about originality or engineering anymore. Shame
No mention of absence of Video out, Netflix certification. Same battery capacity since OP3T. And Axon 7 from 2016 is able to sell a 1440P AMOLED panel at same/less cost with more features like Stereo front facing. AKM DAC wgich works with Lineage. While OP cuts corners at all specs just increase the RAM and Memory that's it. Pixel 2XL shreds OP in speed with low RAM, not against it but this phone is hot garbage.
Oh this phone doesn't have USB3.0 either. CEO says no one uses when we have LTE pure BS. Why are they incl the high capacity UFS then ? Rubbish selfless corp. After being burned at the Op3 promised blobs and the worst part of this phone not mentioned, the T variant. Which makes the Op6 users look like fools, why don't this get mentioned anywhere ? Is OP paying that hefty ? Guess so since they skimp so much and price increase..its a disgrace.
Sultanxda left OP because they never bother to fix the camera or treat it as some high profile golden IP, or the 821 3T high clock crash. Dash charging, another proprietary technology, no way you can find replacement parts for this phone because 5-6Month EOL is extreme ripping.
dshess - Friday, July 27, 2018 - link
I wish one of the second-tier vendors would commit to long-term hardware support. At $600 (or $800), the story is when they manage to make a phone that sucks - but making an excellent phone for $250 is a great story, especially if you can still purchase the same device a few years later. It's like nobody is even trying to take on the iPhone.icalic - Friday, July 27, 2018 - link
Hi @Andrei Frumusanu,why no more gfxbench manhattan es 3.1 / metal test for battery life and final frame rate @ 200 nits?
i think that test is good for us to look GPU efficiency on every device.
for oneplus 6, i suspect high sustained gfxbench followed by higher power consumption than other snapdragon 845 devices.
Andrei Frumusanu - Friday, July 27, 2018 - link
Power consumption between devices doesn't really change. In this regard I'm keeping the power measurements to the initial SoC reviews of the first devices. The power I measured on the OP6 was not much different than on the S9+ which is covered here: https://www.anandtech.com/show/12520/the-galaxy-s9...The final frame rate benchmark has been superseded by the sustained performance measurements across all 3D workloads, so there's no need for a separate metric.
As for battery measurement in 3D workloads, I think the GFXBench test as an actual *battery* test didn't really convey a lot other than a maximum rundown test. I think having a phone loop in an actual commonly released game would be better representative. As a reminder, the SoC's power consumption will differ greatly at different frequencies and real games will be Vsync capped at 60fps, so in actual use-cases the MH3.1 battery test didn't really show a representation of such use-cases. Currently I'm still thinking about a way to do this better and hopefully the methodology will evolve, but for now it's as it is due to practical reasons.
arvindgr - Friday, July 27, 2018 - link
In OnePlus 3 review, AT included note on USB, that they're still using USBv2! But 4 generations later OP6 still packs that same USBv2 tech. Also with Oppo proprietary charging tech, they are least bothered about USB-PD standards which its SoC supports via QCv4. Why such tech implementations weren't discussed at all..?Andrei Frumusanu - Friday, July 27, 2018 - link
USB 3.0 has issue with creating radio frequency interference in the 2.4GHz band which requires vendors to do a lot more engineering in terms of shielding, that's why a *lot* of phones don't support it.In regards to fast charging techniques - first of all ever since I came back to AT I haven't had the opportunity to re-test all the devices over the last year that we were missing data on. Secondly I didn't prioritise this as much because I feel after 5V/2A/10W which essentially every device nowadays supports, faster charging rates become diminishingly relevant.
Batteries are supposed to be charged at a maximum of 1C, for the OP6 that's for example 3.3A since the capacity is 3.3Ah. A 10W charger is already 2.38A, going to 15W is getting near the limit. Devices nowadays also have like 10 different charging profiles depending on what power supply you connect to it, so while true we can resume testing the stock charger, but again in reality how representative is that really for most people, especially among AT readers?
Quantumz0d - Friday, July 27, 2018 - link
Appreciate the response for this. I always wonder what's the reason behind cutting costs for USB spec. And about the Dash charge, it's proprietary you are stuck with using the accessories from OP site only and usable with their HW only. Its not about speed its about adopting standards like USB PD or QC.Andrei Frumusanu - Friday, July 27, 2018 - link
> Its not about speed its about adopting standards like USB PD or QC.I think that's a weird attitude to take. QC or PD are just ways to achieve speed. A 10W 2A non-standard charger will charge the same as a 10W QC, PD, Dash or whatever charger. Electrons don't care about the standard.