We’ve come really far not only in terms of technology but also prices associated with Google’s smartphones – from the Nexus series to the Pixel lineup. This year’s Pixel 3 XL certainly qualifies for your attention if you’re looking for a high-end smartphone in terms of specifications and price. But with the plethora of phones offering the same specifications for less than half of the price of the Pixel 3, you are also right to question what Google’s flagship smartphones offers that others don’t.

Beneath that elegant personality, the industrious processor takes care of all the hard work. As we saw in our full review, the Pixel 3 XL scores impressively on benchmarks although initially, I had concerns about the 4GB RAM. Despite those concerns, the smartphone has held up superbly in AnTuTu and Geekbench scores.

Comparing it with the direct competitors which come with almost equally compelling hardware specifications, including the iPhone XS, Galaxy Note 9, Huawei P20 Pro, and OnePlus 6 (8GB), these are the results we have obtained from the mentioned benchmarking apps:

Out of these, the OnePlus 6 is the only device with a Snapdragon 845. As you can see, the Pixel 3 XL scores better than the entire bunch of Android devices in AnTuTu but when it comes to Geekbench, the Google flagship can be seen lurking lower on the run, with bad scores when compared to other phones.

Gaming

Given the high-grade hardware, the Pixel 3 XL performs admirably well. The Adreno 630 GPU aboard the Snapdragon 845 lets you play PUBG at Ultra graphics settings with no observable lag. The smooth gameplay without any noticeable heating issues on the Pixel 3 XL has allowed me to earn chicken dinners without stutters. While Asphalt 9 is not supported yet, I played Asphalt 8 without any hassle.

Fortnite for Android arrived well in time, too, while I was reviewing the Pixel 3 XL and its performance has not been at par with other games. The graphics have been terrible, the gameplay uneven and inconsistent, while there have been innumerable frame drops as well as instances of lag. Since we’ve seen a similar performance on many other devices, the issues do not seem to be device-dependent. It would help if you lower the settings to Medium for this game.

Before we compare different devices housing the same Snapdragon 845 chipset, it is vital to take a look at the different configurations put to test. For this comparison, we’re using Pixel 3 XL, OnePlus 6, and Poco F1 – all of which are in different price segments. This will give us a fair idea about whether companies take the extra pain of fine-tuning the performance of the hardware.

It is also noteworthy that despite the same Kryo 385 CPU cores, all three smartphones have different clock speeds. Take a look:

After running multiple instances of PassMark, the following are the scores obtained by the three devices. I’ll explain the logic behind the different tests parts of the benchmark at the end of this article. For your information, the higher the numbers per test, the better the performance – except in the case of “Memory Latency” for which the lowest value relates to the best performance.

Note: See our explainer below to know what each of these tests mean

OnePlus 6 Outpaces Poco F1 and Pixel 3 XL

Taking a close look at the RAM specifications used in the three devices, we know all of them have the same LPDDR4X tech, with the same frequency i.e 1,866 MHz. The difference in performance is due to the varying levels of optimization from the three manufacturers, and the memory management kernels they use.

PassMark Tests

Here is what the different tests in the PassMark score mean:

  • Integer Math – This test determines how fast any device can perform arithmetic functions (add, subtract, multiply, and divide) with integers and is measured in MOps/s.

  • Floating Point Math – The test measures the rate at which a device can do perform mathematical operations on floating point numbers i.e. fractions or numbers with decimal points. The results of this test are also measured in MOps/s.

  • Find Prime Number – The test runs an applet which identifies numbers that are not divisible by any other number but themselves and calculates the rate at which a device can run this test – and find prime numbers per unit time.

  • Random String Sorting – In this test, a device’s speed of sorting 500,000 different combinations of words and numbers with 25 characters each is tested, and the result is measured in million strings per second.

  • Data Encryption – The test uses common encryption techniques like AES, Salso20, Two Fish etc. to encrypt data and then calculates the time taken by a device to find the unique key to decrypt the data. This is measured in MB/s.

  • Data Compression – This test is used to calculate the megabytes (KB) of data which can be compressed per second without any loss of data using advanced encoding techniques.

  • Single Thread – This test determines the performance of the CPU in the case of tasks that use a single core at once and measured in MOps/s.

  • Database Operation – The test uses standard C++ container templates to verify how well the RAM can preserve these heavy database templates, and measures this in KOps/s.

  • Reach Cached – Determines the rate at which a small block of memory cache is read.

  • Read Uncached – Determines the rate, in MB/s, at which a 256MB block of the memory is read.

  • Memory Write – This test judges the MB/s rate at which information is written to the memory.

  • Memory Latency – This test determines the time in nanoseconds (ns) to transfer one byte of information from the RAM to the CPU. The lesser this value, the better.

  • Memory Threaded – Just like the Read Uncached test, this one determines the time used for reading multiple blocks of memory at the same time. The information is displayed in MB/s.