Does this look like eMMC speeds to you?

[D13H4RD]

My mom recently picked up an Honor 8 Pro after I suggested that due to her habit of playing a metric crap-ton of games, the high-power and extremely long battery life of the 8P will  be suitable to her.

 

I'm also doing a review on it. After doing some core benchmarks on it (spoiler alert: It's a beast), I fired up Androbench and noticed an immediate oddity;

The sequential and random read and SQLite Update and Delete performance seem to indicate that it is using eMMC storage. However, the Sequential write plus random write along with SQLite Insert performance exceeds even that of some UFS 2.1-touting devices.

 

I'm at an oddity here. I know retail versions of the Honor 8 Pro use eMMC because it turns out that review samples have UFS (why tho?), but even though the read performance is consistent with devices touting eMMC, the write speed is crazy fast.

 

Here's my Moto Z with UFS 2.0 storage in comparison

What gives? Is this some sort of super eMMC that only Huawei has access to?

[SCHISCHKA]

Your new phone should have a newer version of linux kernel, newer android, and more importantly a newer version of SQLite.

Whenever I see really high speeds it looks like caching.

Try a different benchmark.

[D13H4RD]

4 minutes ago, SCHISCHKA said:

Your new phone should have a newer version of linux kernel, newer android, and more importantly a newer version of SQLite.

Whenever I see really high speeds it looks like caching.

Try a different benchmark.

I might try A1.

 

If the results are the same, then something is at an oddity.

[D13H4RD]

Same story.

 

eMMC speeds for read, >UFS 2.1 speeds for write.