Review: Unitek USB3 memory card reader

Review Dec 29, 2018

Recently I decided to get a new microSD card for the camera I use to make my videos, as up until now I had been using a very old 16GB Sandisk SD card. I also took the opportunity to browse Amazon to look for a decent memory card reader. I didn't actually have a memory card reader for my main machine, so whenever I had to copy files over from my camera or microphone or so, I had to resort to odd solutions like using my Zoom H1 microphone as a make-do card reader. Certainly not ideal, and quite slow too.

I stumbled upon an Aluminium Unitek USB3 memory card reader that can read microSD, SD as-well as compact flash. One of the reasons I chose this one was that it mentioned it you can access multiple memory cards at the same time, whereas other card readers often cannot. This would be handy for me, as for my video projects I always record audio separately and so need to import from multiple cards. If I can do that at the same time, all the better.

Another benefit of this card reader was that it doesn't look too bad (although I wish the logo wasn't shown so prominently), and that the USB cable was long enough for my needs — they sell both a ~30cm and ~120cm version, I went for the latter, so I could put the reader in an easily accessible spot on my desk.

The card reader has a small white status light that blinks when there is activity, or stays on when it is not doing anything but is connected/powered. I don't really need this and usually prefer no (status) lights, but this one is not very bright, so it's not obtrusive fortunately.

The status light blinks when there is activity. (GIF)
The packaging is very simple; Just a box with a quick start flyer and thank you card.
A thank you card; A common find when purchasing from eBay or AliExpress stores.

Of course the most important part is; how well does it work? Since I haven't used compact flash cards in many years now I don't have one to test with. Instead, I'll focus on the microSD and SD slots.

Inserting two cards at the same time works as expected; both show up on my machine, and can be accessed, read from and written to at the same time, which is exactly what I wanted. The device shows up with a generic (literally) name, as can be seen here:

Generic- USB3.0 CRW-SM/xD Media (or CRW-SD Media for the SD slot).

To get a sense of what kind of what performance can be expected, I used AmorphousDiskMark (a macOS version of CrystalDiskMark, basically) to run read/write tests. For the card I used a Samsung EVO Select 64GB microSDXC, which is the card I purchased to replace the aging 16GB SanDisk with.

The first test I ran with the card in the SD card slot using the microSD to SD adapter provided with the memory card:

Read and write speeds for the Samsung EVO Select 64GB microSDXC using the card reader's SD slot.

For the second test I ran the exact same tests with the same card, but this time using the microSD slot. The results are pretty much identical, so both slots perform as they should.

Read and write speeds for the Samsung EVO Select 64GB microSDXC using the card reader's microSD slot.

The results pretty much match the promised 100MB/s read speeds (95.45 is close enough, I'd say). Obviously random read/writes are much slower, but that can be expected of this kind of card, and is not due to the card reader not performing as expected.

I also purchased a larger 128GB microSD card for another use-case, so I ran the same tests again with this card to see how it performs compared to its smaller brother, just in case there's a mention-worthy difference here. It seems like the write speeds are higher, but everything else is pretty much identical.

The Samsung EVO Select 128GB microSDXC results are similar to the 64GB one, as can be expected.

Using dd to fill up one of the microSD cards with randomness.

I thought about a good way to test the performance when using both slots at the same time, and I thought one way to run the test would be to use dd to fill up one SD card while at the same time running AmorphousDiskMark on the other drive. My hope is that this way I can generate consistent results. I ran the actual test on the 64GB Samsung card, whilst using dd to fill up the 128GB card. In case this is relevant or useful to you, the command I used was:

sudo dd if=/dev/urandom of=/dev/diskX

Replace diskX with whichever drive or partition you mean to target. You can use diskutil list on macOS to see all connected drives and their partitions.

The results of the 64GB Samsung microSD card while at the same time using dd to fill up another card.

Performance seems to be about half for the Sequential QD32 read and write tests, but only about a third for the 4K QD32 read test. 4K read speeds are also considerably slower, at just one-fourth the speed. Sequential read/write are still very good though, at about 80% the performance for read speeds and about 90% for write speeds.

I guess whatever chip they used cannot sustain full speeds for two slots at the same time, as I don't think these speeds are pushing the limits of USB3 in the slightest. But I might be wrong about that.

Still, I'm glad they at least work at the same time, and at least in my use-case I tend to finish copying files from one stick (audio recordings) at lot faster than the other stick (video footage). But this might be important to know for your use-case.


Just for fun and curiosity, I wanted to see what the old SanDisk 16GB SD card I had been using up until now looks like, performance wise, so I ran the tests with this card too. The results pretty much show that it was a good call for me to replace it. Take a look:

Yikes. I should've replaced this card a long time ago.

I also noticed an oddity in my tests; somehow when I format the microSD card using Apple's Mac OS Extended (Journaled) option, Sequential QD32 read speeds are consistently doubled, with all other results roughly the same. I am not sure what causes that, but I'm guessing there's some caching at play. I ran several extra tests using both the Factory-formatted NTFS as-well as ExFAT to make sure my numbers are at least roughly correct, and both report similar results, so I used those for all the above tests.


In conclusion, I am happy with the new card reader. It is so much more convenient for me to be able to just grab the memory cards and plug them into the reader. The Unitek performs fine, and I can comfortably recommend it if you're in the market for a similar solution. I haven't tested many card readers though, so there might be (much) better options out there. This one just happens to be what I chose.

The same applies to the Samsung memory cards I purchased with this reader, they perform exactly as advertised, and very well priced.

At the time of publishing the Samsung EVO 64GB and 128GB models cost $12.99 and $24.99, respectively, and the Unitek memory card reader was $14.99.


note: This is not a sponsored post, I merely share my thoughts here in case someone else is in the market for a (micro)SD card reader. I purchased the reader with my own money, and none of my links are affiliate links. Thank you for reading!