When I go overseas on travel, shopping tends to be the last thing on my mind. Coming in with almost-overflowing baggage, there’s not much chance to buy anything. After all, if I’m overseas, there’s a good chance I’ve already spent a lot of money just to be there.
But alas, I can’t help being drawn to technology, especially to items which are both high-performance and low-cost. Good value, in a nutshell, is always attractive.
Walking through Golden Computer Center in Hong Kong, I couldn’t help but stand near a rack of memory card readers. My favourite reader that I use on an everyday basis was the Transcend RDF8K. As I had originally kept a spare, but given that away to a relative, I thought I’d go and buy another one in person, rather than via eBay. Maybe it would have been cheaper. Imagine my surprise when I see an updated RDF9K available on the shelf at almost the same price. I was immediately sold.
The reader comes in the same plastic-bubble package as its predecessor, and has the shame shape and size. Included is the reader itself, and a cable. This unit’s main claim is the ability to run UHS-II cards, achieving up to 260MB/s read and 190MB/s write, all on a USB 3.1/3.0 bus. That’s much faster than the 104MB/s limit of UHS-I buses, but only the very expensive and still-rare UHS-II cards can achieve this performance. UHS-II requires the use of a second row of contacts – resulting in a different socket design, but the reader is compatible with older cards. Of interest is that UHS-III which is faster still, has already been specified, although products are even less common.
While I have very little reason to opt for a USB 3.1 UHS-II card reader, as I only have USB 3.0 ports and UHS-I cards, it turned out that the price was not much different. In all, I managed to purchase the reader for just AU$24. With this, came a different chipset which would likely be even better optimized for performance.
If you had an RDF8K like myself, it’s rather difficult to distinguish the two at a glance as they share exactly the same casing, just with slight changes to the label and the addition of the USB 3.1 text on the top.
There is a subtle printing difference near the SD slot – instead of SD MMC as before, UHS-II is now printed. The slot order and position remain the same as the RDF8K, with the exception that the SD slot is deeper to accommodate the extra row of contacts necessary for UHS-II operation. A side effect is that a few scratches may appear on the plastic bodies of non-UHS-II cards due to the scraping of the contacts.
The inside card of the plastic bubble also comes with some catalogues of Transcend products, with a diagrammatic utilization guide. There is also some information about Transcend’s free RecoveRx program which can be used to recover lost data from cards – however, I’ve never used it myself. Warranty and hazardous substance information is also provided.
Just like its predecessor, the top pops off without any tools. It reveals a Realtek RTS5321 chipset, with a small piece of flash containing the firmware. The datasheet doesn’t appear to be easily available, however, products from other vendors including Startech use this chipset as well. For back-up and examination purposes, a dump of the ROM contents was made using the CH341 programmer.
The rear is mostly occupied by the CompactFlash slot, a crystal oscillator and the microUSB-B connector.
Unfortunately, properly testing this reader is not a possibility for me. I do not have any USB 3.1 ports to use it with, thus it is tested using an Intel USB 3.0 port “on location” in Hong Kong instead. I don’t have any UHS-II cards either, thus it was tested with the Samsung Evo+ 128Gb microSDXC card and a Toshiba Exceria Type 2 64Gb SDXC card.
The speeds were in line with expectations, and no errors occured. I also tested writing to the Samsung card.
There was no problem there either, however, I did note that in some rare cases it was possible for a less-than-optimal speed to be negotiated with the card. This was easily fixed by removing and re-inserting the card, and is more likely related to contact “crackle” during insertion disrupting high-speed communication, forcing the chipset to fall-back to a more reliable, slower rate.
In all, it’s actually pretty hard to judge whether this reader is going to be good for USB 3.1/UHS-II operation, but it certainly performs well on UHS-I cards on USB 3.0. No worse than the RDF8 did, and miles better than a lot of older readers.
On further thought, it’s questionable whether USB 3.1 is actually necessary – as they advertise 260MB/s maximums, this is achievable on USB 3.0 with UASP. USB 3.0 without UASP frequently tops out between 120-200MB/s due to back and forth latency – this is not something USB 3.1 would fix, so maybe this reader will offer gains with USB 3.0 alone for UHS-II cards.
No unexpected data integrity issues occured, and testing using the Renesas USB 3.0 chipset was performed in the review of Toshiba Exceria N301 and Samsung Evo+ cards here. The RDF9 produced the best results in sequential tests, but did produce strange 512kB access results with CrystalDiskMark only. This is unlikely to reflect real-life usage.
While the Transcend RDF9K is based around a Realtek chipset solution, it seems that it does offer superior performance compared to the Genesys Logic chipset in the RDF8K. The Genesys Logic chipset is known to have some strange clocks which (at times) may not optimally drive cards at their maximum rate, so this is hardly a surprise. However, compared to the horror that was the Kogan Realtek/Realsil RTS5301 reader, it’s good to see that the Transcend RDF9K doesn’t suffer any notable significant compatibility issues with the cards I use. For its price, it’s a pretty good device even if you don’t need UHS-II or USB 3.1 speeds and works just fine in that role. Maybe it will become my “everyday” reader from now on …
As a plus, now that I have backed up the firmware on my device, maybe you can play with various firmwares if you’re not satisfied …