Showing Results: Most Recent
Pros: HD Tune Pro 4.61 Read Benchmark results using a MX200 M.2 SSD:
Windows 7 Intel M.2: 524.4 MB/s
Windows 7 USB 3.0: 282 MB/s
Windows 7 USB 3.0 with Boost: 312.1 MB/s
Windows 8.1 USB 3.0: 343 MB/s
Windows 7 PCIE 2.0 x16 Sunix USB2312 with USB 3.1: 334.5 MB/s
Windows 8.1 PCIE 3.0 x16 Sunix USB2312 with USB 3.1: 419.3 MB/s
Cons: Doesn't reach USB 3.1 theoretical max speeds.
Other Thoughts: A couple of things to note before I get started.
USB 3.0 bandwidth is 5 Gigabits per second (Gbps). To convert that to Megabytes per second (MB) is 128 × dGb = dMB. Therefore, USB 3.0 speeds have a theoretical max full duplex speed of 640 MB/s. However, since USB uses 8b/10b line code, turning eight-bit symbols into ten-bit ones, it incurs a 20% bandwidth hit. Thus, you'll see peak throughput of 512 MB/s. However, something causes slowdowns up to 45.5% or more, not just 20%, and no one really has specified why that is. That means you could see USB 3.0 flash drives or SSD's connected via USB 3.0 reach speeds close to 348.8 MB/s for sequential reads and writes, but no faster than that.
USB 3.1 bandwidth is 10 Gbps, twice as much bandwidth as USB 3.0, which equates to 1,280 MB/s theoretical speeds. Encoding is 128b/132b, so instead of a 20% bandwidth hit, it's only 3%. Nice! Thus, you'll see peak throughput of 1,241.6 MB/s? That means you could see USB 3.1 SSD's connected via USB 3.1 reach speeds close to 676.62 MB/s (factoring in the 45.5% bandwidth hit theory of mine). Sounds crazy, right?
Unfortunately, we are not going to hit that mark with the Sunix USB2312. Here's why.
The Sunix USB2312 supports PCIe 2.0 x2. Since a single PCIe 2.0 lane supports 512 MB/s, multiply that by two, and that's 1,024 MB/s. However, we run into the same 20% bandwidth hit, so, at best, that's 819.2 MB/s. That means you could see USB 3.1 SSD's connected via USB 3.1 reach speeds up to 446.46.6 MB/s (again, factoring in the 45.5% bandwidth hit theory), but no faster.
With USB 3.0 on Windows 8.1, I've been able to hit around ~343 MB/s. With USB 3.0 on Windows 7, it seems to top out at ~282 MB/s. However, when I'm able to use an ASUS board with USB 3.0 Boost, and enabling UASP, I hit around ~312.1 MB/s.
Using the Sunix USB2312 on Windows 8 with a PCIE 3.0 x16 slot, I get ~419.3 MB/s. On Windows 7 with a PCIE 2.0 X16 slot, I get ~334.5 MB/s. Interestingly enough, the Sunix USB2312 actually seems to support PCIE 3.0, even though the specs doesn't mention it does.
I'll probably play around with this some more in the future. The Sunix USB2312 is definitely nice to have, and it does provide the fastest USB Sequential Read speeds, but it's still not better than using a M.2 or SATA 6G connection. There's just too many known and unknown bandwidth hits preventing USB 3.1 from reaching it's peak theoretical potential.
However, it is nice to have something like this on an older x58 system to get 334.5 MB/s, although I was surprised that on Windows 8.1 using PCIE 3.0, I was able to get up to 419.3 MB/s. I'll need to do another test with Windows 7 and PCIE 3.0 x 16 to see if I get the same numbers. It may just be that Windows 8.1 handles USB 3.0 and USB 3.1 protocols a bit better.
Pros: See Other thoughts.
Cons: See Other thoughts.
Other Thoughts: I'm wondering about Crucial's new business strategy concerning the BX100 and MX200 line of SSD's.
1. There's no official reviews for the BX100, but lots of heavy marketing saying how fast it is compared to a Hard Drive. Really? Comparing it against Hard Drives in this day and age? The BX100's are Crucial's budget line of SSD's. Their first SSD product that uses the Silicon Motion 2246EN controller. It uses 16nm 128Gbit NAND (but we don't know if it's MLC or TLC NAND - most likely the latter, but we don't really know because there's no official reviews for it). It also has no M-class features like hardware-accelerated encryption or SLC caching.
How come no official reviews have come out for it before release like other SSD's?
2. The MX200, from two official reviews, shows it as being identical to the MX100, other than a firmware update having it perform differently. If you applied the same firmware that's on the MX200 on the MX100, would it essentially become a MX200? I'm afraid that if this is true, then consumers are essentially paying for firmware updates. It would be like if NVIDIA releasing a GTX 980, stopped giving out driver updates past 347.52, then releasing a GTX 1080 (which is exactly the same as a GTX 980), but it gets driver updates past 347.52, which increases performance by 10% - 15% over the previous drivers.
If you look at Crucial's Firmware download page, you'll notice that there's no firmware updates for the MX100, even though it was released over 7 months ago (June 2014). You've got to wonder about that.
This review is from: UtechSmart USB 3.0 4-Port Hub, Backward Compatible with USB 2.0 - Black
Pros: Works with USB 3.0 devices that don't need a lot of power.
Other Thoughts: I own this and the UtechSmart 7-Port USB 3.0 Hub with VIA Vl812 Chipset and 5V/4A Power Adapter [2013 Updated Version].
If you have USB 3.0 devices that have their own power supply, then this will work great. If not, then you want to get the 7-Port Hub I mentioned to work with devices like USB 3.0 flash drives.
Remember, this USB 3.0 Hub doesn't have its own power supply, so don't expect it to work with power hungry USB 3.0 devices that don't have one, either.