If you're a regular VR-Zone reader you might remember our massive USB 3.0 host controller test that we did back in August last year. As the first of Intel's new 7-series motherboards are starting to appear, we thought we'd do a quick test of Intel's native USB 3.0 host controller and compare how it stacks up to some of the best discrete controllers.For this test we used a Gigabyte Z77X-UD5H and connected up the same Patriot Wildfire SSD and ASMedia ASM1051E SATA 6Gbps to USB 3.0 bridge as we used in our previous test to keep it as level playing ground as possible. We only ran CrystalDiskMark and HD Tune Pro this time around though, but both tests are giving us something of an indicator as to what to expect from Intel's native USB 3.0 host controller.We used the most recent Intel USB 3.0 drivers which at the time were the 1.0.3.214. Admittedly there might have been some performance improvements for the competition since we did our previous test, especially in the case of Renesas as we used a beta driver for the µD720201. Sadly we didn't have all of the host controllers at hand, but we'll be doing some additional testing with newer drivers for the ASMedia ASM1042 and see if there's any difference worth writing about.Starting with the read performance in CrystalDiskMark Intel is straight away up in the lead with 241.9MB/s, although that's only 8.6MB/s faster than the µD720201 which was the fastest host controller last time around in this test. In the write performance test the Etron EJ168A is still in the lead with its 185.9MB/s whereas Intel is 3.7MB/s behind at 182.2MB/s. It goes to show that despite Intel's host controller is part of the chipset, this doesn't mean it'll be faster than a discrete chip under all circumstances.
Moving on to HD Tune Pro, Intel stays ahead of the competition once more with an average read speed of 235.2MB/s, 8.7MB/s faster than the µD720201 this time around. Moving on to the write test, Intel falls behind all three of the tried in tested host controllers from Asmedia, Etron and Renesas. Here the Intel controller barely breaks the 200MB/s barrier and considering all of the other three host controllers are comfortably past the 200MB/s barrier, this is quite disappointing.
Looking at the max read speed Intel isn't doing quite as well as before, as its host controller is only a mere 2.7MB/s ahead of the µD720201. The maximum write speed sees Intel ahead of Renesas, but behind both ASMedia and Etron. Where the Intel USB 3.0 host controller stands out is in the burst read and write tests, as here Intel has a healthy 11.8MB/s lead in the read test and a massive 28.3MB/s lead in the write test. This is obviously an area where an integrated host controller offers some serious benefits over a discrete one and it's possible that once we get faster device side controllers, we could see improved performance from Intel overall.
On the other hand, it looks like Intel needs to do a fair bit of work on its drivers, as with 5.5 percent CPU usage, Intel is as bad as Renesas original µD720200 host controller, albeit it's only half the CPU usage that the host controller in AMD's A75 chipset utilized when we tested it back in August.
We'll be taking a look at the USB 3.0 performance in Windows 8 and see what kind of difference if any, that Microsoft's USB 3.0 host drivers bring to the table. The good news here though is that Intel appears to have done a good job with its USB 3.0 implementation. What this quick little test also tells us is that if you have a USB 3.0 host controller from ASMedia, Etron or the µD720201 from Renesas, you shouldn't really have to worry about the performance difference, as in most instances it's only a few Megabytes per second difference and sometimes it's even in the discrete controllers favour.

Read more: A quick look and comparison of Intel's native USB 3.0 host controller by VR-Zone.com