ORICO PFU3-2P. Dual-port PCI Express USB 3.0 card.

Not long ago, I posted a review of a hard drive enclosure. Since it already had a USB 3.0 interface, while my home computer still only had USB 2.0, I decided to give my old PC a bit of an upgrade.
As far as I can tell, Orico is something of a brand name in computer accessories, so I decided to see what this Chinese brand was all about.

Unfortunately, I had to give up my extra network card (there's only one PCI-Express slot), even though I'd gotten used to having it. But lately I've needed higher speeds with external USB devices more often, so I decided to replace it with this PCI-Express card with two USB 3.0 ports. I ordered the 2-port version because I figured that would be more than enough; USB 3.0 wouldn't help most of my devices anyway.

It took a very long time to arrive, and I ordered it before the hard drive enclosure, but the important thing is that it got here—even though I'd already had a couple of packages from this store go missing :(

I'll try to keep this review minimalist, since there's not much to say about this card.

The card arrived in a regular mailer, packed in a cardboard box. No color printing—everything was plain and simple.
One side has a picture showing two versions of the card, with two and four ports.
The other side shows four options. As I understand it, they differ in the number of ports, and some cards also have an internal port for connecting the front panel.
Below that are a bunch of pictures: a wine glass, an umbrella, and a crossed-out trash can.
So, before you throw it away, you should at least have a drink—and preferably do it in the rain, I guess :)


And here's what was inside the box with the wine glass.
An instruction manual or warranty card—I can't really tell what it says.
A disc, which I tossed right away since I don't even have a drive to spin it in. It's about time they started including flash drives instead :)
A metal bracket for installing the card in a low-profile case.
A screw for securing the card to the computer case.
And, of course, the card itself.


The manual (or warranty card) is entirely in Chinese. I couldn't quite tell whether it's traditional or simplified, but I doubt anyone cares—it's all Greek to me either way.
The disc is supposed to contain the drivers, but since my computer hasn't had a disc drive for about seven years, I went online and downloaded the driver.
Still, including a disc is a good thing in itself. Last time, with the network card, there wasn't one (fortunately Windows recognized it on its own). Otherwise, you'd need to download the network card driver from the internet, but you wouldn't have internet because the driver wasn't installed.


The bracket and screw: the bracket is blackened, and the lettering is printed cleanly. The shiny screw is neatly packed in a little zip-lock bag :)


The card itself is stuffed into an almost opaque antistatic bag, as if it could give someone an electric shock. They even drew some kind of alien hand with two fingers on it.


And here's the board. The PCB is yellow, the soldering is neat, and the flux has been cleaned off. No complaints from an electronics engineer's point of view.


The board is double-sided, but the components are mounted on one side only, so there are no components on the back. That's a good thing, since the card usually sits in the computer with this side facing up, so there's no risk of knocking anything off the next time you clean out the dust.


Two USB ports are brought out on the rear. It comes with a full-height bracket installed; you can replace it with a low-profile one if needed.
There is a power connector on the back of the board. These connectors used to power hard drives and CD/DVD drives; they’re generally no longer used, so I don’t think finding a spare connector will be a problem. They are still used occasionally, usually to power various expansion cards (though the only ones that come to mind are cards like the one in this review).


Having learned from experience with computer expansion cards, I always check a board beforehand for obvious defects, solder splashes, misaligned components, and so on.
But I had no complaints after inspecting this one.
There are quite a lot of electrolytic capacitors on the board. It’s clear the manufacturer wasn’t trying to cut costs, which is always nice to see.


A small piece of paper is stuck to the controller, hiding the chip’s name: FL1009, made by the very little-known company Fresco Logic. I found only four types of chips in the company’s product lineup.
Honestly, I expected a chip from a much better-known company, one that made at least five or six types :)
It’s odd, really, to use a chip from some obscure company when there’s NEC, for example—and that’s with this card being far from the cheapest.


The controller is powered by two small DC-DC PWM controllers. I don’t know their part numbers; if anyone does, let me know and I’ll add the information to the review.


There’s also another, clearly more powerful, PWM controller on the board. There’s a little more information about this one: it’s a 2 A, 150 kHz, 45 V PWM controller, the td1509p5.
This is the one that left me a bit puzzled. According to the information I found about this controller, it’s connected in a way that makes no sense to me.
It’s a step-down DC-DC converter, so it should have 8–12 V going in, followed by an inductor and a diode to ground.
But the board is laid out completely differently. Power goes to the inductor, as in a boost converter, then there’s a diode to ground, as in a buck converter, and then power goes to… the chip’s output.
It looks like the chip is being used in some very unconventional way.
I suspect it’s being used as electronic overload protection, so knowing its part number could come in handy and make repairs much easier if it fails.


That wraps up the hardware analysis. Now I’ll move on to the software and testing.
I don’t know how it will behave in Windows 7, but Windows XP had never heard of it and was shocked—imagine that, such a well-known controller manufacturer :)
I found a driver in five seconds by searching for the chip name. The manufacturer may be little-known, but it’s quite helpful, and the link to the driver was easy to find.

Installing the driver was trouble-free too: a few mouse clicks and everything works.
In Device Manager, the card showed up as two Fresco Logic devices.


The controller itself:
PCIVEN_1B73&DEV_1009&SUBSYS_00001B73&REV_024&293AFFCC&1&00E0
And the hub:
FLUSBROOT_HUB_FL305&29C60FCE&0
The drivers are fairly recent, from January this year. I suspect the company itself isn’t all that old either.


Since I couldn't find anything faster at home than an external hard drive with a USB 3.0 interface, that's what I'll use for testing. I'll also compare its speed with the USB 3.0 controller built into the motherboard.
For the test, I repeated the same benchmark with the same hard drive, in the same program and under the same conditions.
Screenshots of the card under review are on the left; the motherboard's built-in controller is on the right


Summary.
Pros
The card works perfectly; nothing freezes.
Although I couldn't fully test the speed, I haven't noticed any major differences compared with the built-in controller so far.
High-quality construction
Good set of accessories.

Cons
In my opinion, it's a bit expensive for a card with a controller from an unknown manufacturer.

Conclusion. It's a reasonably good card. Of course, only time will tell, and it may be put under heavier loads, but I ran it for about two days and didn't notice any defects. The controller may be unknown, but it works fine. I was pleased with the included accessories. Owners of low-profile cases can add a couple of USB 3.0 ports, which can certainly come in handy.

I hope this information was useful. If I get something faster, I'll be sure to test it.