The USB 3.0 is a lie D:

[Winter_Wolf]

Sad day. Well, not super sad, but still disappointing. I have an ASUS K53SD laptop which I've grown accustomed to, one of the touted features being a USB 3.0 port.

And it's not working at 3.0 speeds. Couple of different devices, using ATTO benchmark, and transfer speed tops out at USB 2.0 max speed. I am 100% certain that the devices themselves are working properly, but I don't want to put the hard drive back in the desktop box. I have up to date drivers for my machine (which are probably woefully out of date) but I'm stumped, and looking for suggestions as I keep searching for a solution.

Anyone had similar problems and managed to get it sorted out?

[Flickerdart]

Are they USB 3.0 devices? Are you using the right port? It's common for laptop manufacturers to include only one or two 3.0 ports and the rest 2.0.

[Winter_Wolf]

Are they USB 3.0 devices? Are you using the right port? It's common for laptop manufacturers to include only one or two 3.0 ports and the rest 2.0.

100% sure. I have ONE USB 3.0 port, it's on the left side of my laptop, it's got the blue coloring of USB 3.0 ports. Both of the devices are advertised and spec for USB 3.0 functionality. So either both product manufacturers are lying to me through blatantly false advertising, or I'm not getting what I paid for with this laptop.

Frankly I'm somewhat more inclined to blame Asus for screwing up (or being shady) than not, since another family member has had several problems with theirs, although it was monitor and built-in cam problems and not failing to live up to the advertised promise of the product.

[Siosilvar]

USB almost never caps out transfer speed. If the transfer speeds are close to 480 Mbps, it's likely that it is actually a 3.0 port, since USB 2.0 usually only runs at ~300 Mbps or less. It's a bit unusual that it'd be that low, but try comparing to a USB 2.0 device on the same port.

[factotum]

What device is it that you have attached to the USB 3.0 port? The speed may be limited by the device itself, not the USB protocol--you're not going to find a spinning rust hard disc that can transfer data much faster than the peak speed of USB 2.0, for example.

[Winter_Wolf]

What device is it that you have attached to the USB 3.0 port? The speed may be limited by the device itself, not the USB protocol--you're not going to find a spinning rust hard disc that can transfer data much faster than the peak speed of USB 2.0, for example.

The particular device I want to function at high speed is a Seagate Baraccuda 7200.12 (http://www.newegg.com/Product/Product.aspx?Item=N82E16822148433) plugged into a Plugable (that's the actual brand name) USB 3.0 docking station (http://plugable.com/products/usb3-sata-uasp1). I'd like to point out that I did use a regular external hard drive which claimed USB 3.0 speeds that was NOT that drive and did not use that dock, as a "control".

Considering that when I had the hard drive wired into the computer and was getting at minimum 400 MB/s transfers on data with a SATA connection, it seems like a huge disconnect that I'm getting at peak with benchmarking (i.e. not "real" conditions) 110-120 MB/s from the USB 3.0 connection. Actual usage gives me an average of 20 MB/s with uncommon bursts of ~40MB/s and very rare ~65-70 MB/s bursts. Given that it touts itself to have "up to" 5Gbps (~640 MB/s), I must say that I'm really disappointed with the situation.

Support from Plugable says that "you're in the ballpark" with the lower speeds, which means I'm not likely to get anything else from them. Chalk this up to "had I but known" I suppose. I would certainly have just gotten the 10/100/1000 ethernet adaptor card for the box and continued using the crossover cable. I was getting better speeds with it even capping out at 10/100. That's messed up. :smallannoyed:

[Don Julio Anejo]

Considering that when I had the hard drive wired into the computer and was getting at minimum 400 MB/s transfers on data with a SATA connection, it seems like a huge disconnect that I'm getting at peak with benchmarking (i.e. not "real" conditions) 110-120 MB/s from the USB 3.0 connection. Actual usage gives me an average of 20 MB/s with uncommon bursts of ~40MB/s and very rare ~65-70 MB/s bursts. Given that it touts itself to have "up to" 5Gbps (~640 MB/s), I must say that I'm really disappointed with the situation.

Support from Plugable says that "you're in the ballpark" with the lower speeds, which means I'm not likely to get anything else from them. Chalk this up to "had I but known" I suppose. I would certainly have just gotten the 10/100/1000 ethernet adaptor card for the box and continued using the crossover cable. I was getting better speeds with it even capping out at 10/100. That's messed up. :smallannoyed:
There's no way you could be getting 400 MB/s from a regular hard drive, they top out around 120 MB/s sustained (~170 burst), and that's for performance drives like WD Caviar Black, which your Seagate is not. For anything more, you need an SSD. USB has nothing to do with this, and the speeds you're getting are fine.

Real-world usage of USB 2.0 is ~12 MB/s write and 30 read so you're still getting a pretty decent improvement. Also, considering you're using a notebook, your copy speed is just as likely constrained by the notebook drive itself (in either direction). If it's running a 5400 RPM drive, you're lucky if it tops out at 70 MB/s.

If you were copying SSD to SSD, that's a whole other story, but you're not.

tl; dr: USB 3.0 is only about 2x as good as 2.0 if you're using regular hard drives that are inherently limited by how fast the platters can spin.

[nedz]

There could be any number of reasons for your performance issues, but without detailed analysis of the entire communication path tying this down is going to be very hard. You are unlikely to ever see the theoretical maximum throughput on any benchmark.

[Winter_Wolf]

Yeah I was probably pulling some numbers out of somewhere. But going through a SATA II connection in the box, I was getting better transfer speeds (less time) than I am with the USB 3.0. About 80 gigs of assorted file sizes from a 5400 rpm WD to the 7200 rpm Seagate, vs about 4 GB of assorted file sizes between the laptop and the same Seagate.
Based on this list of https://en.wikipedia.org/wiki/List_of_device_bit_rates (scroll down to "storage" for SATA 2 and"peripheral" USB 3.0.
I guess I was just expecting that because SATA II caps out at about 3.0 Gbps/300MB/s) and USB 3.0 is claimed at 5 Gbps(625MB/s) that I'd get better speed (or at least equal!).

Reality clearly has decided that these theoretical numbers shall not be mine to play with. In all seriousness, though, I'm sore tempted to just put the hard drive back in the box.

[factotum]

it seems like a huge disconnect that I'm getting at peak with benchmarking (i.e. not "real" conditions) 110-120 MB/s from the USB 3.0 connection.

Well, that in itself is still 3 times faster than USB 2.0 (480 megabits per second = around 60Mb/s maximum theoretical speed, and you'll never get anywhere near that in real-world scenarios), so you're definitely getting USB 3.0 speeds there. It's OK to be disappointed that they're not as fast as you were expecting, but you've certainly not been gypped into getting a USB 2.0 speed rather than USB 3.0 as you were saying! :smallwink:

[Siosilvar]

Well, that in itself is still 3 times faster than USB 2.0 (480 megabits per second = around 60Mb/s maximum theoretical speed, and you'll never get anywhere near that in real-world scenarios), so you're definitely getting USB 3.0 speeds there. It's OK to be disappointed that they're not as fast as you were expecting, but you've certainly not been gypped into getting a USB 2.0 speed rather than USB 3.0 as you were saying! :smallwink:

Careful with your capitalization there. 480 Mbps is 60 MBps, yes, but that's not what you have.

But otherwise this is true. 120 MBps is 960 Mbps is about twice the theoretical maximum of USB 2.0 and about four to five times what you'd get in practice, so it's definitely a 3.0 port.

[Winter_Wolf]

I'm pretty sure I read or heard somewhere that disappointment is what happens when reality doesn't meet our expectations.

It didn't for me, but I just have to get used to the fact that I didn't have a full understanding of what I was getting into before I got into it. Such is life. All it really means in practice is that the external drive will continue to be the storage device for long term, seldom used things, and anything I want to install/access often will be copied onto my local drive.

I still wonder why it is that my supposedly slower connection was giving me better data transfer rates, but it's not like it's going to keep me up nights.

[Raineh Daze]

Motherboard can directly operate it? It might be that.

[factotum]

Plus SATA is optimised specifically for attaching to hard discs, whereas USB is a more general protocol--not too surprising it does better at the thing it's designed to do! :smallsmile:

[scurv]

Something i remind people when it concerns tech, as distasteful as it may be.
It is made by slave labor in a 3rd world country

So this being said quite likely if there is any corner that could of been cut, It was. If there is any standard that could of been shirked, It was. In general read reviews before you buy and if you want quality expect to pay significantly for it.

I can not count the times that i have encountered components from the factory that are out of tolerance, or had substitutions for lesser chip sets for cost reasons. Or substandard soldering or other forms of shoddy construction.

In general the standard of industry is produce a device that will work long enough to make the consumer not upset, But flakes out often enough that they will acquire a new one.
<edit>




I still wonder why it is that my supposedly slower connection was giving me better data transfer rates, but it's not like it's going to keep me up nights.

A from the waist response on that would be most likely a lower error rate in data transmission. Less data corruption=less reword on the device resending data=more time it is sending new data

[Don Julio Anejo]

Something i remind people when it concerns tech, as distasteful as it may be.
It is made by slave labor in a 3rd world country

1. Politics
2. Blatantly untrue, I highly suggest you do some reading on electronics factories in China. True, they're miserable by Western standards, but they're much better off than most other factories there and workers manage to feed entire families using their salaries.


So this being said quite likely if there is any corner that could of been cut, It was. If there is any standard that could of been shirked, It was. In general read reviews before you buy and if you want quality expect to pay significantly for it.

I can not count the times that i have encountered components from the factory that are out of tolerance, or had substitutions for lesser chip sets for cost reasons. Or substandard soldering or other forms of shoddy construction.

Generally, most businesses try to do it now (as opposed to, say, 50 years ago, when stuff was literally built to last). However, blame does not lie with the factories: they will manufacture precisely what you tell them to. Blame lies with design and marketing bureaus of makers; if they choose to cut a corner somewhere, they do. Otherwise, they don't. A lot of really good stuff is made in China too, it's just that the average cheap off-brand knock-off is yes, usually bad and unreliable.