USB 3.0 'super speed' performance.

[Peter Jason]

Win7 SP1

I have just installed a "USB3.0 4-port Super-Speed
PCIe Host Controller" of the Sunix Company.
SUN-USB4300NS Sata power connector
http://www.anyware.com.au/SUN-USB4300NS.aspx



Its brochure says I should get >300MB/Sec if the
PCIe is a Version 2.
I bought the very expensive Corsair64GB USB3 flash
drive to go with it.
http://webshop.eurosys.be/repository/product/gallery/Pic_14778483_1564291.jpg

I got it working, but how do I test its speed? And
how do I know if the PCIe is Version2?

Peter

 

[Paul]

Win7 SP1

I have just installed a "USB3.0 4-port Super-Speed
PCIe Host Controller" of the Sunix Company.
SUN-USB4300NS Sata power connector
http://www.anyware.com.au/SUN-USB4300NS.aspx



Its brochure says I should get >300MB/Sec if the
PCIe is a Version 2.
I bought the very expensive Corsair64GB USB3 flash
drive to go with it.
http://webshop.eurosys.be/repository/product/gallery/Pic_14778483_1564291.jpg

I got it working, but how do I test its speed? And
how do I know if the PCIe is Version2?

Peter

You can do a read transfer test with the free version of HDTune.

http://www.hdtune.com/files/hdtune_255.exe

*******

As for the PCI Express lanes, we'd need to know more about
your computer motherboard for that.

Apparently, this is the chip on the controller card. From Renesas.
The Rev2 lane runs at either 500MB/sec or 250MB/sec, depending
on the motherboard and chipset.

http://www.renesas.com/products/soc/usb_assp/product/upd720201/index.jsp

The 64GB Flash Voyager GT USB 3.0, is 220MB/sec on read and 110MB/sec
on write. You might not be able to detect the PCI Express lane
type using it.

http://www.corsair.com/en/usb-drive/flash-voyager-gt.html

This slide set from a university, seems to rate PCI Express Rev 1.1 at
200MB/sec practical (page 2). So maybe if your Corsair read benchmark
is capped at 200, that would tell you the lane used for the card
is Rev 1.1 and not Rev 2 or Rev 3.

http://www4.cs.umanitoba.ca/~comp4690/Documents/07HighPerformanceIOandStorage-3up.pdf

Using this document, they have a set of curves on page 2. To achieve
220MB/sec on a 250MB/sec physical layer, would be an efficiency of 0.88.
The document claims that Intel desktop chipsets use relatively
small buffers, which makes getting to 0.88 efficiency unlikely.
And you might be able to see that in your HDTune results, if you
have Rev 1 lanes on the motherboard. So maybe the cap would be
detectable.

http://www.plxtech.com/files/pdf/technical/expresslane/Choosing_PCIe_Packet_Payload_Size.pdf

If your HDTune benchmark curve reports 220MB/sec, then the PCI Express
lane is likely Rev 2.0. If you get 200MB/sec or less, it's probably
Rev 1.1. You can use the PLXTech curves, to estimate what percentage
of a 250MB/sec or 500MB/sec lane is achievable.

Paul

 

[ray carter]

Win7 SP1

I have just installed a "USB3.0 4-port Super-Speed PCIe Host Controller"
of the Sunix Company.
SUN-USB4300NS Sata power connector
http://www.anyware.com.au/SUN-USB4300NS.aspx



Its brochure says I should get >300MB/Sec if the PCIe is a Version 2.
I bought the very expensive Corsair64GB USB3 flash drive to go with it.
http://webshop.eurosys.be/repository/product/gallery/ Pic_14778483_1564291.jpg

I got it working, but how do I test its speed? And how do I know if the
PCIe is Version2?

Peter

One way would be to boot a Linux Live CD and do 'lsusb' in a terminal
window.

 

[s|b]

I bought the very expensive Corsair64GB USB3 flash
drive to go with it.
http://webshop.eurosys.be/repository/product/gallery/Pic_14778483_1564291.jpg

According to

<http://webshop.eurosys.be/nl/data-opslag/usb-sticks/corsair-64gb-cmfvygt3a64gb-p375365>

you paid 80,92 euro for this? I bought a SanDisk Extreme USB 3.0 Flash
Drive 64GB on eBay for 58 euro.

<http://www.ebay.co.uk/itm/SanDisk-E...80-/230805643836?ssPageName=ADME:L:OC:BE:3160>

I'm just saying...

I got it working, but how do I test its speed? And
how do I know if the PCIe is Version2?

Test speed:

CrystalDiskMark
<http://crystalmark.info/download/index-e.html>

These are the results of my SanDisk:
<http://img809.imageshack.us/img809/5638/84693046.png>

As for your second question: I don't know.

 

[Paul]

According to

<http://webshop.eurosys.be/nl/data-opslag/usb-sticks/corsair-64gb-cmfvygt3a64gb-p375365>

you paid 80,92 euro for this? I bought a SanDisk Extreme USB 3.0 Flash
Drive 64GB on eBay for 58 euro.

<http://www.ebay.co.uk/itm/SanDisk-E...80-/230805643836?ssPageName=ADME:L:OC:BE:3160>

I'm just saying...


Test speed:

CrystalDiskMark
<http://crystalmark.info/download/index-e.html>

These are the results of my SanDisk:
<http://img809.imageshack.us/img809/5638/84693046.png>

As for your second question: I don't know.

A poster here got:

http://www.newegg.com/Product/Product.aspx?Item=20-171-649&ParentOnly=1&IsVirtualParent=1

In the sequential read speed test... SanDisk Extreme (179.01 MB/s)
In the sequential write speed test... SanDisk Extreme (118.34 MB/s)

Using the same method (a user review)

"CORSAIR Voyager GT 64GB USB 3.0 Flash Drive Model CMFVYGT3A-64GB"
http://www.newegg.com/Product/Produ...rue&Keywords=(keywords)&Page=1#scrollFullInfo

CrystalDiskMark 3.0.1 64-bit results:
Sequential Read: 225.8MB/s, write 100.9MB/s

I think I'd sooner have the extra write speed.

But since I bought something recently from Sandisk, and it
was a dog, I won't be doing business with them. (I don't mind
doing business with people, as long as the packaging for the
product describes *exactly* how poorly it works. What I don't
like, is being deceived on purpose.)

*******

It's strange how so many people have trouble with that USB3 stuff.
The average review isn't very good. The Corsair seems to
have a problem with certain brands of USB3 cards (chip issue?).

It's the usual deal - don't buy without checking the reviews
first. The inside of the USB flash, could have an inferior
brand of controller chip. That's how they cut costs and
make money. When the stick won't work with certain USB3
cards, that tells me the flash controller didn't actually
pass USB3 certification. Or the PCB the chip sits on,
wasn't designed properly. Or the company making the
tiny PCBs, didn't make them right. (You don't make those
PCBs in house, you contract that out. You have to do incoming
inspection, if you suspect a problem with them.)

A company designing such crap, could send the stick to a
third-party lab, for verification at the interface.
You don't have to buy all the gear and keep it in house.
You can rent expertise from others, especially if you're
a small company, and don't have the budget for all the
lab equipment.

There are even people willing to explain it

http://www.evaluationengineering.com/articles/200901/usb-30-physical-layer-measurements.php

Paul

 

[Peter Jason]

You can do a read transfer test with the free version of HDTune.

http://www.hdtune.com/files/hdtune_255.exe

*******

As for the PCI Express lanes, we'd need to know more about
your computer motherboard for that.

Apparently, this is the chip on the controller card. From Renesas.
The Rev2 lane runs at either 500MB/sec or 250MB/sec, depending
on the motherboard and chipset.

http://www.renesas.com/products/soc/usb_assp/product/upd720201/index.jsp

The 64GB Flash Voyager GT USB 3.0, is 220MB/sec on read and 110MB/sec
on write. You might not be able to detect the PCI Express lane
type using it.

http://www.corsair.com/en/usb-drive/flash-voyager-gt.html

This slide set from a university, seems to rate PCI Express Rev 1.1 at
200MB/sec practical (page 2). So maybe if your Corsair read benchmark
is capped at 200, that would tell you the lane used for the card
is Rev 1.1 and not Rev 2 or Rev 3.

http://www4.cs.umanitoba.ca/~comp4690/Documents/07HighPerformanceIOandStorage-3up.pdf

Using this document, they have a set of curves on page 2. To achieve
220MB/sec on a 250MB/sec physical layer, would be an efficiency of 0.88.
The document claims that Intel desktop chipsets use relatively
small buffers, which makes getting to 0.88 efficiency unlikely.
And you might be able to see that in your HDTune results, if you
have Rev 1 lanes on the motherboard. So maybe the cap would be
detectable.

http://www.plxtech.com/files/pdf/technical/expresslane/Choosing_PCIe_Packet_Payload_Size.pdf

If your HDTune benchmark curve reports 220MB/sec, then the PCI Express
lane is likely Rev 2.0. If you get 200MB/sec or less, it's probably
Rev 1.1. You can use the PLXTech curves, to estimate what percentage
of a 250MB/sec or 500MB/sec lane is achievable.

Paul

I have just done the HD-Tune test with flash
drives on the new PCIe card.

The Motherboard manual says I'm using the PCIEX1
slot. Is this version 1? If so, can it be
upgraded?
The Motherboard is:
Gigabyte-X58A-UD7 (installed Dec2010)

Anyway the results came in as:

Read/Write MB/Sec
21 / 10 Corsair Flash voyager 4GB usb2
22 / 10 Sandisk 8GB Backup Ultra
32 / 6 Sandisk Cruzer Blade 4GB
18 / 7 Transcend Jetflash 8GB
17 / 8 Sandisk U3 cruzer Micro 2GB
70 / 70 Corsair Voyager GT3.0 32GB
160 / 100 Corsair Voyager GT3.0 64GB (The new
one).

The HD-Tune seems to have no way of telling a USB2
from a USB3.

Why the huge difference in the last two drives?

I guess I'll have to be happy with an improvement.


Peter

 

[Paul]

I have just done the HD-Tune test with flash
drives on the new PCIe card.

The Motherboard manual says I'm using the PCIEX1
slot. Is this version 1? If so, can it be
upgraded?
The Motherboard is:
Gigabyte-X58A-UD7 (installed Dec2010)

Anyway the results came in as:

Read/Write MB/Sec
21 / 10 Corsair Flash voyager 4GB usb2
22 / 10 Sandisk 8GB Backup Ultra
32 / 6 Sandisk Cruzer Blade 4GB
18 / 7 Transcend Jetflash 8GB
17 / 8 Sandisk U3 cruzer Micro 2GB
70 / 70 Corsair Voyager GT3.0 32GB
160 / 100 Corsair Voyager GT3.0 64GB (The new
one).

The HD-Tune seems to have no way of telling a USB2
from a USB3.

Why the huge difference in the last two drives?

I guess I'll have to be happy with an improvement.


Peter

70 / 70 Corsair Voyager GT3.0 32GB
160 / 100 Corsair Voyager GT3.0 64GB (The new one)

That difference is caused by the number of occupied flash
channels inside the device.

The 32GB one is wired like this.

USB3 ---- controller ----X
----X
---- 16GB_flash_chip
---- 16GB_flash_chip

When they make the 64GB version, then all channels are
occupied. That's the basic idea. There could be other
details I'm missing there. If they needed to make a 128GB
one, they can use 32GB chips, or two 16GB chips sharing a
channel, and so on. But when the device has a small capacity,
some channels end up unused, and bandwidth drops.

What a user does, when shopping, is compare the 16GB, 32GB,
64GB, 128GB ones and so on, and sees where the read and write
bandwidth stop increasing. That then tells you, what is a
"good" size to buy. Based on your "sample of two", we now
know the 64GB one is better than the 32GB one But you can
also get this from a Newegg advertisement for the product
(read/write bandwidth).

*******

You are in luck. Gigabyte usually has pretty good motherboard
manuals.

X58A-UD7

"(The PCIEX16_1, PCIEX16_2, PCIEX8_1 and PCIEX8_2 slots
conform to PCI Express 2.0 standard.)"

"2 x PCI Express x1 slots" <-- no spec = Rev 1.1 ( 250MB/sec )

And confirming, from the Intel ICH10R spec

"PCI Express Interface

The ICH10 provides up to 6 PCI Express Root Ports, supporting
the PCI Express Base Specification, Revision 1.1." <-- Your x1 slots...

Try moving your new USB3 card, into one of the four video card
slots. Then rerun your two tests.

I presume, somewhere along the way, your system picked up a UAS driver.
Maybe you wouldn't have got this far, without it.

http://en.wikipedia.org/wiki/USB_Attached_SCSI

This mentions Windows 8 has UAS driver built-in. For the others,
you use the driver CD that came with the USB3 card.

http://msdn.microsoft.com/en-ca/library/windows/hardware/gg487327.aspx#Do_I_need

Paul

 

[Wolf K]

On 4/13/2013 9:36 PM, Peter Jason wrote:
[...]

The HD-Tune seems to have no way of telling a USB2
from a USB3.

[...]


Correct me if I'm wrong, but of you connect a USB2 to flash memory to a
USB3 port, the speed drops back to USB2 levels. I don't see how any
utility could change this, so even if HD Tune could tell what was what,
seems to me it wouldn't make a difference.

Clarification(s) requested,

 

[Paul]

On 4/13/2013 9:36 PM, Peter Jason wrote:
[...]

The HD-Tune seems to have no way of telling a USB2
from a USB3.

[...]


Correct me if I'm wrong, but of you connect a USB2 to flash memory to a
USB3 port, the speed drops back to USB2 levels. I don't see how any
utility could change this, so even if HD Tune could tell what was what,
seems to me it wouldn't make a difference.

Clarification(s) requested,

There is a way to tell what "mode" a port is in.

Utilities like USBView or UVCView are examples.

There is a field in the config information,
that reflects the negotiated connect rate.

On a USB3 device (like a flash stick), both
a USB2 set of pins, and a USB3 set of pins exist.
But they don't try to operate them in parallel.
It's either one runs, or the other. And the
config information readout, done any time, should
reflect that.

I don't know if UVCView has been updated for
USB3. It probably requires you to get the raw data
from config, and interpret it yourself.

In Linux, the "lsusb" probably gets updated
enough from release to release, to keep track
of this stuff. Microsoft apparently doesn't
see the value in maintaining the USBView/UVCView
tools. They're a pain to get.

*******

Another thing being discussed here, is the impact
the motherboard bus has, on a USB3 card. If you add
a USB3 card, it's possible for the card performance
to be limited by the motherboard bus.

PCI Express x1 slots come in Revision 1 or Revision 2.
Revision 1 is 250MB/sec. Revision 2 is 500MB/sec. The
PCI Express traffic is in "packets", and the packets
have overhead. You look at a graph of buffer size
versus efficiency, to see what effect the efficiency has.
If the chipset buffers were small, perhaps the packets
would have 20% overhead, and only 80% of a packet carries
user traffic (disk transfer). Then, you get
250MB/sec * 0.8 = 200MB/sec. Exactly what you get,
will depend on the buffer size chosen by the chipset
maker.

I was shocked to learn, that my motherboard with x16
graphics slot (rated at 4GB/sec), actually gets
slightly less than 2GB/sec in the "real world". And
that's due to the usage of small buffers in the chipset
design. Enthusiast sites have no way of measuring this,
which is why it slips under the radar. And you get
to notice the effect, if you're a USB3 user (where the
USB3 is via an add-on PCI Express chip).

Paul

 

[Gene E. Bloch]

On 4/13/2013 9:36 PM, Peter Jason wrote:
[...]

The HD-Tune seems to have no way of telling a USB2
from a USB3.

[...]

Correct me if I'm wrong, but of you connect a USB2 to flash memory to a
USB3 port, the speed drops back to USB2 levels. I don't see how any
utility could change this, so even if HD Tune could tell what was what,
seems to me it wouldn't make a difference.

Clarification(s) requested,

I'm a little unclear interpreting what you wrote, but assuming I got it
right, the following should be a clarification

1. Any USB2 device connected to a USB3 port runs at USB2 speeds. In
other words, the USB3 port acts like a USB2 port.

2. Also, any USB3 device connected to a USB2 port runs at USB2 speeds.
In other words, the USB3 device acts like a USB2 device.

3. If you use a USB2 cable to connect a USB3 device, then it acts like a
USB2 device even on a USB3 port.

USB3 Type A connectors have extra pins on the blue blade. When a USB2
connector joins a USB3 connector the USB3 pins are isolated.

 

[Wolf K]

On 4/13/2013 9:36 PM, Peter Jason wrote:
[...]

The HD-Tune seems to have no way of telling a USB2
from a USB3.

[...]


Correct me if I'm wrong, but of you connect a USB2 to flash memory to
a USB3 port, the speed drops back to USB2 levels. I don't see how any
utility could change this, so even if HD Tune could tell what was
what, seems to me it wouldn't make a difference.

Clarification(s) requested,

There is a way to tell what "mode" a port is in. [snip]

Thanks to you and Gene for information. It is just as I thoughted: the
USB level of the memory stick or port, as the case may be, is the
governing factor.

 

[Wolf K]

On 4/13/2013 9:36 PM, Peter Jason wrote:
[...]
The HD-Tune seems to have no way of telling a USB2
from a USB3.
[...]


Correct me if I'm wrong, but of you connect a USB2 to flash memory to
a USB3 port, the speed drops back to USB2 levels. I don't see how any
utility could change this, so even if HD Tune could tell what was
what, seems to me it wouldn't make a difference.

Clarification(s) requested,

There is a way to tell what "mode" a port is in. [snip]

Thanks to you and Gene for information. It is just as I thoughted: the
USB level of the memory stick or port, as the case may be, is the
governing factor.

More concisely: If either port or memory stick is USB2, then you run at
USB2 speed.

 

[John Williamson]

Wolf K wrote:
On 4/13/2013 9:36 PM, Peter Jason wrote:
[...]
The HD-Tune seems to have no way of telling a USB2
from a USB3.
[...]


Correct me if I'm wrong, but of you connect a USB2 to flash memory to
a USB3 port, the speed drops back to USB2 levels. I don't see how any
utility could change this, so even if HD Tune could tell what was
what, seems to me it wouldn't make a difference.

Clarification(s) requested,


There is a way to tell what "mode" a port is in. [snip]

Thanks to you and Gene for information. It is just as I thoughted: the
USB level of the memory stick or port, as the case may be, is the
governing factor.

More concisely: If either port or memory stick is USB2, then you run at
USB2 speed.

And if it's a cable old enough to be pre-USB2, then it will all run at
USB 1 speeds. I have one or two such in the drawer, and I keep meaning
to "lose" them permanently, every time they surface.

 

[Wolf K]

On 4/14/2013 1:05 PM, Paul wrote:
Wolf K wrote:
On 4/13/2013 9:36 PM, Peter Jason wrote:
[...]
The HD-Tune seems to have no way of telling a USB2
from a USB3.
[...]


Correct me if I'm wrong, but of you connect a USB2 to flash memory to
a USB3 port, the speed drops back to USB2 levels. I don't see how any
utility could change this, so even if HD Tune could tell what was
what, seems to me it wouldn't make a difference.

Clarification(s) requested,


There is a way to tell what "mode" a port is in. [snip]

Thanks to you and Gene for information. It is just as I thoughted: the
USB level of the memory stick or port, as the case may be, is the
governing factor.

More concisely: If either port or memory stick is USB2, then you run at
USB2 speed.

And if it's a cable old enough to be pre-USB2, then it will all run at
USB 1 speeds. I have one or two such in the drawer, and I keep meaning
to "lose" them permanently, every time they surface.

H'm. Any way to tell, apart from actually using them?

TIA

 

[John Williamson]

On 4/15/2013 9:28 AM, Wolf K wrote:
On 4/14/2013 1:05 PM, Paul wrote:
Wolf K wrote:
On 4/13/2013 9:36 PM, Peter Jason wrote:
[...]
The HD-Tune seems to have no way of telling a USB2
from a USB3.
[...]


Correct me if I'm wrong, but of you connect a USB2 to flash memory to
a USB3 port, the speed drops back to USB2 levels. I don't see how
any
utility could change this, so even if HD Tune could tell what was
what, seems to me it wouldn't make a difference.

Clarification(s) requested,


There is a way to tell what "mode" a port is in. [snip]

Thanks to you and Gene for information. It is just as I thoughted: the
USB level of the memory stick or port, as the case may be, is the
governing factor.

More concisely: If either port or memory stick is USB2, then you run at
USB2 speed.

And if it's a cable old enough to be pre-USB2, then it will all run at
USB 1 speeds. I have one or two such in the drawer, and I keep meaning
to "lose" them permanently, every time they surface.

H'm. Any way to tell, apart from actually using them?

TIA

Unfortunately, no, unless they have clear covers and you can check for
screening. Anything other than full size A - B or A - A cables is new
enough not to suffer from the problem, though.

 

[Paul]

On 4/15/2013 9:28 AM, Wolf K wrote:
On 4/14/2013 1:05 PM, Paul wrote:
Wolf K wrote:
On 4/13/2013 9:36 PM, Peter Jason wrote:
[...]
The HD-Tune seems to have no way of telling a USB2
from a USB3.
[...]


Correct me if I'm wrong, but of you connect a USB2 to flash memory to
a USB3 port, the speed drops back to USB2 levels. I don't see how
any
utility could change this, so even if HD Tune could tell what was
what, seems to me it wouldn't make a difference.

Clarification(s) requested,


There is a way to tell what "mode" a port is in. [snip]

Thanks to you and Gene for information. It is just as I thoughted: the
USB level of the memory stick or port, as the case may be, is the
governing factor.

More concisely: If either port or memory stick is USB2, then you run at
USB2 speed.

And if it's a cable old enough to be pre-USB2, then it will all run at
USB 1 speeds. I have one or two such in the drawer, and I keep meaning
to "lose" them permanently, every time they surface.

H'm. Any way to tell, apart from actually using them?

TIA

Shielded cable is USB2.

In this example of a USB2 cable, you can see the shield through the
transparent insulating covering on the cable.

http://g-ecx.images-amazon.com/images/G/01/ciu/57/85/d882810ae7a0bace13559110.L._SX300_.jpg

I like that style of cable, because you know for sure it is
shielded, and more likely to run at USB2 rates.

*******

The keyboard cable, that runs at very low rates, doesn't have
a shield. The cable is also thinner as a result, because there
is no shield to take up space. Some keyboards, that cable is
captive, and connected to the keyboard so it can't be removed.
But occasionally, you'll find a detachable, unshielded keyboard
cable, and using that for a USB flash will give very low rates.
Or might not work at all.

You should be able to use a shielded cable with a USB keyboard,
so in fact you could run your computer room with nothing
but shielded cables. Those evil keyboard cables should
just disappear. The evil keyboard cables are usually
grey in color (as they predate the "colorful" era).
And significantly thinner than your other good USB cables.

Paul

 

[Gene E. Bloch]

On 4/15/2013 9:28 AM, Wolf K wrote:
On 4/14/2013 1:05 PM, Paul wrote:
Wolf K wrote:
On 4/13/2013 9:36 PM, Peter Jason wrote:
[...]
The HD-Tune seems to have no way of telling a USB2
from a USB3.
[...]


Correct me if I'm wrong, but of you connect a USB2 to flash memory to
a USB3 port, the speed drops back to USB2 levels. I don't see how any
utility could change this, so even if HD Tune could tell what was
what, seems to me it wouldn't make a difference.

Clarification(s) requested,


There is a way to tell what "mode" a port is in. [snip]

Thanks to you and Gene for information. It is just as I thoughted: the
USB level of the memory stick or port, as the case may be, is the
governing factor.

More concisely: If either port or memory stick is USB2, then you run at
USB2 speed.

And if it's a cable old enough to be pre-USB2, then it will all run at
USB 1 speeds. I have one or two such in the drawer, and I keep meaning
to "lose" them permanently, every time they surface.

H'm. Any way to tell, apart from actually using them?

TIA

Look a the printing on the cable. There will usually be a long string of
characters, small and of low contrast, repeated along the length of the
cable, and if it says USB 2.0 or High Speed USB, then it's probably safe
to assume that it's as it says.

If it doesn't say USB 2.0 (even if it's shielded) it goes into my
recycling bin, like what John Williamson said above.

 

[Gene E. Bloch]

like what John Williamson said above

More precisely, "analogous to what JW said".

 

[Char Jackson]

On 4/13/2013 9:36 PM, Peter Jason wrote:
[...]
The HD-Tune seems to have no way of telling a USB2
from a USB3.
[...]


Correct me if I'm wrong, but of you connect a USB2 to flash memory to
a USB3 port, the speed drops back to USB2 levels. I don't see how any
utility could change this, so even if HD Tune could tell what was
what, seems to me it wouldn't make a difference.

Clarification(s) requested,

There is a way to tell what "mode" a port is in. [snip]

Thanks to you and Gene for information. It is just as I thoughted: the
USB level of the memory stick or port, as the case may be, is the
governing factor.

I would have phrased it something like, "the overall speed is limited to the
speed of the slowest link in the chain."

 

[Gene E. Bloch]

Wolf K wrote:
On 4/13/2013 9:36 PM, Peter Jason wrote:
[...]
The HD-Tune seems to have no way of telling a USB2
from a USB3.
[...]


Correct me if I'm wrong, but of you connect a USB2 to flash memory to
a USB3 port, the speed drops back to USB2 levels. I don't see how any
utility could change this, so even if HD Tune could tell what was
what, seems to me it wouldn't make a difference.

Clarification(s) requested,


There is a way to tell what "mode" a port is in. [snip]

Thanks to you and Gene for information. It is just as I thoughted: the
USB level of the memory stick or port, as the case may be, is the
governing factor.

I would have phrased it something like, "the overall speed is limited to the
speed of the slowest link in the chain."

Short, sweet, and correct.

Cool...