USB vs SATA II transfer rate

If you look at simple stats on buffer to platter transfer rates on hard drives, a typical rate is 780Mbs, and for USB its 480Mbs. I'm not convinced this reflects real world performance.

Lets say you have an external SATA II drive that has both SATA II and USB 2.0 connectors. Let's say you hook up the drive to your external SATA II port and do an RAR file transfer of 1 GB to your internal hardrive, and then you repeat the transfer using USB 2.0.

Does anyone know the REAL difference in speed you will see between the USB 2.0 and SATA II connections using the above test methods?

A friend of mine who thinks he knows it all (Masters in CS) says that the speed difference is negligible. As far as stats go, and stats alone, he is wrong-- 480 Mbps (USB 2.0) vs 780Mbs 7200 RPM SATA 150 HD transfer rates. But I'm wondering if he is more right when comparing real world testing parameters.

Thanks--

Add a reply

Register or log in to remove.

Quote :

If you look at simple stats on buffer to platter transfer rates on hard drives, a typical rate is 780Mbs, and for USB its 480Mbs. I'm not convinced this reflects real world performance.

Lets say you have an external SATA II drive that has both SATA II and USB 2.0 connectors. Let's say you hook up the drive to your external SATA II port and do an RAR file transfer of 1 GB to your internal hardrive, and then you repeat the transfer using USB 2.0.

Does anyone know the REAL difference in speed you will see between the USB 2.0 and SATA II connections using the above test methods?

A friend of mine who thinks he knows it all (Masters in CS) says that the speed difference is negligible. As far as stats go, and stats alone, he is wrong-- 480 Mbps (USB 2.0) vs 780Mbs 7200 RPM SATA 150 HD transfer rates. But I'm wondering if he is more right when comparing real world testing parameters.

Thanks--

The 480Mb/s for USB2.0 is not pour DATA. There is control packet as well as data sent.

My USB2.0 drives goes at about 25 MBytes/s. Far from my SATA hdd that goes at about 60 some Mbytes/s.

Reply to pat

Pat have you timed this using an external HD as I suggest in my initial?

I really want to take this asswipe to task on this, so I really need hard core evidence that he is wrong.

I mean the real world test would be as I've described it above, since we were talking about external HDs and I was wanting an enclosure that would mount a SATA II 3.5 drive and had both USB 2 and SATA II capabilities TO the computer. He said "Why, USB 2 is just about as fast at SATA II." So we're talking about real world transfer rates here from an internal BARE drive mounted externally enclosed in a external bare drive conversion kit (SATA to USB or SATA combo). I'm sure you both know what I'm talking about.

As a general bench mark, if you were transfering 1 GB of iinformation, what would the time difference be between USB and SATA II given the above parameters?

Reply to butthead

Yeah that's just what I want. I need it in writing froma trusted source. Thanks man.

Reply to butthead

Quote :

Pat have you timed this using an external HD as I suggest in my initial?

I really want to take this asswipe to task on this, so I really need hard core evidence that he is wrong.

I mean the real world test would be as I've described it above, since we were talking about external HDs and I was wanting an enclosure that would mount a SATA II 3.5 drive and had both USB 2 and SATA II capabilities TO the computer. He said "Why, USB 2 is just about as fast at SATA II." So we're talking about real world transfer rates here from an internal BARE drive mounted externally enclosed in a external bare drive conversion kit (SATA to USB or SATA combo). I'm sure you both know what I'm talking about.

As a general bench mark, if you were transfering 1 GB of iinformation, what would the time difference be between USB and SATA II given the above parameters?

I have 2 hdd in external enclosure (WD120 Gigs and Maxtor 200gigs) and they run as fast as USB allow.. between 22 and 25 MBytes /s.

You can seatch for motherboard review, on anandtech, for review of the a8r-mvp and some nforce4 motherboard. they test the speed of usb and sata so you'll have the proof you need.

but anyway.. usb2.0 run at 480Mbits/seconds whlie sata2 specs allow for 3.0 Gigabits/seconds, about 6 time faster interface speed..

Reply to pat

Pat, those specs can't be compared like that. We're talking about a real world test where you take information off of a hard drive and transfer that same information to another hard drive. It other words, overall transfer rate, or the time it takes to write a file(s) using USB and then compared that to SATA II, all other things being equal.

I've searched like 2 hours last night and couldn't find anything like that. I found something close, but no real tests like that.

Any links would be appreciated.

Reply to butthead

USB is quick, Firewire quicker, IDE Fast, SATA Faster, SATAII Faster still...

Message quoted 1 times

Reply to RichPLS

The exact test you are talking about is done with this enclosure. It is made by Vantec. I just bought one from zipzoomfly. I expect it early next week along with a Seagate 250GB sata drive.

Anthony

http://www.bigbruin.com/reviews05/ [...] ns3&file=1

Reply to aiannar974

Thanks nfor the link Antony. However, those are only burst speeds, which can be no doubt misleading when you are talking about the METHOD ABOVE!!!!!!!!!<--Real Time Transfer Rate. You would think someone, uh like Tom's, would do that test.

Quote :

The exact test you are talking about is done with this enclosure. It is made by Vantec. I just bought one from zipzoomfly. I expect it early next week along with a Seagate 250GB sata drive.

Anthony

http://www.bigbruin.com/reviews05/ [...] ns3&file=1

Reply to butthead

Yeah that will definitely convince a person who has their Masters Degree in CS from University of St. Luois.

Quote :

USB is quick, Firewire quicker, IDE Fast, SATA Faster, SATAII Faster still...

Reply to butthead

Quote :

Pat, those specs can't be compared like that. We're talking about a real world test where you take information off of a hard drive and transfer that same information to another hard drive. It other words, overall transfer rate, or the time it takes to write a file(s) using USB and then compared that to SATA II, all other things being equal.

I've searched like 2 hours last night and couldn't find anything like that. I found something close, but no real tests like that.

Any links would be appreciated.

listen.. I'm nothome righ now but I c a assure you that moving a file to my usb enclosure take 3-4 more time than moving the same file between 2 sata or pata hdd..

Reply to pat

Wusy,

Yeah I found that too. Too bad they didn't compare eSATA also, but I didn't check the date--perhaps it wasn't available. What I'll need to do is find the same drive tested using eSATA then find the same drive and enclosure on this essay and comapre them--however that will be near impossible unless I can find teh exact same drive, enclosure, and test program+platform.

My friend isn't going to buy anything, I am. He simply was telling me I was an idiot for wanting a eSATA connection instead of (or now with) USB 2.0, since he thinks USB 2.0 compared to eSATA is "not much faster."

This is just a war of EGO, and his is too big and he is wrong. I just want to chop him down by showing him incontrovertible evidence that he is WRONG. Plus, and I will make this coment to him, for someone in the IT business and with a Master Degree in CS, he better get his $hit together. I mean if he tells his clients that USB is almost as fast as SATA II, he is grossly misleading them.

And since this discussion we had, he said he doesn't want to talk to me about technical things anymore becsaue I don't know what I'm talking about. I've pretty much written him off as any type of friend, but revenge in this way is indeed sweet. I know you feel the same way, so let's keep looking bro.

Quote :

That is the best I found. It's between USB2 and Firewire on the same enclosure.
If you were to strip the drive out and put it on IDE it would be even faster.

If your friend still isn't convinced, then suggest him to get Firewire800 instead.

Reply to butthead

Here is another review of the Vantec Esata unit. Is this speed Real time transfer rate?

http://www.3dgameman.com/vr/vantec [...] ew_03.html

Anthony

Reply to aiannar974

Easy way to settle this is the two of you pick out a USB SATA Combo Enclosure and a hard drive. Put agreed upon data on it and then time the 2 transfers with a stop wacth. The loser pays and the winner has a new external hard drive.

Reply to g-paw

Guys,

I just bought a Lacie Two Big 1TB drive which is basically two 500GB Maxtor drives in an external enclosure with a RAID controller bolted on.
I have configured the drives as a RAID1 (mirror) setup. As I rely on this for my work (as a photographer) it would be crazy for me to choose any other option.
This provides both USB2 and eSATA connectors. In order to get the eSATA working I ended up buying the Lacie PCI-X card (as my mobo supports PCI-X which is rare. I couldn't get the eSATA port on my Asus mobo (Marvell controller) working, so gave up choosing the card instead.
A point of note for anyone buying eSATA cards - unless you get a PCI-X or PCIe card, you are limited to SATA 150.

I'm going to do some timings tonight using both the eSATA and USB2 interfaces to see the differences on both reads and writes.

Will let you know!

Cheers, Paul

Reply to athertop

Okay, I did a couple of tests tonight:
Test 1: Copying a single 5.7GB avi file between the Lacie and my internal Raptor comparing eSATA with USB2 both reading and writing to the Lacie.

Test 2: Copying 811 smaller files in 2 directories, totalling 4.07GB between the Lacie Drive and my internal Raptor comparing eSATA with USB both reading and writing to the Lacie.

Test 1 - eSATA Write 1m 32sec, Read 1m 31sec
Test 1 - USB2 Write 3m 10sec, Read 3m 39sec

Test 2 - eSATA Write 1m 00sec, Read 1m 20sec
Test 2 - USB2 Write 2m 57sec, Read 1m 51sec

Okay, the read/write (above) refer to the external Lacie drive. I used my internal Raptor drive (my temporary/swap/ PS scratch file drive) for this test as its the fastest drive I have.

I was surprised with the USB Test 2 on the read, as this was very fast compared to the other tests - not as fast as the eSATA, but a lot closer to eSATA compared with other tests.

I started off doing these tests using my internal RAID5 array (LSI Logic 300x8 hardware PCI-X controller) instead of the Raptor and soon found that whilst the read speed from the RAID5 was comparable with the Raptor, writing to the RAID5 was totally crap - I need to find out why I spent £290 on a hardware PCI-X SATA2 RAID controller and its this sloooow! Looking at Test 1 eSATA Read above - I did this using my RAID5 controller first and it took nearly 7 minutes - highlighting the RAID5 writing issue! Checking for new firmware on the LSI site now!

Anyway, it conclusively shows that eSATA is faster than USB2 in all cases above, using the same external drive. The biggest difference between eSATA and USB2 being seen when writing smaller files to the external drive - eSATA being approx 3 times faster. The least difference being the reading of smaller files from the external drives.

The onboard interfaces and other PC info are as below:

USB2 - integrated controller on Asus P5WDG2 WS Professional mobo
eSATA - Lacie PCI-X 300 4 port eSATA controller
OS: Win XP Pro x64
Mem: 4GB (4x1GB)

Hope this is useful.
Cheers, Paul
www.pjamedia.com

Message quoted 1 times

Reply to athertop

Does this help?
http://www.extremetech.com/article [...] 098,00.asp

You go smack his CS down! hehe

Reply to nobly

Check this real-world test by Tom's - it is exactly what you ask.

http://www.tomshardware.com/2006/1 [...] index.html

Actually copying files from external to internal and back, using either USB2 or eSATA interfaces with the same external enclosure and HDD. Average transfer rates are 27 to 32 MB/s for USB2, and from 61 down to 32 MB/s (depends on how full the disk is) for eSATA. MAX transfer rate specs do NOT mean much!!

See also this test of a Seagate eSATA External Drive:

http://www.tomshardware.com/2006/0 [...] index.html

It has real-world file transfer rates for external drives with IEEE 1394b (Firewire 800), IEEE 1394a (Firewire400), eSATA and USB2 interfaces, and some for internal directly-connected SATAII drives.

Reply to Paperdoc

Hey thanks for everyone who replied, even though i haven't replied in over a year here. The test exteme tech did was perfect. that pretty much throws mister I know all about everything siilicon's argument under the bus. Actually, I took this information (I found it myself after posting) and put all the documentation, including my own tranfer time test using an internal SATA 300 and eSATA 300 in a Sans Digital enclodure vs an external USB drive (A Hyperdrive model that acieves around 16MBs sustained).

When I saw this guy I opened easily telling him that i know he doesn't want to talk about computers with me becsaue "I don't know what I'm talking about because I'm stupid according to you." Then I told him I had one proposition for him: "I'll give you 5000US dollars if I'm wrong about USB vs. SATA transfer speeds, and you give me 1000US dollars if your wrong, a 5-1 odds, since I'm stupid and you seemingly kow pretty much everything about computers." I did this after several of our friends were around that knew why we weren't talking anymore, just to rub his face in it. So I said, "Do we have a deal?" He just looked like a deer that got caught in the headlights. So I aksed again, "Do we have a deal, you know, you said in front of these people a few months ago that I was stupid, so here is your chance to prove it." He just smiled and had egg all over his face. All of our mutual freinds were just staring at him. I ended the relationship with, "You're a Fing punk. Don't ever talk to mme again." I haven't spoken to him since then, nor do I even acknowledge that he exists when I see him around.

Anyway, thanks for all the responses.

Message quoted 1 times

Reply to butthead

wusy wrote :

Quote :

Let's say you hook up the drive to your external SATA II port and do an RAR file transfer of 1 GB to your internal hardrive, and then you repeat the transfer using USB 2.0.

SATAII will be clearly faster if you timed both.

One only has to look at the data rate chart to see at the beggining where 480Mb/s is clearly the bottleneck.

Megabits/Second vs Megabytes/Second, and the bus limits, sharing etc have a large impact on performance.

Message quoted 1 times

------------------------------ Q6600@3510/1560 + TT BigTyphoon+Mod
8gb Kingston 800mhz
Gigabyte EP35-DS3P
XFX 8800GT/512
Reply to apache_lives

apache_lives wrote :

Megabits/Second vs Megabytes/Second, and the bus limits, sharing etc have a large impact on performance.

Indeed bus and any other hardware limiting variable does have an impact. Tonight I transfered a 19GB ISO file from my Western Digital 500GB SATA 300 internal drive to my 160GB WD SATA 150 internal drive and got around 43MB/s.

Transferring from an external Toshiba 2.5 SATA 300 drive to an internal Toshiba SATA 300 drive (using a new HP Laptop) I achieved around 17MB/s using a USB cable. The external drive case is a HyperdriveSPACE model. I formatted the external drive from FAT to NTFS beforehand because of the FAT 4GB size limit.

So we can see even in a separate drive to drive internal transfer to a SATA 150 drive, the actual transfer rate of 187MB per second for SATA 150 is far below that in the real world. However, I'd of thought that it would be more than 50MB/s.

I used the Vista 64 transfer speed dialog box, which may or may not be accurate. But it reports in MB/s.

So in these two tests, the USB transfer about 1/3 as fast. Also, maximum transfer rates for a very good USB device is suppose to be a maximum of about 45MB second I think. This tells me that both USB and SATA transfer speeds are greatly limited by their peripheral hardware. Think about it. If we could get 40MB/s from our USB devices, we'd be transfering about as fast as our hard drives do now at SATA 100 (If calculations are fairly accurate above, which I admit something may be wrong).

Reply to butthead

What you have to understand here is that there are individual limits on each component. The entire system will transfer data as fast as the slowest component involved.

The SATA-150 interface is rated to transfer data at 150MB/sec. That is NOT the actual transfer speed. That's the rating (i.e. the upper limit). If the hard drive cannot transfer data that fast (and no hard drive does), you will not get that speed.

It's just like a fire hose that's rated to carry 150 gallons per minute. If you hook that fire hose up to a pump that can only pump 60 gallons per minute, then that's all you get. The limit of the fire hose will never come into play until you get a much bigger pump.

USB has limits due to the protocol overhead. The best external USB2 drives can only transfer about 27 MB/sec, even though the theoretical limit for USB2 is 480Mb/sec (60 MB/sec). If your hard drive can to 70 MB/sec, and USB2's cabling can do 60MB/sec, but the protocol limits it to 27 MB/sec, then you only get 27 MB/sec.

Relating that to the fire hose analogy, this is like the fire hose being able to carry 60 GPM, the pump can pump 70 GPM, but there's valves at the ends of the hose than can only allow 27 GPM through. Thus, the whole system runs at 27 GPM.

------------------------------ - SomeJoe7777

"Did he dazzle you with his extensive knowledge of mineral water? Or was it his in-depth analysis of, uh, uh, Marky Mark that finally reeled you in?" - Troy Dyer (Ethan Hawke), Reality Bites, 1994
Reply to SomeJoe7777

ever wonder why hard drives never have an internal USB connection? if USB is that fast, then they would have replaced PATA with USB long-time ago.

Reply to bitbugin

I have an external hard drive with USB2 and eSATA connectors.

file 1,048,574kB

transfer using SATA link

external SATA to internal SATA 25 seconds
internal SATA to external SATA 25 using SATA link seconds

transfer using USB2 link

external SATA to internal SATA 45 seconds
internal SATA to external SATA 30 seconds

The times are accuratish (+/- 1 second)

The above demonstrates that eSATA is faster than USB2 to a significant degree in one direction and slightly faster in the other direction.
Actual speeds will be dependant on system component parameters.

Interestingly, simultaneously (almost) transferring a 1G file from A to B and a different 1G file from B to A using the USB2 link took 60 seconds despite the noticeable chatter of the hard drive mechanisms switching from read to write.

The above "tests" were performed with an Athlon 64 X2 Dual Core 5200+ system

I then connected the external drive to an Intel Q6600 system using the USB2 link.
The recorded times for both directions were close to 33 seconds.

I could repeat the SATA link test but am waiting for a eSATA to SATA conversion cable.

Reply to rexiebaby

butthead wrote :

Hey thanks for everyone who replied, even though i haven't replied in over a year here. The test exteme tech did was perfect. that pretty much throws mister I know all about everything siilicon's argument under the bus. Actually, I took this information (I found it myself after posting) and put all the documentation, including my own tranfer time test using an internal SATA 300 and eSATA 300 in a Sans Digital enclodure vs an external USB drive (A Hyperdrive model that acieves around 16MBs sustained).

When I saw this guy I opened easily telling him that i know he doesn't want to talk about computers with me becsaue "I don't know what I'm talking about because I'm stupid according to you." Then I told him I had one proposition for him: "I'll give you 5000US dollars if I'm wrong about USB vs. SATA transfer speeds, and you give me 1000US dollars if your wrong, a 5-1 odds, since I'm stupid and you seemingly kow pretty much everything about computers." I did this after several of our friends were around that knew why we weren't talking anymore, just to rub his face in it. So I said, "Do we have a deal?" He just looked like a deer that got caught in the headlights. So I aksed again, "Do we have a deal, you know, you said in front of these people a few months ago that I was stupid, so here is your chance to prove it." He just smiled and had egg all over his face. All of our mutual freinds were just staring at him. I ended the relationship with, "You're a Fing punk. Don't ever talk to mme again." I haven't spoken to him since then, nor do I even acknowledge that he exists when I see him around.

Anyway, thanks for all the responses.

I really got a kick out of this thread, enough to register so I could give it a http://img.tomshardware.com/forum/ [...] icon14.gif

Reply to ferante

Regarding the maximum transfer rate on USB2.0:
The fastest read rates I have seen so far are in the range of 29 to 33 MByte/s from Harddisks or Memory Sticks. But ... read ...

People are talking about USB2.0 protocol limits and overhead. Does anybody know what is really going on here?

At 33 MByte/s, the USB2.0 bus is clearly NOT saturated, even accounting for protocol overhead.

I connected two Mass Storage Devices (Memory Stick or HDD) to the same port via a Hub, both reading at 31 MByte/s alone. Then I tested reading from both in parallel, everyone giving approx 22, the sum was 43.8 MByte/s.

I added a third Mass Storage Device (a little slower) and got a sum of 47.0 MByte/s.

That test clearly shows that the USB2.0 resources are not completely used by 1 device alone. The results are quite the same for Windows (XP, Vista) and Linux. I still do not see the point in USB protocol which causes this limit for a single device.

Reply to A20_user

It really sounds like you have an involuntary (probably sexual) attraction to this guy, you want him to be the bottom to your improbable top. There you are the wee guy with the small stick and no CS degree ( which obviously makes you feel inferior as per the thread contents) sliding it in slow to his welcoming brown door. Trawling the web looking for evidence that you are right, so concerned about what mutual friends will say/think.

Why dont you just get past the innuendo and kiss the guy, mutter slowly into his soft eyes about your irrestible need to be one with his sphincter, and you guys can lie back exhausted later and play with a macbook pro as your mutual friends sms you incessantly with "SO HOW WAS IT?"

that's the truth as I see it, sorry for being honest.

Reply to Anonymous

I'm running a real world test using linux to give you guys some numbers.

I'm doing a sector for sector copy of a 1 Terabyte SATA II western digital drive to another 1 Terabyte SATA II Western digital thru usb.

I'll post the results on my blog.

For the record I'm using a quad core Intel chip, Biostar board, Thermaltake USB docks (2 of them)

Right now it looks like drive to drive copy using:
ddrescue -n /dev/sdb /dev/sdc

it looks like it's going to take about 16 hours of transfer time (estimated at this point) to copy between the drives.

I'm going to try 2 different platforms and drives and give you guys some real world numbers here.

I use SATA II External Thermaltake drives for data transfer of striped array members for their ease of use and hook them up with ESATA as a rule.

It will be interesting to see the performance difference for the copy, of course copying from drive to drive amplifies difference in speed.

(My blog is at http://www.coredatarecovery.com)

Reply to coredatarecovery

Regardless of what anyone says, eSATA is extremely quicker in transfer rates than USB or Firewire.

I got fed up with trying to back up my 300GB hard drive through a USB port because it took 1 hour and 25 minutes. I bought an eSATA capable Western Digital Backup Drive and when I did the exact same back up through the eSATA connection it took 28 minutes!!! That's nearly 66% faster.

Reply to Anonymous

RichPLS wrote :

USB is quick, Firewire quicker, IDE Fast, SATA Faster, SATAII Faster still...

USB 1.1 = Slowww
USB 2.0 = Faster, but still not THAT fast
IDE = Rated slower than Firewire
SATA = WAY Faster than IDE
SATA II = Even Faster still
eSATA = faster than USB2.0 or IDE

Why they don't include a power source for eSATA ports I don't know...

Reply to linkolng35

athertop wrote :

Okay, I did a couple of tests tonight:
Test 1: Copying a single 5.7GB avi file between the Lacie and my internal Raptor comparing eSATA with USB2 both reading and writing to the Lacie.

Test 2: Copying 811 smaller files in 2 directories, totalling 4.07GB between the Lacie Drive and my internal Raptor comparing eSATA with USB both reading and writing to the Lacie.

Test 1 - eSATA Write 1m 32sec, Read 1m 31sec
Test 1 - USB2 Write 3m 10sec, Read 3m 39sec

Test 2 - eSATA Write 1m 00sec, Read 1m 20sec
Test 2 - USB2 Write 2m 57sec, Read 1m 51sec

Okay, the read/write (above) refer to the external Lacie drive. I used my internal Raptor drive (my temporary/swap/ PS scratch file drive) for this test as its the fastest drive I have.

I was surprised with the USB Test 2 on the read, as this was very fast compared to the other tests - not as fast as the eSATA, but a lot closer to eSATA compared with other tests.

I started off doing these tests using my internal RAID5 array (LSI Logic 300x8 hardware PCI-X controller) instead of the Raptor and soon found that whilst the read speed from the RAID5 was comparable with the Raptor, writing to the RAID5 was totally crap - I need to find out why I spent £290 on a hardware PCI-X SATA2 RAID controller and its this sloooow! Looking at Test 1 eSATA Read above - I did this using my RAID5 controller first and it took nearly 7 minutes - highlighting the RAID5 writing issue! Checking for new firmware on the LSI site now!

Anyway, it conclusively shows that eSATA is faster than USB2 in all cases above, using the same external drive. The biggest difference between eSATA and USB2 being seen when writing smaller files to the external drive - eSATA being approx 3 times faster. The least difference being the reading of smaller files from the external drives.

The onboard interfaces and other PC info are as below:

USB2 - integrated controller on Asus P5WDG2 WS Professional mobo
eSATA - Lacie PCI-X 300 4 port eSATA controller
OS: Win XP Pro x64
Mem: 4GB (4x1GB)

Hope this is useful.
Cheers, Paul
www.pjamedia.com

What allocation table did you use on the drives? That could make a significant difference since FAT32 and NTFS have much different cluster allocation sizes (so a text document with 1 letter in FAT32 might take up 64kb but only 4kb in NTFS). Granted this shouldn't affect the transfer speed, but if you are transferring files the files might take longer to transfer to a FAT32 formatted drive than an NTFS formatted drive.

Food for thought...

Reply to linkolng35

Become a fan

Forum Storage : Hard Disks - USB vs SATA II transfer rate