Esata / sata speed query

mikeo

mikeo

mikeo

Soldato
Joined
17 Aug 2005
Posts
2,828
Location
SW Scotland
Hi folks

I've had a Samsung F3 1TB in an external enclosure connected via Esata for a while now.

Just replaced one of my internal drives with a Samsung F3 1TB and decided to compare speeds of the two drives via ATTO. Was expecting pretty much the same profile in ATTO. But while the internal one looks fine, the one connected via Esata looks no where as near as fast!

Esata.jpg


Is this just a case that the Jmicron controller (Esata) is not as good as the Intel one?
or is there something else that I can "tweak"?
 
wkingsnorth said:
I'm no expert, by I think e-SATA is just a bottleneck due to its age
Click to expand...
No. Even if the eSATA connector was only 1.5 Gbps, that drive isn't reaching that.

It could be to do with the controller the eSATA port is on. For example, my motherboard's built-in eSATA port (which never works but that's by the by) is on the crappy Marvell controller, whereas my case's eSATA port is connected to an Intel SATA port, so is as fast as if it were connected internally.

EDIT: Just saw the last bit of the first post...yep, that JMicron controller looks like the culprit. Grab a eSATA bracket to go on the back of your PC case (blocking an unused or missing PCI(-E) slot) and plug that into one of your Intel ports. That'll work much better.
 
Last edited:
DragonQ said:
No. Even if the eSATA connector was only 1.5 Gbps, that drive isn't reaching that.

It could be to do with the controller the eSATA port is on. For example, my motherboard's built-in eSATA port (which never works but that's by the by) is on the crappy Marvell controller, whereas my case's eSATA port is connected to an Intel SATA port, so is as fast as if it were connected internally.

EDIT: Just saw the last bit of the first post...yep, that JMicron controller looks like the culprit. Grab a eSATA bracket to go on the back of your PC case (blocking an unused or missing PCI(-E) slot) and plug that into one of your Intel ports. That'll work much better.
Click to expand...

+1 it isnt the interface, its the controller
 
Thanks guys.

I did sort of wonder if it would be the Jmicron controller. I would have thought that on a decent motherboard like I have, that they would have used something decent (assuming there is something decent).

Question... I'm assuming we are talking about:

http://www.overclockers.co.uk/showproduct.php?prodid=CB-025-AK

Emmmm. Though the only spare sata ports that I have left, is either one of the two sata-3 ports, or the 2 on the Marvell controller.
 
Only looks to be either the Akasa one, or one from startech. I'll order the one off OCuk (free P&P).

Thanks again.
 
Right.... got my Esata-sata connector.

Initially tried it on one of the Marvell 6Gb/s sata ports (as none of the Intel 3Gb/s are free). While this was a bit better than when connected via the Jmicron E-sata connector, it was still not as good as my other Samsung F3 1TB connected via the Intel controller. So swapped it to the free Intel 6Gb/s sata port and now get:

EsataIntel6Gbs.jpg


Pretty close. I'll assume the difference may be something to do with the amount of data on each drive.

Just out of interest I ran HDtune and got:

HDtuneSamsung.jpg


Now I'm really confused!!!!! The external drive doesn't look so good here! I was rather expecting for it to look pretty similar to the other F3 drive that's fitted internally! I'm starting to wonder if the bit of circuitry in the external enclosure makes a difference??? But why ATTO makes it look OK but HDtune makes it look like it's still running slower than the internal drive remains a mystery to me.

Any ideas . suggestions anyone? (besides "stop running speed tests"!).

Both F3's appear to have the same firmware.
 
You must log in or register to reply here.
Back
Top Bottom