| 1 |
On Wednesday 15 December 2010 12:41:51 Mark Knecht wrote: |
| 2 |
> On Wed, Dec 15, 2010 at 12:07 PM, Volker Armin Hemmann |
| 3 |
> <volkerarmin@××××××××××.com> wrote: |
| 4 |
> <SNIP> |
| 5 |
> |
| 6 |
> >> The logic and timing of the signals on SATA and eSATA cables is |
| 7 |
> >> (TTBOMK) intended to be identical. What those signals look like at |
| 8 |
> >> different places in the cable chain will be different. |
| 9 |
> >> |
| 10 |
> >> - Mark |
| 11 |
> > |
| 12 |
> > and you have sources to support that claims and did not just make it up. |
| 13 |
> |
| 14 |
> Ouch! Man, you like to put people in their place, don't you! :-) |
| 15 |
> |
| 16 |
> |
| 17 |
> How about: |
| 18 |
> |
| 19 |
> http://www.serialata.org/technology/why_sata.asp |
| 20 |
> http://www.serialata.org/technology/esata.asp |
| 21 |
> |
| 22 |
> Wikipedia is an interesting place to go for information but the |
| 23 |
> organization that is responsible for the spec itself is better. |
| 24 |
> |
| 25 |
> SATA is currently defined as supporting up to 6Gb/S (now Gen3) while |
| 26 |
> eSATA supports up to 1.5Gb/S (Gen1) or 3Gb/S (Gen2). Internal cables |
| 27 |
> have highly limited lengths. External cables can be much longer. (2 |
| 28 |
> meters) |
| 29 |
> |
| 30 |
> TTBOMK no one on this list is using a single drive that would exceed |
| 31 |
> 1.5Gb/S (roughly 187MB/S) so it's unlikely anyone would see a |
| 32 |
> difference in speed, but that doesn't mean the specs are the same. |
| 33 |
> |
| 34 |
> The 'logic' of what's on the cable during data transferred is |
| 35 |
> identical. The _rate_ at which it's transferred on eSATA is slower and |
| 36 |
> the electrical levels are modified to provide more reliability across |
| 37 |
> longer cables and reflections at cable/connector boundaries. |
| 38 |
> |
| 39 |
> I trust that you can read the Serial-ATA org web site yourself to get |
| 40 |
> properly educated on the matter. |
| 41 |
> |
| 42 |
> Cheers, |
| 43 |
> Mark |
| 44 |
|
| 45 |
and where do you get that internal ports can't do hotplug? |
| 46 |
|
| 47 |
From the site you just linked: |
| 48 |
|
| 49 |
Currently, most PC motherboards do not have an e-SATA connector. eSATA is |
| 50 |
readily enabled, however, through the addition of an eSATA HBA or bracket |
| 51 |
connector (as shown above) for desktop systems or with a Cardbus or Express |
| 52 |
Card for notebooks. New motherboards introduced in 2005 will start to |
| 53 |
incorporate e-SATA connectors directly, making the addition of external |
| 54 |
storage an easy option. |
| 55 |
|
| 56 |
Oh wow, all you have to do is to plug in the cable from the bracket... hmm... |
| 57 |
|
| 58 |
A little hint for you: all AHCI sata ports can do hotplugging. Every single |
| 59 |
one. The sockets are designed to allow hotplugging, so it is down to the |
| 60 |
controller and driver. And AHCI is completely fine in both regards. |
| 61 |
Some very early looks-like-ide controller chips could not do hotplugging but |
| 62 |
coldplugging (you had to tell the driver that you were offline the device), some |
| 63 |
could not do hotplugging at all - but not because of 'cheapness'. |
| 64 |
|
| 65 |
Again, all AHCI chips can do hotplugging. No matter where they are. |
| 66 |
|
| 67 |
And about esata: the only difference: cables are a bit fatter, connectors a |
| 68 |
little bit more robust. That is all. You can turn any ahci-sata port into an |
| 69 |
esata port without any problems. |
| 70 |
|
| 71 |
https://ata.wiki.kernel.org/index.php/SATA_hardware_features |
| 72 |
|
| 73 |
https://ata.wiki.kernel.org/index.php/AHCI |
Archives