| 1 |
On Oct 24, 2011 7:21 PM, "walt" <w41ter@×××××.com> wrote: |
| 2 |
> |
| 3 |
> On 10/24/2011 10:28 AM, walt wrote: |
| 4 |
> > ... |
| 5 |
> > Over at least six trials on each docking station I consistently |
| 6 |
> > get 105 seconds for USB and 84 seconds for e-sata, a 5:4 ratio |
| 7 |
> > in favor of e-sata/sata over USB3/sata... |
| 8 |
> |
| 9 |
> Wow, lots of great answers, guys, thanks. Enough material to give |
| 10 |
> me lots more questions to ask you :) |
| 11 |
> |
| 12 |
> Like, for example, in theory the raw bit-rate for USB3 is more than |
| 13 |
> enough to keep up with any existing consumer hard drive, right? The |
| 14 |
> speed of usb/sata protocol translation should be very fast compared |
| 15 |
> to the speed of a spinning mechanical disk (I think?) |
| 16 |
> |
| 17 |
> Now, lack of DMA is another story for hard disks, certainly. Here's |
| 18 |
> where my ignorance of hardware limits my thinking: |
| 19 |
> |
| 20 |
> AFAIK the device driver *always* sits between the disk drive and the |
| 21 |
> DMA hardware, doesn't it? |
| 22 |
|
| 23 |
DMA means a device is told where in the system's address space it may write |
| 24 |
to, and it writes directly to that place without further CPU involvement. |
| 25 |
Since drivers run on the CPU, the drivr isn't a go-between. |
| 26 |
|
| 27 |
When the CPU *is* involved in the passing of bits around, things slow down. |
| 28 |
IIRC, that's called PIO--programmed IO. |
Archives