• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Serial ATA

mrman3k

Senior member
When is Serial ATA comming out? I really want to know about this information because ideally I want to do my big motherboard/CPU upgrade only once Serial ATA is out and integrated onto the motherboard. Any help, opinions, knowledge will be appreciated.
 
another good 6-7 month for implementation and 9-10 month for a good integration and end of year to appear... dont know if it can be any faster.. but i'm waiting too...
 
What would make Serial ATA better?

Something seems inherently better about sending multiple bits at once instead of just one. 😛
 
SerialATA could be better for the same reason that a P4 can be better than a PIII. With a PIII, you get a lot of work done every clock cycle. With a P4 you get less work done, but you're able to ramp it up to a LOT more clock cycles. With SerialATA it's the same thing.

Also SerialATA has support for more devices on a single chain, and a few other things that make it almost SCSI-like. (SerialSCSI is in the works too I believe.)
 
One cable to the HD, and a thin one at that.

No jumpers, no bigass flat cables, no more of the current crappy powercables that sometimes get stuck in the HD.
 


<< Something seems inherently better about sending multiple bits at once instead of just one. 😛 >>



It may appear that way on the surface but there are limitations to the speed at which a parallel bus can run. As speed increases the effects of the smaller bit time (how many nSec the bit is on the bus) combined with the cable capacitance make the ability of the parallel data to all arrive at the receiver at the same time suspect. The data may leave all nice and aligned at the silicon in the chip but then it has to go thru the IO pad, down the cable and then into the receiver. At that point if it is not aligned then you get garbage instead of data. Going serial eliminates this alignment issue (also called signal skew) and lets you run MUCH faster. The logic is harder and bigger but its just silicon...

Serial ATA is going to be a winner in the desktop space. But IMHO it is not going to replace SCSI since it is using the same basic HDA that the current parallel ATA drives are using. Nobody wants to displace the much more profitable SCSI drive with a cheaper comodity product... thats business suicide. Now folks are talking about a eSATA drive which will be more like SCSI but at a more SCSI like price tag.
 
Okay, I have visited Serialata.org many times, but I guess I will still be upgrading with standard IDE and only get Serial ATA with the K8 or whatever platform a good year or so from now.
 
The first versions of SerialATA won't perform much better than ATA133 anyway. And the drives themselves will still be the same slow-ass things (relative to the available bus speed). The only advantage would be the increase in the number of drives on the channel (I don't think SerialATA allows multiple devices to be active at once, like SCSI, just more devices, so you still only have one drive attempting to use the bandwidth, unlike SCSI where several drives could easily flood the bus). The one advantage SerialATA will have over ATA100 is the larger capacity, but ATA133 gets around that already.
 
"When is Serial ATA comming out? "

Adaptec just announced their controller will go in to production in the 4th quarter. So frocesho's estimate of end of the year is probably on target. MB integration probably won't be until 1st or 2nd quarter next year, so I wouldn't put off an upgrade waiting for it.
 
Back
Top