Is ATA100 really any faster than ATA66?

[OS]

I bought a promise ultra100 tx2 to replace an ultra66 during a troubleshooting session. They both work more or less equally well now. So is the 100 really any faster than the 66?

 

[risen]

I did not notice much of a difference going from 66 to 100.

 

[RIGorous1]

DATA transfer is supposedly upto 100mbps but you must remember that this is a rate of change.... thus only is prolonged sustained tests you would be able to see the difference.... ofcourse 100mbps is not constant as most ide controllers only achieve this usually in first burst situations....

 

[OS]

Just so we're all clear, all my drives do support ATA100.

 

[everman]

It has the potential to be faster, but that depends on your drives and the type of work you do. If you do very disk intensive things like video editing and have high performance HD's then you may see a difference. I doubt it would be much, and you should really be using scsi if you're doing that kind of thing a lot anyways.

 

[teqwiz]

Originally posted by: OS
Just so we're all clear, all my drives do support ATA100.

YeeHaw!

 

[Jhhnn]

The ata100 standard is supposedly better in terms of error correction and other technical details.

As I understand it, the speed difference is non-existent, simply because even the fastest drives aren't capable of using the full capabilities of either interface. If you're riding a moped, it doesn't matter if you're on a two lane road or a superhighway, it's still just a moped.....

 

[Alphazero]

Most IDE drives can't even achieve throughputs of 50 MB/s, let alone 66 or 100. I'd prefer the newer ATA-100 because it's more mature, but there is almost always no difference in performance.

 

[OS]

Originally posted by: Jhhnn
The ata100 standard is supposedly better in terms of error correction and other technical details.

Got a link or something? IMO that alone is quite noteworthy.

 

[DeeTees]

If you run a striped 4 drive RAID array using drives with 8 meg buffers you may be able to measure a difference between an ATA 66 and an ATA 100 interface. It is very unlikely that you could generate sufficient data through-put to saturate an ATA 66 interface with 2 drives of any kind.

However when you start to look at the ATA serial specifications (i.e. ATA 150) and the drive characteristics of the serial ATA hard drives you see a much higher load on the interface and the need from more headroom in burst situations. The specifications are usually not designed for todays work load but for tomorrow's extreme.

 

[Goi]

The fastest drives today have the ability to almost reach the ATA66 theoretical bandwidth of 66MB/s. Some of them achieve over 50MB/s sustained bandwidth, falling gradually to ~30GB. Burst speed is another matter. The fastest of them exceed the ATA66 limitation, and reach to around 80+MB/s. None exceed ATA100 yet though, making ATA133 pretty moot except for the ability to support really huge drives.