Ick. While I'll grant that the latest generation of parallel and serial ATA drives are much-improved over the older ATA drives, they still don't compare to a good SCSI bus. If you look beyond published specifications, and actually perform tests, you'll find that command queueing and better bus protocols present in SCSI products make SCSI a much better choice for real-world performance than ATA in a busy multi-tasking machine.
ATA does have some good points, I'll grant. Large amounts of inexpensive storage with reasonable performance is not a bad thing, but it's hardly the only thing to base a drive decision on.
To provide you with a real-world example that's somewhat similar, I'll tell you what I found when messing around with my ATA/66 controller in my 7600. I have a Western Digital 80GB IDE drive on the first channel of the controller, and seeing as I had the second channel free, I disassembled my QPS Firewire CD-RW, and stuck it in the CD-ROM bay on my 7600. The drive is an older unit, but still supports a read speed of 32x, and supports booting on my computer. While ripping a CD in iTunes (with a 400MHz G4), I could get a combined read/encode speed of no better than 1.7x in iTunes, and actually averaged around 1.4x. Disgusted, I pulled the drive out, put it back in the Firewire enclosure, and connected it to a port on an OrangeMicro USB/Firewire combo card. The exact same CD, the exact same CD-RW, and using firewire I averaged 3.0x, with peaks as high as 3.4x.
Now, I don't know if this is a function of the chipset in the controller, or the IDE/ATAPI drivers in OS X, but there is something not right with this picture. Note also that this was not a master and slave on the same IDE channel, but two masters on two separate channels. I'd suggest you test multiple read and write operations on multiple SCSI devices; you'll find little of that sort of performance degradation. Even multiple read and write operations on a single SCSI drive tend to be better handled than IDE, thanks to the command queueing present in the later SCSI implementations.
</rant>
Gary
On Wednesday, Jan 14, 2004, at 15:20 US/Pacific, Jesse Stanford wrote:
If you run ATA/133 or SATA/150 with a PCI controller host and a compatible drive (the same high speed interface) you can get a clear speed advantage over the build in UW-SCSI. You'll have to buy a PCI controller card (so you will need to have a free pci slot) and a compatible drive (I have an ATA/133 Maxtor 40GB it cost me 50$)
-- PCI-PowerMacs is sponsored by <http://lowendmac.com/> and...
Small Dog Electronics http://www.smalldog.com | Refurbished Drives | -- Sonnet & PowerLogix Upgrades - start at $169 | & CDRWs on Sale! |
Support Low End Mac <http://lowendmac.com/lists/support.html>
PCI-PowerMacs list info: <http://lowendmac.com/lists/pci-powermacs.shtml> --> AOL users, remove "mailto:"; Send list messages to: <mailto:[EMAIL PROTECTED]> To unsubscribe, email: <mailto:[EMAIL PROTECTED]> For digest mode, email: <mailto:[EMAIL PROTECTED]> Subscription questions: <mailto:[EMAIL PROTECTED]> Archive:<http://www.mail-archive.com/pci-powermacs%40mail.maclaunch.com/>
Using a Mac? Free email & more at Applelinks! http://www.applelinks.com
