"IDE = consumer market = semi-experimental & number-driven - newest fastest technology pushed out ASAP
SCSI = business/enterprise market, restricted to technology which has proven reliable (proven means its been in use for at least a couple years, and is pretty much outdated relative to whats getting put in the IDE drives with the ridiculously fast rate drives have been improving lately)"
It's pretty much the exact opposite of this. SCSI is purely driven by performance and reliability, ATA is purely driven by cost. SCSI uses all the latest technologies, that's why it costs so much, manufacturers have to recoup the costs of R&D. When manufacturers figure out how to produce the technology cheap enough to mass produce it, it trickles down to ATA.
"Numbers to back this up, specific to your situation:
(media transfer rate = physical disk <-->internal buffer, not buffer<--> SCSI/IDE bus)
Seagate 15K 36GB Cheetah peak media transfer rate : 69MB/s (552Mb/s)
WD1200JB (120GB) peak media transfer rate: 75.25MB/s (602Mb/s)
WD2000JB (200GB) peak media transfer rate: 92MB/s (736Mb/s)"
You're comparing apples to oranges. The WD numbers are not media transfer rate (ITR-Internal Transfer Rates) values. The WD number is some bogus buffer to media transfer rate which is a pretty much a useless stat, not that ITR is any more useable in comparing drives. Since WD doesn't quote ITR, you have to use a Seagate to Seagate comparison:
Barracuda V ATA: 570Mbits max
Cheetah 15k.3: 688Mbits max (formatted)
Note that the Cheetah number is formatted speed, which is significantly lower than the ZBR value which is 891Mbits. It's not stated whether the Barracuda number is formatted or not, so the gap might be quite a bit larger. All these numbers are ITR values which in the real world are useless compared to STR numbers. Let's take your original comparison and use realworld STR numbers:
Seagate X15-36LP: 60.5MB-45MB/s
WB 1200JB: 48.8MB-29.2MB/s
Seagate 15k.3: 76.4MB-51.1MB/s
That's a king sized butt whooping especially at the end of the drive by the even the old Seagate. The new one is faster at its slowest point than the WD is at its fastest point.
"Hint: if you use a 36GB partition on that 200GB drive, the seek times are lower because it never has to go full stroke, bringing the access times pretty much in line with the Cheetah."
That's bogus. The 2.16ms advantage the Seagate has in average latency alone will make sure the WD can't compete. Even without that, the 15k drives use much smaller platters (2.6in vs 3.5in), which means that worst case full stroke seeks are still better than even the average seek time for an ATA drive. Then add on the latency advantage and you see where this is going...
"Yes, that's right, I'm saying the IDE drive you already own is technically faster than high-end SCSI."
Not on this planet. Take a look at Storage Review and compare the 15k.3 to the WD 1200JB, it's not a pretty sight, with the Seagate winning a number of the benchmarks by more than 100%.
"Worth the cost? I would definately say no."
For most people, probably not, but don't make wild claims that ATA is the superior technology, in no uncertain terms, it isn't.