Google it, man!
Though short on explanation here's the math:
http://english.aopen.com.tw/tech/techinside/ata133.htm

Signature
Richard G. Harper [MVP Shell/User] rgharper@gmail.com
* PLEASE post all messages and replies in the newsgroups
* for the benefit of all. Private mail is usually not replied to.
* My website, such as it is ... http://rgharper.mvps.org/
* HELP us help YOU ... http://www.dts-l.org/goodpost.htm
> Just wondering how 133 or 150 MBs data rate can be accomplished through a
> 33MHz PCI bus...
[quoted text clipped - 15 lines]
>> Joe
>> > Any way a SATA disk can be connected to an IDE header please?
33X4B/cycle~133MBs, as always, an upper limit figure.
I expect you're right about the 150; if it's ver. 2.2, it would offer
66MHz operation, but not on the secondary, or card, bus. My chipset is
presumably 2.2 and has SATA onboard, but that's clearly the primary PCI bus.
If it happens to be PC-X 1.0, which dates from late 1999, but probably not
on any machines seen before 2001, then it could handle the 150MBs card since
it shares bus width between 32- and 64-bit devices.
My impression is that this might be more detail than ATK wanted, though.
Hope this satisfactorily amends my previous remarks.
Joe
> Just wondering how 133 or 150 MBs data rate can be accomplished through a
> 33MHz PCI bus...
[quoted text clipped - 14 lines]
> > Joe
> > > Any way a SATA disk can be connected to an IDE header please?
ATK - 05 Nov 2004 11:39 GMT
Many thanks for interest shown - I suspect cost is not worth it.
> 33X4B/cycle~133MBs, as always, an upper limit figure.
> I expect you're right about the 150; if it's ver. 2.2, it would
[quoted text clipped - 25 lines]
>>> Joe
>>>> Any way a SATA disk can be connected to an IDE header please?