** This is the quasi-official and semi-temporary T13 email list server. ** To follow up on Hale's drive performance issues....... Those of us in the drive industry and drive controller industry know that there is virtually no way to compare the data coming off the cable to the data on the media any longer. I am continuously amused when I still see drives listed in sales literature with spin rate, track-to track seek time, etc...I can guarantee that if you take two different drive firmware programmers and give them identical drives with the same electronics board and let them each generate their own firmware there will be major differences in "performance" (By performance, I am referring to what is seen as data transfers across the cable). Hale pointed out just a few of the items that go on in a drive that are no longer seen by the interface. The old days of the host telling a drive what to do are gone. The drives are calculating ramping speeds for long seeks, re-generating lost data from ECC bits, seeking to alternate tracks where "bad" sectors have been re-allocated, micro-positioning due to thermal changes, and a host of other things that a drive has to do to keep track of where it is at all times. There are no longer any headers for data, and most of the drives have proprietary methods to "increase performance" depending on the phase of the moon and the most popular OS of the day. As Hale also pointed out, firmware cheats! The Core Test benchmark was used for many years to test transfer performance. It would continually read Track 0 Head 0 Sector 0 to calculate a "performance" number. There was at least one major drive manufacturer (that I know of) that had firmware that looked for this pattern, and when it was found, it would disable all other internal operations, cache the data, and blast it across the cable as fast a possible to get a high rating. Does this have any relation to a real-world situation??? Since I was only associated to the drive industry from the controller chip support side, I am sure there are many other tricks that I don't know of, so if anyone else wants to continue this thread, go for it.... But from my point of view, "performance measurements" are like counting fairies on the head of a pin! Everyone has their own point of view! Rick Kalish -- If you have any questions or wish to unsubscribe send a message to Hale Landis, [EMAIL PROTECTED] To post to this list server send your message to [EMAIL PROTECTED] For questions concerning Thistle Grove Industries or TGI's list services please send email to [EMAIL PROTECTED]
