Re: [t13] Posted new proposal

[Frank Shu]

Since so many people started asking questions on proposal E05142r0_PRCS, I do think it does deserve a quick answer because just about all questions are regarding SMART vs. the new proposal.   SMART and the E05142r0_PRCS proposal server different purpose. OEM’s and disk drive manufactories use SMART to try catching quality issue before product shipping and after a product returned. It can identify problems like handling and warranty issues.   The new proposal proposes a design for average computer users. Neither a computer user nor the host can, or want to interpret a SMART data. For instance, what does the SMATR “The previous self-test completed having the electrical element of the test failed” mean to the user? Does it mean a harmful condition existed, or just a glitch. What does host and the user should do with it? What about “Vendor Specific” data? Plus, just for getting this information, SMART has to be enabled and self test has to be run. Then in which way you want SMART to be enabled and self test to be run is another issue which I do not want to get into now.   At the same time, we all know that the drive today has much more intelligent than SMART, and it knows its condition better than anything else. The new proposal basically says that when a drive detects a harmful condition during its real time operation, it should be able to request the host to take a proper action to correct the condition in real time. For instance, if temperature over, the drive requests the system fan be turned on. If duty cycle over, the drive requests a load balance. If servo system is just about break, the drive requests a drive backup. There must be many things we can do. We just can not let drive fail without any warring.          Thanks,   Frank Shu Program Manager Windows – Storage Device Driver [EMAIL PROTECTED] Tel  425-707-1779 Fax  425-936-7329