In Reply to: No, John, my ears are not so "untrained" as you put it... posted by rlw on May 23, 2015 at 09:38:16:
"exactly what measurements I should be looking at"
If you take the dive, try looking at read-latency variations and powersupply current variations and spectra. I think that at least the large drives do quite a bit of wear-leveling and the "seek times" (or whatever you call it with flash) change with use and mapping and not necessarily in any sort of linear way. The controller may decide that you've got too many bad sectors and dump a block in favor of using a whole new one so they might even "improve" with time.
Sure the data is ultimately the same but unfortunately the systemic result may not be.
Just a thought...
Rick
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- RE: No, John, my ears are not so "untrained" as you put it... - rick_m 07:42:08 05/24/15 (0)