I find it hard to understand what benefit there could be for a HDD manufacturer to deliberately make a HDD that virtually builds in a weaker than normal bit errors correction. Why wouldn't a bit error that can be corrected not be corrected.
I could understand a drive spreading video over the entire set of cylinders or not stopping the motor to save startup wear or running the motor slower where higher rotation speed is not required. But not correcting errors? That sounds implausible.
Because it's cheaper both in terms of the few dollars they save in manufacturing costs and the time penalty it takes to attempt correction.
Complete bollocks...
I don't know what the difference between these drives is...
Would you care to explain why you think it's complete bollocks (since you basically admitted in the same post that you don't know what the difference is)?
Perhaps my 17 years in IT may add a little weight to the discussion; and I don't just mean I answered IT helpdesk calls for the past 17 years, my expertise includes (among other things) server, NAS and SAN builds and administration for (at the time) one of Australia's top three largest telcos and forensic data recovery from both magnetic and solid state media (or course I'm not going to divulge what I did specifically, but for the purposes of this forum, consider me well versed).
Unless you believe the marketing wank, these drives are nothing special. If you have a look at the Western Digital AV-GP drives as an example, they are marketed as being "firmware optimised for write-intensive operations". In other words, it utilises the
ATAPI Steaming Feature Set. First line straight out of the ATAPI documentation:
The Streaming feature set allows a host to request delivery of data within an allotted time, placing a priority on the time to transfer the data rather than the integrity of the data.Specifically the sections we are interested in:
4.23.2.4 - The READ CONTINUOUS bit in the read stream commands (see 7.30.3.3) specifies that the device shall transfer the requested amount of data to the host within the time specified by the DEFAULT CCTL field (see 7.4.3.4) or the COMMAND CCTL field (see 7.30.3.2) even if an error occurs. The data sent to the host by the device in an error condition is vendor specific.4.23.2.5 - The WRITE CONTINUOUS bit in the write stream commands (see 7.69.3.2) specifies that the device shall transfer the requested amount of data from the host within the time specified by the DEFAULT CCTL field (see 7.4.3.4) or the COMMAND CCTL field (see 7.30.3.2) even if an error occurs. If the device is unable to resolve an error within the time specified by the DEFAULT CCTL field or the COMMAND CCTL field, the erroneous section on the media may be unchanged or may contain undefined data. A future read of this area may not report an error, even though the data is erroneous.As I said in my previous posts, if this sort of drive behaviour is something you don't care about (such as in the case of streaming video that isn't used for broadcasting or post-production purposes), then this drive might be quite alright. If you're thinking about storing anything critical on it, you run the risk of corrupted files.
You'll also note that in their specifications, their MTBF figures are quoted based on hard errors (which I think most manufacturers are). You could have endless amounts of soft errors that go uncorrected.
There is a reason why these drives are specifically targeted at low-end CCTV systems and consumer streaming NAS type devices.
It basically comes down to the old mantra: FAST, GOOD, CHEAP; Pick two.