Refurbished Seagate, need advice

[Floyd]

Hi, I'm new on this forum.

I've just received this HDD from Seagate RMA in exchange of an older 1.5TB drive. Could someone explain this strange HD Tune graph?



Model: ST2000DM001
Part Number: 9YN164
Serial number: W240xxxx
Firmware version: CC4H


Seeing the average speed of 140 MB/s I think it's the 3 platter version (this model has another version with only two 1TB platters), but why it peaks over 220 MB/s? I saw a lot HD Tune graphs from users of this particular model but I didn't find one like this. The drive is empty and it wasn't accessed at all while doing the test.

My concern is about reliability of this drive, to know if I should fill it with data without having problems later. Is it normal?

Thanks in advance.

[BGman]

The drive is OK, but the soft is lousy....

[longlife]

what is the meaning of "soft" here...

[BGman]

what is the meaning of "soft" here...

short of software

[Floyd]

The drive is OK, but the soft is lousy....

Thanks for the answer, but do you think HD Tune is lousy? I think it is one of the best tools for Windows to test hard drives performance. If you search for benchmarks with this tool for this particular model, you get results like this one:

Two 1TB platters version, 164 MB/s average


I've made a short stroked test of my drive, measuring the first 40 gB of the drive:



Could this be the result of mixing different density platters? I mean, 667 GB platters mixed with 1 TB platter, this could explain this particular pattern of the spikes on the graph. Is it technically viable to mix different density platters on the same drive?

Thanks!

[BGman]

Most likely there was another program (or virus) running in background when you run the test.

Try on another computer and the result will be different.

[Floyd]

Most likely there was another program (or virus) running in background when you run the test.

Try on another computer and the result will be different.

Thanks for the tip but this test has been replicated in another computer and booting from a mini-XP bootCD with the same result. No programs running in the background, no antivirus real-time monitor active, HDD is empty, and I've done the same test with the boot hard drive (a 2TB Toshiba) attached to the same controller and the speed is constant (no spikes) like in the example of the two 1TB platters drive above.

Is it technically viable to mix different density platters on the same drive? Any other tool or sugestion to benchmark this?

Thanks!

[labtech]

Can use hddscan or mhdd for other benchmark tests.
In your opinion, what is the correlation between that benchmark test and the possible mix of different areal density platters?

Though I do not have a fact to back my thoughts, I believe that it would not be viable for a manufacturer to mix platters of different areal densities because that would mean the read-write heads would have to be highly flexible in their ability to adjust to various unique reading /writing adaptives. In my opinion, such flexibility is likely not possible.

[Floyd]

I thought that because a 667 GB platters drive can't reach 220 MB/s and because the short stroked benchmark shows a very particular pattern, with symmetrical highs and lows on read speed for the very first 40 gB of the drive.

But your explanation has incontestable logic, I'm not an expert in hard drives but I think that the R/W heads must suffer a lot exposed to different density platters on the same drive.

I'm testing now with HDD Scan, some zones from the start of the disk are being read at 120 MB/s, then jumps to 224 for a while and returns to 120 (like in HD Tune Pattern). I'll post my results later.

Thanks a lot!

[Floyd]

HDDScan results, short stroked to LBA 78140583

Block Size: 256 (HD Tune uses 64 Kb blocks by default).

Same pattern

Attachments

[arztt]

Do the Error Scan for some minutes and look what the Disk monitor shows you.
Or post a screenshot.

[Floyd]

Do you mean error scan in HD Tune or in HDDScan?

By the way, do you think the procedure to measure the speed and number of platters described in this article is valid? Take a look please: http://alienbabeltech.com/main/hitachi- ... lien-view/

Thanks!

[whocares]

It looks like Seagate is accepting defective drives in warranty, remapping a ton of bad sectors to extra cylinders, clearing SMART and just pushing them back to the unsuspecting.

I just got a 3TB backup plus with a weird hdtune profile, full of gaps. First I thought it was some app interfering with the measurement, killed everything I could, reran with IDENTICAL results - exact same dips at exact same places. This indicated it is really a drive characteristic, confirmed by running hdtune on another drive, which gave smooth response.

Since they just give the remainign of warranty on the exchanges, Seagate is just playing this dishonest game. Your soup has a hair in it? No problem - we take it back to the kitchen, serve to someone else. A 'recertified' soup will be right out to you.

Drives are Seagate Backup Plus 3TB, which have a ST3000DM001 Barracuda inside.

Seagate no more...

Attachments

A healthy, original 3TB backup has great data rates, smooth decay.

Same model, 'recertified' from Seagate's used car lot has lower top rate and several dips across the disk