Hi,

I'm trying to figure out why a 250GB WD is slow. I've run chkdsk, hdtune, and the extended SMART test, and the drive comes through fine. But it slows when writing at some points. I noticed in HDTune that the file transfer graph is pretty choppy. And larger block sizes transfer much faster than small. Why do those sectors snag, yet not record errors, and is there anything I can do about this?

Thanks much.

I tried to help in your previous thread on this drive, but unfortunately you weren't able to provide the information I requested. Therefore I'll just say that if a zerofill (which is obviously destructive to the data) doesn't improve the drive's behaviour, there is nothing more that you can do. A company that refurbishes drives, may be able to assist.

Thanks. I appreciate your input. That other thread was a different drive, actually. I've been coming up against several drives lately that have been underperforming, and wondering what I can do about it. From my elementary perspective, "massaging" the drive, for lack of a better term, has been doing some good things. By that I mean running chkdsk, seatools or a like extended self-tester. Currently I'm running repeated HDTune benchmarks on this drive and the read graph is running MUCH better.

If I understand what you've said here, I should do a backup of the drive and then run a full drive zero write. I will try that. It seems the cobwebs are being brushed off the drives sectors. Whatever it is, the precipitous drops are mostly gone, the average xfer rate is significantly up and access time, down.

Thanks again.

That's interesting new info... In that case, be aware that this drive behaviour can also have external causes (e.g. power & vibration to name just two).

To a large degree it comes with the turf - I buy laptops, many of them distressed, and refurb them. Thus it would be a huge plus for me to be able to restore the drives to a significant degree.

Ok, I overwrote zeros to the drive using DBAN. Then, using MHDD I scanned with erase delays ON, then scanned with remap ON. I didn't do anything between scans - If I was supposed to manually mark the delays for remapping, I missed that.

When I restored the image it wouldn't boot - because I had forgotten to toggle back to AHCI mode in the bios. That fixed, windows booted up. I tested the drive to find that I had actually lost ground. It was averaging around 44MB/s with a 23ms access time, on reads. The graph now had several precipitous drops into the 2MB/s area.

So I did a chkdsk /r on the drive, and that brought it back up (chkdsk certainly has been my friend). The graph is pretty linear, and it's currently averaging 47MB/s and 19ms. File transfers still appear rather choppy (all according to HDTune). The drive is a WD1600BEVT. I have a WD2500... scorpio blue that performs much better, and even it has some choppy delay problems.

So in this case results seem marginal. Unless I missed something to do with remapping those delays, I guess I'll consider trying this again on a different drive, or maybe swapping drives to isolate the drive/machine factor.

Thanks.

What's the model number?

_________________
http://www.datasaversllc.com

Sorry for the delay. The drive in question was a WD1600BEVT, don't know the suffix. That machine is gone now, so this is history. I was able to get the reads into the 50s, and that seemed adequate, so I passed it.

I'm not sure if MHDD gives extra SMART attribute info, but what I saw via HDScan and others seemed good.

So it's on to other drives. My current one has some pretty high access times that I'd like to do something about.

Thanks again.