OCZ Vertex3 - HUGE raw read error rate

How, what, where and why - when using the software.
rgrosz78
Posts: 3
Joined: 2012.01.31. 21:12
Location: Atlanta, GA

OCZ Vertex3 - HUGE raw read error rate

Post by rgrosz78 »

I installed HD Sentinel Pro yesterday. It informs me that my OCZ Vertex3 has excellent health. But the values shown for "raw read error rate" look BAD. Yesterday the value was about 104 million, and today the value is 131 million.

If these values are correct, my drive health should not be "excellent". Everything has been running fine on my system since I updated the OCZ firmware on 12-18-2011. Is it possible that the "raw read error rate" values are incorrect?
Attachments
OCZ Vertex3 01-31-12.jpg
OCZ Vertex3 01-31-12.jpg (434.97 KiB) Viewed 18454 times
Life is too short to drink bad wine (or bad coffee)
rgrosz78
Posts: 3
Joined: 2012.01.31. 21:12
Location: Atlanta, GA

Re: OCZ Vertex3 - HUGE raw read error rate

Post by rgrosz78 »

You have to scroll that prior image down to see that the same numeric value is shown for three other items:

On-the-fly ECC Uncorrectable error count
Uncorrectable soft read error rate
Soft ECC correction

This is NOT true of the other hard drive in the same PC ... but it does not have the same data items to display.
Attachments
WD10EALX 1TB 01-31-12.jpg
WD10EALX 1TB 01-31-12.jpg (444.93 KiB) Viewed 18452 times
Life is too short to drink bad wine (or bad coffee)
User avatar
hdsentinel
Site Admin
Posts: 3128
Joined: 2008.07.27. 17:00
Location: Hungary
Contact:

Re: OCZ Vertex3 - HUGE raw read error rate

Post by hdsentinel »

Thanks for the images and excuse me for the confusion.
First of all, I can confirm that no, there are no problems with SSD in general and with the "raw read error rate" attribute and yes, the displayed value is correct.

As the name of this attribute indicates (".... RATE") it does not show the count of errors or events for this device.
In contrast, you may see counter attributes (".... COUNT") which really reflect the count of events (for example start/stop count) or count of errors (eg. retired block count).

There are some statistical attributes (like this one) where the "Data" field contains statistical information, for example the total number of read operations, instead of the count of errors. The "Data" for these attributes can increase to 100 millions and have huge jumps / sudden drops - and these are completely normal. This is described in the Help also. For these attributes, the "Value" and "Threshold" fields (and their relation) determines how "good" is the status.

Of course Hard Disk Sentinel checks all attributes and if any problem found with this (or other) attribute, it immediately show the problem in both the text description and the actual health % value.
User avatar
hdsentinel
Site Admin
Posts: 3128
Joined: 2008.07.27. 17:00
Location: Hungary
Contact:

Re: OCZ Vertex3 - HUGE raw read error rate

Post by hdsentinel »

The S.M.A.R.T. specification allows the manufacturers to use the "Data" field as they prefer.
As you can see, WD drives do not show high numbers there - this may caused the confusion.
However, Seagate drives work the same way as the OCZ Vertex3: the "Data" field of these statistical attributes can be high on them.
(This is discussed on this forum previously about Seagate drives).

In general, directly comparing 1-1 such attribute of different hard disks (or a hard disk vs SSD) can cause false conclusion,
this is why Hard Disk Sentinel is developed: to properly report the error(s) reported by different devices of all manufacturers.
sligionse
Posts: 1
Joined: 2012.01.17. 13:52
Location: Germany

Re: A question ...

Post by sligionse »

Now I understand, it is not difficult.

I think after the log-in, it is under "view active topics".

;-)
rgrosz78
Posts: 3
Joined: 2012.01.31. 21:12
Location: Atlanta, GA

Re: OCZ Vertex3 - HUGE raw read error rate

Post by rgrosz78 »

hdsentinel wrote:Thanks for the images and excuse me for the confusion.
First of all, I can confirm that no, there are no problems with SSD in general and with the "raw read error rate" attribute and yes, the displayed value is correct.

As the name of this attribute indicates (".... RATE") it does not show the count of errors or events for this device.
In contrast, you may see counter attributes (".... COUNT") which really reflect the count of events (for example start/stop count) or count of errors (eg. retired block count).

There are some statistical attributes (like this one) where the "Data" field contains statistical information, for example the total number of read operations, instead of the count of errors. The "Data" for these attributes can increase to 100 millions and have huge jumps / sudden drops - and these are completely normal. This is described in the Help also. For these attributes, the "Value" and "Threshold" fields (and their relation) determines how "good" is the status.

Of course Hard Disk Sentinel checks all attributes and if any problem found with this (or other) attribute, it immediately show the problem in both the text description and the actual health % value.
Thank you for the explanation. I thought those values might mean something else, and you confirmed that.

I assumed HD Sentinel was correct about my SSD, but I just wanted to be sure. The OCZ Vertex has a "bad rep" :geek:
Life is too short to drink bad wine (or bad coffee)
Post Reply