S.M.A.R.T. - where do I find out the Western Digital meaning of the values

I am thinking about buying a refurbished harddrive and I found out the smart values of the drive. I also found out that manufacturers determine what the values mean.

Their “(ATTRIBUTE_NAME) FLAGS-VALUE- WORST-THRESH-FAIL-RAW_VALUE” can hava a slightly different meaning depending on the maker of the drive. Where does WD keep this info?



Hi there and welcome to the WD community.

In order to check the SMART values of the drives we have a tool called DLG which can be obtained on the download section of the WD website.

Here are some links that might help you with your questions:




And here is a link to the DLG software download:


There you will see a Dos version and a Windows version. 

See the following thread:


A WD user asks WD Tech Support for the meaning of the Raw Read Error Rate SMART attribute. The support person is unable to answer the question, so it is escalated to Engineering. Engineering are unable to find an appropriate WD technical document, so they plagiarise one that they find on the Internet. It turns out that the plagiarised document is one of mine. Worse still, my article was written about Seagate’s unique SMART attributes and does not apply to WD’s drives.

@Khahhhd, I received an email, with a case number, from “Brandon” at WD Tech Support. I don’t know why, because I didn’t raise a support case. Perhaps it was meant for you?

Anyway it didn’t contain any information pertinent to your question, merely a reiteration of the link to WD’s SMART page and a reference to Wikipedia’s article on the subject:


Perhaps if you gave WD’s support people a clearer indication of the kind of information that you are looking for, then one of them may delve into the SMART specification that undoubtedly exists somewhere within the bowels of the organisation. Presumably WD’s firmware programmers would have access to this specification, or do they get their information from the Internet as well?

Here is a hypothetical example that might help them to understand the question:

Attribute name Current value Raw value

Raw Read Error Rate 120 200,000,000
Seek Error Rate 60 1,000,000

If a Seagate drive were to report the above attribute values, it would be 100% healthy and error free, whereas a WD drive with those same values would be terminally ill.

BTW, I apologise to those people who have been trying to contact me via PM. After receiving three “private” messages which had been edited by WD’s moderators, I no longer check my mailbox.

Hello “fzabkar”.

I guess it “was” misdirected because as you said, the info given was something really generic and didnt answer the question. Whats even more questionable was that I sent a text file highlighting the values that pertained to my question.

You must be a bit sensitive to my dilemmna because the example you stated is where my questions are. Here are the values that I sent:


  1 Raw_Read_Error_Rate     POSR--   200   200   051    -    0

  3 Spin_Up_Time            PO----   182   182   021    -    7858

  4 Start_Stop_Count        -O–CK   100   100   000    -    18

  5 Reallocated_Sector_Ct   PO–CK   200   200   140    -    0

  7 Seek_Error_Rate         -OSR--   200   200   051    -    0

  9 Power_On_Hours          -O–CK   044   044   000    -    41236

 10 Spin_Retry_Count        -O–C-   100   253   051    -    0

 11 Calibration_Retry_Count -O–C-   100   253   051    -    0

 12 Power_Cycle_Count       -O–CK   100   100   000    -    16

192 Power-Off_Retract_Count -O–CK   200   200   000    -    13

193 Load_Cycle_Count        -O–CK   001   001   000    -    4456930

194 Temperature_Celsius     -O—K   128   100   000    -    24

196 Reallocated_Event_Count -O–CK   200   200   000    -    0

197 Current_Pending_Sector  -O–C-   200   200   000    -    0

198 Offline_Uncorrectable   ----C-   200   200   000    -    0

199 UDMA_CRC_Error_Count    -OSRCK   200   200   000    -    0

200 Multi_Zone_Error_Rate   —R--   200   200   051    -    0

                            ||||||_ K auto-keep

                            |||||__ C event count

                            ||||___ R error rate

                            |||____ S speed/performance

                            ||_____ O updated online

                            |______ P prefailure warning

As you can see, the load cycle count is crazy high. I did a little detective work and found that load cycle count is attributed to the number of times the hd head parks itself to save power. On a green drive this time based feature can be as low as 8seconds. After a little more investigation I found that WD has a small program that can reset this time to something more realistic like 5 minutes. I did the calculations and figured out that if the drive had a original load cycle (count) time of 5 minutes that the drive would have a load cycle count of about 530K, still high according to WD’s numbers…but on another manufactures drive thats an acceptable number especially since the drive was running linux which does the park hd heads thing under the os’s commands in addition to the drives self maintenance.

When you add the two together do you get an hd with an artifically high load cycle count and the drive is ok, or is the load cycle thing a purely mechanical count and the drive has parked its heads 4.5 million times and still working!

That would also mean that wd’s own numbers are really, really low considering that they state that a load cycle count of around 300K is where you should start paying attention to the drives reliability. 4.5 million could mean that the drives reliability is about 15 times better then they (know about?) state, especially when the other important values are acceptable. Maybe time on is high, but the drive was used in a backup server that backup’d, read/wrote data once a day, as was told to me.

So we know that the person answering my question did’nt have the experience necessary to answer my question. I even asked if I could e-mail a tech that may better know how to answer the question.

As of now, the refurbished drive is a misnomer, I should have said a pulled drive from a working environment, but the geek in me wants to know what these number mean if a drive has done the park hd heads thing 4.5 million times but the threshold is 300K.

Concerning the power on hours, I was planning to use the drive as storage for videos to move stuff off my desktop, so wasnt largely concerned by those numbers - more (or much more) than 1/3 a year of 24/7 on time safely remains. But now the issue has changed and the drive is off my list because I know that there will be a day where I leave the now external drive connected, 120 days later…anyway.

But the puzzeling questions about the drive remains to be answered.


I notice that some Scorpio Blue drives are rated for 600,000 load/unload cycles under “controlled ambient conditions”. Some Caviar Greens are rated for 300K cycles.


That doesn’t necessarily reflect the maximum count allowed for by SMART, though.

You didn’t mention the model of your drive, so I haven’t been able to compare your results against an identical drive. However here is a WD example:


9 Power On Hours 94 4795
193 Load Cycle Count 181 59701

The normalised value of the Load Cycle Count has dropped from 200 to 181. The current normalised value could have recently dropped from 182 to 181, or it could be about to drop from 181 to 180. This enables us to calculate the range for the maximum rated cycles.

59701 / (200 - 181) x 200 = 628 432
59701 / (200 - 180) x 200 = 597 010

The results would suggest that the rated figure is 600K.

Similarly, assuming that the initial normalised value for Power On Hours is 100, we arrive at the following results:

4795 / (100 - 94) x 100 = 79 917 hours = 9.12 years
4795 / (100 - 93) x 100 = 68 500 hours = 7.81 years

That would suggest an 8 or 9-year 24/7 lifetime.

In the case of your drive, it would appear that its remaining lifetime (according to SMART) is …

41236 / (100 - 44) x 44 = 32 400 hours = 3.7 years

Hello again.

here’s the model #:

Model Family:     Western Digital Caviar Green
Device Model:     WDC WD10EACS-00ZJB0
Serial Number:    WD-[[Deleted - Trancer]]

I see that there is a lot more to understanding smart values then I realized. So…the values arent supposed to be taken directly. Raw value is not to be understood as a linear value its more a value to be weighed into the threshold.

How can normalized value change and still be valuable information to understand how healthy a drive is?

How can anyone understand load cycle count if the numbers that it displays arent directly related to what the data says its supposed to be related to.

Is there anyway to get a relationship with the numbers without using algebra to make sense of it all? 

uuh boy. I think I gotta do some more reading before I get comfortable using the values to gain an understanding of a drive, well other than linear values - which should be highlighted so a person understands.

So what does 4million load cycles mean when looking directly at the data?

My apologies to the helpdesk person that was unable to point me in the right direction. I have more information now and am as confused as I was when I first asked for a solution.

Maybe more.

ISTM that your drive really has recorded 4.5 million load/unload cycles. IMO your results should allay any concerns that users might have about this particular feature. The normalised value of the attribute is sitting at 1 and will probably remain just above the threshold for the remainder of the drive’s life. I suspect this is to avoid triggering a SMART failure were the attribute allowed to fall to 0.

Personally, I would be very happy to have a drive that has lasted as long as yours and has kept a clean sheet during that time. My only beef is that HDD manufacturers keep their vendor specific SMART specifications to themselves. This means that most, if not all, of the publically available SMART information has been derived from speculation and experimentation. The Wikipedia article seems to be about the best there is, but it is still basically inadequate. As you say, not all raw or normalised values are linear. Some attribute data are logarithmic (eg Seagate’s), while others are comprised of two or more hexadecimal values (eg Hitachi).

To take another example, notice that your Temperature_Celsius attribute has a normalised value of 128 and a raw value of 24. My guess is that the normalised value is calculated as follows:

Current value = 152 - Raw value

… where the raw value is the actual temperature in degrees C and the Current value is a health score chosen so that a value of 100 reflects 52 degC. In this way temperatures lower than 52C score higher than 100 and vice versa.

I should add that some WD models record additional SMART attributes which are not reported to the user. IIRC, these include uncorrectable error counts and total LBAs read and written.

You can retrieve the drive’s firmware modules (including SMART modules 20 - 26?) using the demo version of SeDiv:

SeDiv WD Read ROM & Modules:

I have written a program to extract these SMART attributes, including the hidden ones:

Extracting SMART data from WD MODs 20 - 26: