Jump to content


 


Register a free account to unlock additional features at BleepingComputer.com
Welcome to BleepingComputer, a free community where people like yourself come together to discuss and learn how to use their computers. Using the site is easy and fun. As a guest, you can browse and view the various discussions in the forums, but can not create a new topic or reply to an existing one unless you are logged in. Other benefits of registering an account are subscribing to topics and forums, creating a blog, and having no ads shown anywhere on the site.


Click here to Register a free account now! or read our Welcome Guide to learn how to use this site.

Photo

G-sense error rate readings, are they a probem?


  • Please log in to reply
19 replies to this topic

#1 rp88

rp88

  • Members
  • 3,022 posts
  • OFFLINE
  •  
  • Gender:Not Telling
  • Local time:08:37 PM

Posted 18 April 2015 - 08:56 AM

I wanted to clarify what the "G-sense error rate" reading in the smart data of a hard drive is.

Is G-sense error rate something which indicates imminent failure if it rises or is it a normal thing which can be expected to grow at a rate of about 1 per week? This refers to the readings got from speccy but I expect other hard-drive data gathering tools also show this quantity amongst the smart data. On a new machine it started off at zero when the machine was brand new, it was at 1 after about a week, now after around 2 weeks it is at 2.

I read somewhere that this reaing acts as a log of how many times a hard drive has been knocked around whilst working, caused by things like pulling a stiff cable or USB out of the computer and the machine shaking a bit when you do or things like tilting or turning a machine while it is running, is this true or does G-sense error rate indicate prolems with the drive?

Thank You

Edited by rp88, 18 April 2015 - 08:56 AM.

Back on this site, for a while anyway, been so busy the last year.

My systems:2 laptops, intel i3 processors, windows 8.1 installed on the hard-drive and linux mint 17.3 MATE installed to USB

BC AdBot (Login to Remove)

 


#2 rp88

rp88
  • Topic Starter

  • Members
  • 3,022 posts
  • OFFLINE
  •  
  • Gender:Not Telling
  • Local time:08:37 PM

Posted 19 April 2015 - 09:29 AM

Any thoughts, Thank You.
Back on this site, for a while anyway, been so busy the last year.

My systems:2 laptops, intel i3 processors, windows 8.1 installed on the hard-drive and linux mint 17.3 MATE installed to USB

#3 signofzeta

signofzeta

  • Members
  • 421 posts
  • OFFLINE
  •  
  • Local time:03:37 PM

Posted 26 April 2015 - 04:05 AM

Important question.  Do you have a Western Digital hard drive?

 

I have a 1 year old laptop with a Western Digital hard drive.  I have been logging my G-sense error rate for my laptop, and the raw value keeps on rising, and as such the normalized value drops by the same amount.  The raw value would start at 0, while the normalized value would start at 100.  Note that in my case, the normalized value and the decimal form of the raw value must add to 100.

 

Currently, my raw value for the G-sense error rate is at 000000000056 (hexadecimal), while the normalized value is at 14.

 

I notice a few things though.  When I had windows 8, I treat my laptop like a baby.  I don't move it.  I don't bump it, nor do I subject the laptop to any physical shock.  The G-sense error normalized values still decreased regardless.  I upgraded to 8.1 and the G-sense error rate seems to have stabilized.

 

There was one stretch where the G-sense error rate for me was stable at a normalized value of 18 and 17.  Just recently, it went down to 16 because I was trying to create an image file from a CD, and it went down to 15 when I did the same thing.  It is at 14 because someone bumped my laptop.

 

What I don't understand is why the G-sense error rate sensor is so sensitive.  So basically, assuming that normalized values can never be less than 1, if I burn 100 CD's, my G-sense error rate will be at 1.

 

I have another laptop that has a seagate drive.  The normalized values doesn't drop, while the raw values increase, which I think has to do something with the fact that only larger physical shocks would probably cause the normalized value to go down.  Not my laptop with the Western Digital drive.  The normalized value would go down from a tiny vibration, which is kind of stupid.

 

Does anyone know if the G-sense error rate alone can be ignored?  As I said before, I made a log of the G-sense error rate, mostly every time I try to hibernate, shut down, or restart my computer, and every time I see that the G-sense error rate has changed.


Edited by signofzeta, 26 April 2015 - 04:16 AM.


#4 Sintharius

Sintharius

    Bleepin' Sniper


  • Members
  • 5,639 posts
  • OFFLINE
  •  
  • Gender:Female
  • Location:The Netherlands
  • Local time:09:37 PM

Posted 26 April 2015 - 04:24 AM

From Wikipedia on S.M.A.R.T.:

191 0xBF G-sense Error Rate Lower
The count of errors resulting from externally induced shock & vibration.

So G-sense Error rate depends on how you use your machine, and if the hard drive is subjected to mechanical forces or not.

#5 signofzeta

signofzeta

  • Members
  • 421 posts
  • OFFLINE
  •  
  • Local time:03:37 PM

Posted 26 April 2015 - 03:36 PM

I know what G-sense error rate means.  I am asking about the sensitivity of the sensor that causes the G-sense error rate raw values to rise.

 

For example, I treat my laptop like a baby.  I don't bump it.  The G-sense error rate would still go up.

 

My brother moves his laptop while it is on.  His G-sense error rate won't go up, unless he hits his laptop very hard or drops it.

 

The real question is, how large of a mechanical force does it trigger the G-sense error rate sensor to cause the raw value to increase by 1 unit?  For me, it was a tiny bump, or the vibration from the Blu-Ray drive that caused it.  As I said, I can burn 100 CD's, and the G-sense error rate would go from a normalized value of 100 down to 1.

 

Just like the OP and I are asking, we want to know if the G-sense error rate SMART attribute alone can be ignored, and the raw value can be super high and the hard drive would still be fine, or should we be concerned about it if it gets too high?

 

We are not asking for the definition of G-sense error rate.



#6 Sintharius

Sintharius

    Bleepin' Sniper


  • Members
  • 5,639 posts
  • OFFLINE
  •  
  • Gender:Female
  • Location:The Netherlands
  • Local time:09:37 PM

Posted 26 April 2015 - 04:00 PM

The G-sense Error Rate is not considered very important IMHO. I move my laptop around plenty (mainly taking it to school and so, both when it's on standby and when it's off) but that parameter on my SSHD has always been at baseline.

#7 signofzeta

signofzeta

  • Members
  • 421 posts
  • OFFLINE
  •  
  • Local time:03:37 PM

Posted 26 April 2015 - 05:11 PM

Ok, I check my brother's laptop's G-sense error rate.  Just to let you know, when both my brother's and my laptop were brand new, they both started with real value of 0 and normalized value of 100.  Real values would go up, and normalized values would go down by the same amount.

 

The real value of the G-sense error rate on my laptop is 86, and the normalized value is 14.  I don't even move my laptop, and I probably just only burned 2 CD's during the lifetime of this laptop.

 

My brother's laptop has G-sense error rate real value of 9, and the normalized value is 91.  He moves his laptop around while it is on all the time.

 

Explain to me this backwards logic.  Why does my G-sense error rate real value increase when I don't even move my laptop while it is on at all, but for my brother, he moves his laptop while it is on all the time, and his G-sense error rate is 1/9 of mine?

 

Forgot to add.  My laptop is 3 months older than my brother's laptop, and at January 2015, the G-sense error rate normalized value was at 18, so it had nothing to do whether my brother's laptop was newer or not.


Edited by signofzeta, 26 April 2015 - 06:15 PM.


#8 Sintharius

Sintharius

    Bleepin' Sniper


  • Members
  • 5,639 posts
  • OFFLINE
  •  
  • Gender:Female
  • Location:The Netherlands
  • Local time:09:37 PM

Posted 27 April 2015 - 03:54 AM

The normalized value is more commonly used in interpreting disk health. So I would say that your G-sense error rate value is normal.

#9 signofzeta

signofzeta

  • Members
  • 421 posts
  • OFFLINE
  •  
  • Local time:03:37 PM

Posted 27 April 2015 - 04:14 AM

So if the normalized value of the G-sense error rate is at 1, would it still be normal?

 

I also want to know why the G-sense error rate is so inconsistent across different hard drives, even of the same brand.  As I said before, I don't move my laptop at all, and my G-sense error rate increased to a real value of 86, but my brother's laptop, in which he moves all the time while it is on, has a real value of 9.


Edited by signofzeta, 27 April 2015 - 04:19 AM.


#10 SleepyDude

SleepyDude

  • Malware Response Team
  • 3,052 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Portugal
  • Local time:09:37 PM

Posted 27 April 2015 - 04:21 AM

Hi,

 

Only to add my 2 cents...

 

Maybe your laptop have some movement sensor very sensitive and it passes the information to the HDD.

 

Just recently I worked on a Toshiba machine with one of those sensors that is so sensitive that I had to disable the HDD protection software because it blocked everything when active. Didn't check the G-sense error information on the drive.


• Please do not PM me asking for support. Post on the forums instead it will increases the chances of getting help for your problem by one of us.
• Posts in the Malware section that are not replied to within 4 days will be closed. PM me or a moderator to reactivate.
• Please post your final results, good or bad. We like to know! Thank you!

 
Proud graduate of GeekU and member of UNITE
___
Rui

 
 


#11 rp88

rp88
  • Topic Starter

  • Members
  • 3,022 posts
  • OFFLINE
  •  
  • Gender:Not Telling
  • Local time:08:37 PM

Posted 27 April 2015 - 11:42 AM

signofzeta post#3: no my hard-drive is of a toshiba type, atleast that is what it says in the speccy log and in msinfo32.

alexstrasza post#4: I keep my laptop as still as is practical, that means it sits on a desk and is not moved significanty while turned on. It might shake a little bit from pressing keys and it might vibrate when I have to pull out stiff USB sticks and ethernet cables, in some cases I have had to lift it slighty to get the ethernet cable to come out. But otherwise it doesn't get moved at all while turned on.

Also I don't really understand all this business with raw, normalized, readable, real and hex values. Speccy gives a series of numbers for each SMART quantity in it's logs I have been mostly looking at the first vakue listed. The current readings from speccy are: Real value=3 Current=100 Worst=100 Threshold=0 Raw Value=0000000003 Status=Good. I contacted the shop I bought it from to tell them about this G sense error rate but they seemed rather un-concerned by it, I don't know if that is because it is not very important or because they are ot familiar with it.

Edited by rp88, 27 April 2015 - 11:43 AM.

Back on this site, for a while anyway, been so busy the last year.

My systems:2 laptops, intel i3 processors, windows 8.1 installed on the hard-drive and linux mint 17.3 MATE installed to USB

#12 Sintharius

Sintharius

    Bleepin' Sniper


  • Members
  • 5,639 posts
  • OFFLINE
  •  
  • Gender:Female
  • Location:The Netherlands
  • Local time:09:37 PM

Posted 27 April 2015 - 11:53 AM

The thing is, how a machine interprets S.M.A.R.T. value is up to the vendor - there are no *universal* standards. So each value will be different depending on the machine, and the vendors do not explain them (at least not in any documentations I've found - most of the docs I've found are from third-party monitoring software vendors).

#13 rp88

rp88
  • Topic Starter

  • Members
  • 3,022 posts
  • OFFLINE
  •  
  • Gender:Not Telling
  • Local time:08:37 PM

Posted 27 April 2015 - 12:57 PM

So do hard-drive manfacturers design HDDs to report SMART data without themselves even really knowing what the data means? Surely there is no point having SMART data collection systems on a hard-drive unless the data being collected has clear meanings which a user can understand the overall implications of.
Back on this site, for a while anyway, been so busy the last year.

My systems:2 laptops, intel i3 processors, windows 8.1 installed on the hard-drive and linux mint 17.3 MATE installed to USB

#14 Sintharius

Sintharius

    Bleepin' Sniper


  • Members
  • 5,639 posts
  • OFFLINE
  •  
  • Gender:Female
  • Location:The Netherlands
  • Local time:09:37 PM

Posted 27 April 2015 - 01:17 PM

What I mean is, the definitions of S.M.A.R.T. values are known - but how much does a value mean is up to the manufacturer.

It just means that it is best to use the vendor's own diagnostic tool if possible.

#15 signofzeta

signofzeta

  • Members
  • 421 posts
  • OFFLINE
  •  
  • Local time:03:37 PM

Posted 27 April 2015 - 02:07 PM

What about the normalized value?  For Western Digital drives, when the raw value goes up, the normalized value goes down by the same amount.  If you add the decimal form of the raw value, and the normalized value, it always adds up to 100, which is why I ask if OP has a Western Digital hard drive.

 

Seagate on the other hand, when the raw value goes up, the normalized value stays at 100, unless the raw value is huge or something.

 

Real value is really the decimal form of the raw value, which is in hexadecimal form.

 

I have a year old laptop with a Western Digital hard drive.   Real value 86.  Normalized value 14.  I don't move this laptop around and do not bump it nor do I exert any physical force or shock on it.

 

My brother has a 9 month old laptop with a Western Digital hard drive.  Real value 9.  Normalized value 91.  He carries his laptop while it is on all the time.

 

I have a 6 year old laptop with a Seagate hard drive.  Real value is 186.  Normalized value is 100.  There are some points where I exert this laptop;to some physical force, but recently, I haven't been moving this laptop around nor have I exerted any physical force to this laptop.  Real value still at 186.

 

 

Let's talk about Western Digital hard drives.  How come for 2 Western Digital drives, a small bump would cause one to have the G-sense error rate raw value to go up by one unit, and for another Western Digital hard drive, it takes a large bump to cause the G-sense error rate raw value to go up by one unit?  A sensitive G-sense error rate makes sense if the normalized value doesn't drop, but with Western Digital hard drives, every time the G-sense error rate goes up by one unit, the normalized value goes down by one unit.  If that is the case, then it should mean the sensors in Western Digital hard drives aren't as sensitive, but it isn't the case with my Western Digital hard drive.  If the normalized value is an indicator of disk health, and the normalized value of the G-sense error rate on my hard drive goes down as much as the raw value goes up, and given the sensitivity of the sensor, it wouldn't even be one and a half years before the G-sense error rate normalized value goes to 1.

 

So I ask again.  Is it a problem if the normalized value of the G-sense error rate is at 1?


Edited by signofzeta, 27 April 2015 - 02:08 PM.





0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users