All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 24 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Surface test of 5 HDDs.
PostPosted: February 14th, 2011, 22:39 
Offline

Joined: November 8th, 2010, 18:07
Posts: 68
Location: Earth
I tested 5 hard drives using different tests (mainly surface access times tests):

1. Samsung HD204UI in Verbatim 2 TB 3.0 USB enclosure.
2. Samsung SP1614N
3. WDC WD6400AAKS
4. WDC WD1600BEVS
5. Seagate ST320410A

Attached to this message are zipped 149 files with screenshots of these tests.

Your comments, as well as the results of your tests are welcome.


Attachments:
HDDs_surface_scans.zip [7.52 MiB]
Downloaded 1295 times
Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 15th, 2011, 2:11 
Offline

Joined: December 12th, 2010, 9:50
Posts: 121
Hi great initiative,
I request to include latest breed of Hitachi as well.


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 15th, 2011, 7:38 
Offline

Joined: November 8th, 2010, 18:07
Posts: 68
Location: Earth
1. Generally Samsung drives (even the new 2TB Samsung HD204UI) seem to have considerably poorer quality of the surface comparing to the 3 years old (or so) Western Digital WDC WD6400AAKS.

2. I would like to see similar tests of other brands 2 TB hard drives. Please upload your tests results in this thread.

3. I wonder why in some hard drives slower access times can bee seen aligning in some diagonal patterns (in HD204UI) and why in some other drives (in WDC WD1600BEVS) a checkered pattern appears (only from the second half of the scan) which "shape" changes regularly depending on the part of the disk surface scanned, while in some other drives no such regular patterns are present?

4. Why is there such a difference in SMART status health notification between various software for Samsung SP1614N (it is red for most of the tests in HDD Health and everything OK in HD Tune)? There is also a significant difference in MHDD scan result and HDDSCan result for SP1614N.

5. BTW, it took ca. 31 h to test Samsung HD204UI in Verbatim 2 TB 3.0 USB enclosure on USB 2.0 port.


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 15th, 2011, 11:36 
Offline
User avatar

Joined: September 29th, 2005, 12:02
Posts: 3561
Location: Chicago
dfr wrote:
1. Generally Samsung drives (even the new 2TB Samsung HD204UI) seem to have considerably poorer quality of the surface comparing to the 3 years old (or so) Western Digital WDC WD6400AAKS.

Your guess is incorrect
It's because you don't know how hard drives work

1. Read this article - http://hddscan.com/doc/HDD_Tracks_and_Zones.html
2. Look on the last picture in the article. Samsung uses from 1 to 20 million sectors per a serpent chunk. WD uses from 35000 to 65000 sectors per a serpent chunk

_________________
SAN, NAS, RAID, Server, and HDD Data Recovery.


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 15th, 2011, 12:05 
Offline

Joined: November 8th, 2010, 18:07
Posts: 68
Location: Earth
Doomer wrote:
dfr wrote:
1. Generally Samsung drives (even the new 2TB Samsung HD204UI) seem to have considerably poorer quality of the surface comparing to the 3 years old (or so) Western Digital WDC WD6400AAKS.

Your guess is incorrect
It's because you don't know how hard drives work

1. Read this article - http://hddscan.com/doc/HDD_Tracks_and_Zones.html
2. Look on the last picture in the article. Samsung uses from 1 to 20 million sectors per a serpent chunk. WD uses from 35000 to 65000 sectors per a serpent chunk


So how to interpret those maps?
And what about those patterns mentioned in p. 3? Can you please explain their nature?


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 15th, 2011, 12:44 
Offline
User avatar

Joined: September 29th, 2005, 12:02
Posts: 3561
Location: Chicago
dfr wrote:
So how to interpret those maps?

Can you please read the article?

_________________
SAN, NAS, RAID, Server, and HDD Data Recovery.


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 15th, 2011, 19:00 
Offline

Joined: November 8th, 2010, 18:07
Posts: 68
Location: Earth
Doomer wrote:
dfr wrote:
So how to interpret those maps?

Can you please read the article?


I have just finished reading it, and I also read this: http://hddscan.com/doc/HDD_from_inside.html
but unfortunatelly those articles did not help me to answer the questions I raised.

What I and certainly some other users of HDDScan would appreciate is if you or somebody else would make a FAQ or Help for HDDScan explaining on examples (I allow you to use my screenshots for that purpose) different scan results / different patterns, which can be seen on the maps of the scans of different drives. Without that many users would be clueless as to the meaning of the results, which the maps show.

Also please explain why you say that "Signal B should be read with half amplitude considering increasing amplitude when head moves to the left off the track. Signal C should have full amplitude and Signal D should have amplitude close to nothing."

Looking at the picture I thought that B should have amplitude close to nothing, and D should have half of the amplitude, as B is the least close to Servo N track. If the picture and description is correct then apparently I do not understand something.

I also did not quite understand (or "feel" or "imagine") the last picture...


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 15th, 2011, 22:11 
Offline
User avatar

Joined: September 29th, 2005, 12:02
Posts: 3561
Location: Chicago
Read "Servo Track N" as "Servo Track N+2" sorry for typo

_________________
SAN, NAS, RAID, Server, and HDD Data Recovery.


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 15th, 2011, 22:13 
Offline
User avatar

Joined: September 29th, 2005, 12:02
Posts: 3561
Location: Chicago
dfr wrote:
I also did not quite understand (or "feel" or "imagine") the last picture...

Perhaps you should ask questions then
Make them narrower to the subject

_________________
SAN, NAS, RAID, Server, and HDD Data Recovery.


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 16th, 2011, 6:04 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15440
Location: Australia
dfr wrote:
4. Why is there such a difference in SMART status health notification between various software for Samsung SP1614N (it is red for most of the tests in HDD Health and everything OK in HD Tune)? There is also a significant difference in MHDD scan result and HDDSCan result for SP1614N.

The differences are due to subjective interpretations and misinterpretations by the authors.

My preferred SMART tools are HD Sentinel and smartmontools. Both utilities will output the full 48-bit values of each SMART attribute for HDDs, and 64-bits for SSDs. Some utilities, including HDDScan, will hide the actual raw values, preferring to display only the author's interpretation. HDDScan doesn't often get it wrong, but ActiveSmart is one utility that doesn't understand Seagate's counterintuitive attributes, for example. HD Tune's author restricts the output to the lower 32 bits, and some of the attributes display negative values for counts.

In the case of HDD Health and your Samsung SP1614N drive, the T.E.C. (Threshold Exceeded Condition) warnings are estimated by monitoring the rate of decline of the normalised attribute value over a period of time. Some warnings are misleading (eg. attributes 08, C4, C6) because they are the result of the attribute values changing from their initial values of 253. I suspect that a value of 253 signifies a "virgin" attribute. I'm also speculating that it is only after such an attribute sustains its first error that its value falls within the normal 0-100 range. Hence a change from 253 to 99 really only amounts to a loss of 1 point, not 154.

_________________
A backup a day keeps DR away.


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 16th, 2011, 7:23 
Offline

Joined: November 8th, 2010, 18:07
Posts: 68
Location: Earth
fzabkar wrote:
dfr wrote:
4. There is also a significant difference in MHDD scan result and HDDSCan result for SP1614N.

The differences are due to subjective interpretations and misinterpretations by the authors.


And what about the differences in map results in HDDScan and MHDD?

This is someones post from another forum, which confirms both that HDDScan shows more bad times than MHDD, and that HD204UI has more bad times than WD and Hitachi:

"I don't know if HDDscan timing results can be trusted because when I ran it on a 2TB Hitachi 5K3000, it reported dozens of >150ms sectors, but under MHDD the number dropped to 0-7 per pass, compared to 1-3 for my Samsung HD204UI.

OTOH my 2TB WD did as follows:

<3ms 14672666
<10ms 648967
<50ms 51
<150ms 0
<500ms 0
>500ms 0

A Hitachi 1TB 7K1000.C also did well:

<3ms 7649978
<10ms 6630
<50ms 4272
<150ms 4
<500ms 0
>500ms 0


Samsung 2TB HD204UI:

<3ms 14951778
<10ms 368563
<50ms 1293
<150ms 48
<500ms 2
>500ms 0 "

http://hardforum.com/showpost.php?p=103 ... tcount=389

There are also unfinished test results for another HD204UI:
http://hardforum.com/showpost.php?p=103 ... tcount=387


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 16th, 2011, 7:36 
Offline

Joined: November 8th, 2010, 18:07
Posts: 68
Location: Earth
Doomer wrote:
dfr wrote:
I also did not quite understand (or "feel" or "imagine") the last picture...

Perhaps you should ask questions then
Make them narrower to the subject


Here you go again:
1. Why in some hard drives slower access times can be seen aligning in some diagonal patterns (in HD204UI) and why in some other drives (in WDC WD1600BEVS) a checkered pattern appears (only from the second half of the scan) which "shape" changes regularly depending on the part of the disk surface scanned, while in some other drives no such regular patterns are present?
So in other words how to interpret those maps? If you do not know yourself, just be brave and say so. ;)

2. Why is there such a significant difference in MHDD scan results and HDDScan results (in case of my tests I compared it for SP1614N)?

3. What is where in the last picture of the article? Does it show a surface of one disk or more? If one then what other 3 heads are doing here? In which direction the plate is spinning?


Last edited by dfr on February 16th, 2011, 7:45, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 16th, 2011, 7:39 
Offline
User avatar

Joined: May 5th, 2004, 20:06
Posts: 2782
Location: England
In some cases internal HDD thermal calibrations can delay the odd sector during a sequential read / write etc.

In some cases internal S.M.A.R.T background tests/operations/log updates can cause minor delayed sector access

_________________
All went well until I plugged the drive in.


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 16th, 2011, 7:45 
Offline

Joined: November 8th, 2010, 18:07
Posts: 68
Location: Earth
Doomer wrote:
Read "Servo Track N" as "Servo Track N+2" sorry for typo


OK. So, so many people were allegedly reading this article for the last two years, and nobody noticed that, until I did it. :?: :shock: That raises a question do they really read it with understanding, or just are playing wise guys on forums and really do not understand a bit of it? :(

PS. There also some grammar mistakes, like 'is' or 'are' missing in a few places, and perhaps also it should be rather 'in factory' or 'at factory' instead of 'on factory'. There are probably more - you need help of some native speakers.


Last edited by dfr on February 16th, 2011, 7:59, edited 4 times in total.

Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 16th, 2011, 7:50 
Offline

Joined: November 8th, 2010, 18:07
Posts: 68
Location: Earth
guru wrote:
In some cases internal HDD thermal calibrations can delay the odd sector during a sequential read / write etc.


Even if that is the case, in case of 2TB WD such occurencess seem to be less frequent (or do not appear at all). Of course it would be good to have more such test results from many more HDs to reach any binding conclusions, but at the moment the trend is that HD204UI has a bit more bad timings than other drives.

Also your explanation doesn't explain the gross differences between HDDScan and MHDD scans.


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 16th, 2011, 8:17 
Offline
User avatar

Joined: May 5th, 2004, 20:06
Posts: 2782
Location: England
HDDScan uses windows API's AFAIK. Maybe MHDD uses a more direct path to the drive via it's own driver?

_________________
All went well until I plugged the drive in.


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 16th, 2011, 8:21 
Offline

Joined: November 8th, 2010, 18:07
Posts: 68
Location: Earth
guru wrote:
HDDScan uses windows API's AFAIK. Maybe MHDD uses a more direct path to the drive via it's own driver?


Yes, MHDD uses much more direct path, but we still are just guesssing if that is really the explanation of such huge differences or not.

"How generic DOS program talks to the drive.


PROGRAM <---> MSDOS <---> BIOS <---> IDE/SATA controller <---> Hard disk



And now how MHDD works:


MHDD <---> IDE/SATA controller <---> Hard disk

The main difference: MHDD does not use BIOS functions and interrupts."

http://hddguru.com/software/2005.10.02- ... al.en.html


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 16th, 2011, 8:32 
Offline
User avatar

Joined: May 5th, 2004, 20:06
Posts: 2782
Location: England
"<10 MS" 1223396 (MHDD) 1217450 (HDDSCAN) diff 5946 (46 reads @128 block read*)

"10<>50" 2412 (MHDD) 3087 (HDDSCAN) diff 675 (5 reads @ 128 block read*)

* Block count is a guess at 128 Blocks per read


Not all SW is created the same. Some people like to use native CPU ticks for timing and other people use stock functions already included with the source code. Windows API probably has a rather large overhead, perhaps you can measure the CPU load while running both SW tools?

_________________
All went well until I plugged the drive in.


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 16th, 2011, 10:19 
Offline
User avatar

Joined: September 29th, 2005, 12:02
Posts: 3561
Location: Chicago
dfr wrote:
3. What is where in the last picture of the article? Does it show a surface of one disk or more? If one then what other 3 heads are doing here? In which direction the plate is spinning?

Imaging you reading drive from LBA 0 toward the end
The last picture represent how drive would read actual surfaces when you do linear LBA reading
Hx represent heads. Read lines with arrows (serpents) represent sectors which were read from physical tracks and possible heads movements

You decided to scan let's say Quantum Atlas 10K5 from LBA 0 toward the end. And let's say this drive has 4 heads
Here is how drive is gonna read physical tracks
1. Tracks 0-63 on H0
2. Tracks 0-63 on H1
3. Tracks 0-63 on H2
4. Tracks 0-63 on H3
5. Tracks 64-127 on H0
6. Tracks 64-127 on H1
7. Tracks 64-127 on H2
8. Tracks 64-127 on H3
9. Tracks 128-191 on H0

And so on

If you do the same scan on a Samsung drive it could be something like this
1. Tracks 0-6000 on H0
2. Tracks 0-5900 on H1
2. Tracks 0-6100 on H2
3. Tracks 0-5800 on H3
4. Tracks 6001-12000 on H0

And so on

Each track may have from 1000 to 2500 sectors

Generally Samsungs have a lot more sectors per serpent. This leads to longer seeks when drive needs to switch heads and linear speed drops when drive reads toward OD

_________________
SAN, NAS, RAID, Server, and HDD Data Recovery.


Top
 Profile  
 
 Post subject: Re: Surface test of 5 HDDs.
PostPosted: February 16th, 2011, 10:22 
Offline

Joined: November 8th, 2010, 18:07
Posts: 68
Location: Earth
Ridiculous Samsung replies:
verbatim-usb-external-with-samsung-f4eg-hd204ui-t18455-20.html#p124735
verbatim-usb-external-with-samsung-f4eg-hd204ui-t18455-20.html#p124618

Thay claim that applying their patch won't change the firmware. LOL :shock:


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 24 posts ]  Go to page 1, 2  Next

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 50 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group