TwoRails
09-10-2006, 02:57 PM
As many of you know, I've been a big Maxtor fan for over 20 years. Now that they have been absorbed by Seagate, I have to start looking for another "favorite" which I suspect will be the ".10" perpendicular drives by Seagate. As chance may have it, I just picked up a WD drive. While I've never slammed WD, they have not been my choice. But, a 500 GB SATA 3GB drive by a major manufacture for only $159 on sale with No rebates required... well that was just to tempting to pass up.
This is my benchmark review of that Western Digital Caviar SE16 WD5000KSRTL 500GB 7200 RPM 16MB Cache SATA 3.0Gb/s Hard Drive.
In the past I have found that there really was a difference in an ATA 133 drive versus ATA 100. I've posted benchmarks showing this, but more important for me was that I had zero dropped frames during analog video capture using ATA 133 over too many dropped frames when using ATA 100. Using similar logic, I was expecting a notable speed increase with the WD SATA 3.0Gb/s over the SATA 1 drive I am benchmarking it against.
Being use to Maxtor whopping WD in my own systems, I was braced for a reversal. However, I was surprised at how close the two units are in testing. Some of the results are a little surprising, and the difference between the different benchmarks also gives some confusion on how they actually run the testing. While I usually run my own tests, which is consists of copying a massive number of various size files, I decided to stick with standard benches. I used Sandra 05 Lite, HD Tach 2.7 Lite, and Fresh Diagnose 7.2 for the testing.
The setup is simple. I cloned over the Maxtor to the WD for testing, keeping the ratio of the logical drives the same. The Maxtor is a 300 GB unit while the WD is 500 GB. All logical drives were defragged before testing. Since both drives are in the same system, the CPU, etc., do not play a part as it's a straight comparison. Note that the test motherboard supports only SATA I, however, since none of the tests came close to saturating the theoretical limits of SATA I, I doubt the scores would differ any in a newer mobo.
Remember, the Maxtor is an older SATA I, and the WD is a newer SATA 3.0Gb/s. Both have a 16 MB cache. All specs are listed with Maxtor on the left, WD on the right. All tests are run from the inside to the outer logical drive where applicable. The logical drives are "O," "P," and "S" for the Maxtor, and "T," "U," "V" for the WD.
First, Sandra. I was surprise here as the Maxtor had low buffered read scores and the WD had high ones, but the WD had low buffered write scores and the Maxtor was high in this area. Even though there was quite a difference, they averaged out in the overall score. Both were close, but as expected, the 3.0Gb/s drive edged out in the total score (index).
All scores are MB/s unless otherwise mentioned.
...O versus T.......Maxtor.....WD
--------------------------------------------
Total score (index).....55.....58
Buffered Read...........40.....77
Sequential Read.........63.....69
Random Read............42.....46
Buffered Write.........119.....46
Sequential Write........63.....58
Random Write...........46.....39
Average Access Time...8 ms..7 ms
...P versus U.....Maxtor.....WD
--------------------------------------------
Total score (index).....49.....52
Buffered Read...........40.....77
Sequential Read.........56.....59
Random Read............38.....42
Buffered Write.........119.....48
Sequential Write........55.....55
Random Write...........41.....38
Average Access Time...9 ms..7 ms
....S versus V..........Maxtor WD
------------------------------------------------
Total score (index).....37.....38
Buffered Read...........30.....77
Sequential Read.........41.....43
Random Read............30.....33
Buffered Write.........120.....44
Sequential Write........41.....41
Random Write...........32.....30
Average Access Time...9 ms...7 ms
Next is HD Tach Lite 2.70. Being the Lite version, only Read tests are supported. Also, for the WD, the Burst Read test failed multiple attempts. I suspect this is due to having an older HD Tach version. Also, the HD Tach only benches the physical drive, not logical drives, so only one test was run for each unit. In this test, the Maxtor did much better in the Read area, but again the average is fairly close. Without the Burst test for WD, however, it's difficult to see a clear winner.
...................................Maxtor......WD
-----------------------------------------------------------
Random Access Time...13.8 ms 13.2 ms
Read Burst Speed..........128.4.....0.1
Read Speed Max............116.0.....77.9
Read Speed Min..............36.7.....36.4
Read Speed Avg.............55.1.....58.2
CPU Utilization................35.6%.....36.3%
Finally, Fresh Diagnosis 7.2. FD's tests are not as elaborate as Sandra's, giving just average test scores. Once again, the drives are fairly close with Maxtor taking the outer two logical drives but WD scoring better further out on the last logical drive.
....O versus T..........Maxtor.....WD
----------------------------------------------------
Write Speed Avg........29.14.....28.56
Read Speed Avg.....1748.25..1713.48
....P versus U..........Maxtor.....WD
-----------------------------------------------------
Write Speed Avg........33.01.....30.78
Read Speed Avg.....1980.77..1846.68
....S versus V..........Maxtor.....WD
----------------------------------------------------
Write Speed Avg........29.01.....31.01
Read Speed Avg.........35.41.....41.85
In conclusion, I was a little surprised that the WD didn't have a bigger edge over the Maxtor. Still, the benches show that it's a strong drive. I have not used it in "normal" use yet, but it feels pretty good overall. I base that statement on other (IDE) WD drives that I've benched that were notably slower than what I am use to. My other drives are bolted into the drive cage with rubber grommets and therefore are very quite. The only spot I had open to mount the WD is in the external bay under the floppy drive. This location is the standard metal to metal contact. Even so, the drive is quite and can not be heard.
I would not hesitate to recommend this drive at or around this price. At the drives full price, which is very close to Seagate's newest line, I would personally opt for a Seagate 10 series drive (perpendicular) due to their high rating in the speed department.
Below are the screen shots of the above. All are straight shots except for the Fresh Diagnosis ones. Fresh diagnosis is more graphical (it shows a chart) which would be too large to attach. So, I simply cut-n-pasted the data section of each screen shot and put them all onto one pic.
(If you catch any typos in the data above, please post it!)
PS: I just found the 5 attachment limit, so I will have to make an additional post to cover all the screen shots.
This is my benchmark review of that Western Digital Caviar SE16 WD5000KSRTL 500GB 7200 RPM 16MB Cache SATA 3.0Gb/s Hard Drive.
In the past I have found that there really was a difference in an ATA 133 drive versus ATA 100. I've posted benchmarks showing this, but more important for me was that I had zero dropped frames during analog video capture using ATA 133 over too many dropped frames when using ATA 100. Using similar logic, I was expecting a notable speed increase with the WD SATA 3.0Gb/s over the SATA 1 drive I am benchmarking it against.
Being use to Maxtor whopping WD in my own systems, I was braced for a reversal. However, I was surprised at how close the two units are in testing. Some of the results are a little surprising, and the difference between the different benchmarks also gives some confusion on how they actually run the testing. While I usually run my own tests, which is consists of copying a massive number of various size files, I decided to stick with standard benches. I used Sandra 05 Lite, HD Tach 2.7 Lite, and Fresh Diagnose 7.2 for the testing.
The setup is simple. I cloned over the Maxtor to the WD for testing, keeping the ratio of the logical drives the same. The Maxtor is a 300 GB unit while the WD is 500 GB. All logical drives were defragged before testing. Since both drives are in the same system, the CPU, etc., do not play a part as it's a straight comparison. Note that the test motherboard supports only SATA I, however, since none of the tests came close to saturating the theoretical limits of SATA I, I doubt the scores would differ any in a newer mobo.
Remember, the Maxtor is an older SATA I, and the WD is a newer SATA 3.0Gb/s. Both have a 16 MB cache. All specs are listed with Maxtor on the left, WD on the right. All tests are run from the inside to the outer logical drive where applicable. The logical drives are "O," "P," and "S" for the Maxtor, and "T," "U," "V" for the WD.
First, Sandra. I was surprise here as the Maxtor had low buffered read scores and the WD had high ones, but the WD had low buffered write scores and the Maxtor was high in this area. Even though there was quite a difference, they averaged out in the overall score. Both were close, but as expected, the 3.0Gb/s drive edged out in the total score (index).
All scores are MB/s unless otherwise mentioned.
...O versus T.......Maxtor.....WD
--------------------------------------------
Total score (index).....55.....58
Buffered Read...........40.....77
Sequential Read.........63.....69
Random Read............42.....46
Buffered Write.........119.....46
Sequential Write........63.....58
Random Write...........46.....39
Average Access Time...8 ms..7 ms
...P versus U.....Maxtor.....WD
--------------------------------------------
Total score (index).....49.....52
Buffered Read...........40.....77
Sequential Read.........56.....59
Random Read............38.....42
Buffered Write.........119.....48
Sequential Write........55.....55
Random Write...........41.....38
Average Access Time...9 ms..7 ms
....S versus V..........Maxtor WD
------------------------------------------------
Total score (index).....37.....38
Buffered Read...........30.....77
Sequential Read.........41.....43
Random Read............30.....33
Buffered Write.........120.....44
Sequential Write........41.....41
Random Write...........32.....30
Average Access Time...9 ms...7 ms
Next is HD Tach Lite 2.70. Being the Lite version, only Read tests are supported. Also, for the WD, the Burst Read test failed multiple attempts. I suspect this is due to having an older HD Tach version. Also, the HD Tach only benches the physical drive, not logical drives, so only one test was run for each unit. In this test, the Maxtor did much better in the Read area, but again the average is fairly close. Without the Burst test for WD, however, it's difficult to see a clear winner.
...................................Maxtor......WD
-----------------------------------------------------------
Random Access Time...13.8 ms 13.2 ms
Read Burst Speed..........128.4.....0.1
Read Speed Max............116.0.....77.9
Read Speed Min..............36.7.....36.4
Read Speed Avg.............55.1.....58.2
CPU Utilization................35.6%.....36.3%
Finally, Fresh Diagnosis 7.2. FD's tests are not as elaborate as Sandra's, giving just average test scores. Once again, the drives are fairly close with Maxtor taking the outer two logical drives but WD scoring better further out on the last logical drive.
....O versus T..........Maxtor.....WD
----------------------------------------------------
Write Speed Avg........29.14.....28.56
Read Speed Avg.....1748.25..1713.48
....P versus U..........Maxtor.....WD
-----------------------------------------------------
Write Speed Avg........33.01.....30.78
Read Speed Avg.....1980.77..1846.68
....S versus V..........Maxtor.....WD
----------------------------------------------------
Write Speed Avg........29.01.....31.01
Read Speed Avg.........35.41.....41.85
In conclusion, I was a little surprised that the WD didn't have a bigger edge over the Maxtor. Still, the benches show that it's a strong drive. I have not used it in "normal" use yet, but it feels pretty good overall. I base that statement on other (IDE) WD drives that I've benched that were notably slower than what I am use to. My other drives are bolted into the drive cage with rubber grommets and therefore are very quite. The only spot I had open to mount the WD is in the external bay under the floppy drive. This location is the standard metal to metal contact. Even so, the drive is quite and can not be heard.
I would not hesitate to recommend this drive at or around this price. At the drives full price, which is very close to Seagate's newest line, I would personally opt for a Seagate 10 series drive (perpendicular) due to their high rating in the speed department.
Below are the screen shots of the above. All are straight shots except for the Fresh Diagnosis ones. Fresh diagnosis is more graphical (it shows a chart) which would be too large to attach. So, I simply cut-n-pasted the data section of each screen shot and put them all onto one pic.
(If you catch any typos in the data above, please post it!)
PS: I just found the 5 attachment limit, so I will have to make an additional post to cover all the screen shots.