 Home Members Only Membership Donations Reviews Store Mail List Legal  |
||
February 3, 2006
Review: Evaluating the Seagate 7200.8 300GB 16MB SATA Hard Drive
By Arthur Whalem
The Seagate 300GB 16MB Cache
I have used a Seagate Barracuda 7200.8 300GB 8MB SATA hard drive model ST3300831AS for the last eight months and it works well. It does not perform quite as fast as my Maxtor Maxline 300GB SATA hard drive but it does a good job. The local Fry's Electronics was offering a new 16MB version of the Seagate 300GB for $10 more than the 8MB version and I wondered how much difference there might be in performance between the two models. The 16MB model number is ST3300631AS. Basically, it has the same model number as the 8MB version except the "8" is a "6" in the model
number. The hard drive label has no other information printed on it that would indicate the hard drive has a 16MB cache.You can identify the 16MB model in the store by the 16MB Cache text in the lower right hand corner of the front of the retail box. I have not seen the bare drive version on-line as much as the 8MB model. Searching for ST3300631AS on Google provides considerably less hits and information than the standard Seagate 300GB 8MB SATA model ST3300831AS. Even a search at Seagate fails to produce a hit for the ST3300631AS. The Seagate 300GB 16MB is simply a Barracuda 7200.8 SATA NCQ hard drive with the larger 16MB cache. Both the 8MB and the 16MB support Native Command Queing.
What is Native Command Queuing (NCQ)?
Seagate explains the function of NCQ in simple terms as, "Somewhat similar to an elevator. The elevator travels sequentially from floor to floor and lets people on and off. It does not follow the sequence in which users push the floor request buttons, because that would mean an excessive mechanical load and much
longer times for waiting or riding." A SATA drive that supports NCQ uses an internal queue to dynamically reschedule actions in the most efficient manner. It monitors existing drive requests and allows computers to add additional commands while the drive continues to update and complete the existing queue of requests in a logical order. The hard drive is suppose to function faster and more efficiently when NCQ is functioning. However, recent testing of several brands of hard drives indicates that using NCQ may not always provide the best results. For more information on NCQ there is a joint white paper in PDF format on this topic authored by Intel and Seagate at this link.
PerformanceIn order to see how the 8MB and the 16MB Seagate 300GB SATA hard drives compare, both drives were erased and mounted in a SeriTek/1EN2 enclosure. The enclosure was attached to a SeriTek/1eVE4 PCI-X host adapter, which was mounted in the slot 4 of a PowerMac G5 2.0GHz Dual. Mac OS X 10.4.4 was used for all tests in this article. The Seagate 300GB had difficulty duplicating files quickly with Mac OS X 10.4.2 and previous versions. However, this was fixed in Apple system software with the release of Mac OS X 10.4.3 update. Lets see how the two Seagate hard drives compare.
Test #1
Using Drive Genius 1.2 the read and write performance benchmarks of the 8MB and 16MB Seagate SATA hard drives were recorded and displayed in the graphs below. The Seagate 300GB 16MB results are shown in green and the Seagate 300GB 8MB performance shown in blue.
Sustained Read
Seagate 300GB 16MB SATA HD is green. Seagate 300GB 8MB SATA HD is blue.
In the Sustained Read test both the 8MB and the 16MB Seagate 300GB SATA hard drives have similar performance results.
Sustained Write
Seagate 300GB 16MB SATA HD is green. Seagate 300GB 8MB SATA HD is blue.
In the Sustained Write test above the Seagate 300GB 16MB cache SATA hard drive provided a little better performance than the 8MB version.
Random Read
Seagate 300GB 16MB SATA HD is green. Seagate 300GB 8MB SATA HD is blue.
In the Random Read test both Seagate 300GB hard drives perform almost identically.
Random Write
Seagate 300GB 16MB SATA HD is green. Seagate 300GB 8MB SATA HD is blue.
In the Random Write test, the Seagate 300GB 16MB cache SATA hard drive once again pulled ahead of the Seagate 8MB cache. This was especially evident in the smaller file sizes. The real question is how will these Drive Genius "benchmarks" stand up compared to actual real world usage of these hard drives?
Test #2In this test, the Seagate 300GB 16MB and the Seagate 300GB 8MB SATA hard drives each contain the same 51.42 GB of data on them and were mounted as boot drives on a PowerMac G5 2.0GHz Dual with 2GB of memory and Mac OS X 10.4.4. Several boot configurations were tested. Each drive was booted and tested on the PowerMac G5 while on the "A" SATA internal drive bus, the "B" SATA internal drive bus, the FirmTek 1eVE4 external SATA adapter (Slot 4) and on the FirmTek 1VE4 external SATA adapter (Slot 3). Energy settings were on highest and the drives were not allowed to sleep. A 4.25GB folder (containing two video files) was copied to and from a Maxtor 1.1TB four drive striped RAID set. Both Seagate 300GB SATA hard drives were booted from each configuration. The copy time was recorded along with the time to repair permissions and duplicate the 4.25GB folder on each drive. Xbench and DiskTester readings of the drives performance and the overall computer performance were also recorded as a reference for this data.
| Seagate 300 16MB | Internal A | Internal B | 1eVE4 (Slot 4) | 1VE4 (Slot 3) |
| Copy 4.25GB to RAID | 1:17 minutes | 1:18 minutes | 1:15 minutes | 1:16 minutes |
| Copy 4.25GB from RAID | 1:18 minutes | 1:19 minutes | 1:17 minutes | 1:17 minutes |
| Repair Permissions | 32 seconds | 32 seconds | 32 seconds | 32 seconds |
| Duplicate 4.25GB | 2:36 minutes | 2:36 minutes | 2:37 minutes | 2:37 minutes |
| Xbench 1.1.3 Disk/Overall | 110/214 | 109/214 | 113/215 | 112/214 |
| DiskTester 1GB Read/Write | 48.7/49.5 | 49.1/48.7 | 49.9/49.3 | 49.9/49.2 |
| - | ||||
| Seagate 300 8MB | Internal A | Internal B | 1eVE4 (Slot 4) | 1VE4 (Slot 3) |
| Copy 4.25GB to RAID | 1:18 minutes | 1:18 minutes | 1:18 minutes | 1:18 minutes |
| Copy 4.25GB from RAID | 1:18 minutes | 1:21 minutes | 1:18 minutes | 1:18 minutes |
| Repair Permissions | 33 seconds | 33 seconds | 32 seconds | 32 seconds |
| Duplicate 4.25GB | 2:37 minutes | 2:37 minutes | 2:37 minutes | 2:37 minutes |
| Xbench 1.1.3 Disk/Overall | 110/214 | 109/213 | 110/214 | 110/214 |
| DiskTester 1GB Read/Write | 48.1/49.3 | 48.4/49 | 49.4/48.5 | 49.4/48.8 |
As you can see above, the Seagate 300GB 16MB SATA is slightly faster than the Seagate 300GB 8MB. However, the difference is very small. In most cases, the difference was less than 3%. Based on these results, I would not expect that most users will actually realize much additional benefit from the larger 16MB cache of the Seagate 300GB SATA model "ST3300631AS". As there is only $10 difference between the 8MB and the 16MB, I would probably still get the 16MB. However, buyers should not expect a major boost in the performance of the Seagate 300GB SATA hard drive just because it has a 16MB cache.
In a previous AMUG review of the Seagate 300GB SATA 8MB cache model, running Mac OS X 10.3.9 produced significantly slower duplicate file performance. Apple Mac OS X 10.4.3 dramatically improves the performance of duplicating files with Seagate 300GB hard drives. Apple also reduced the time it takes to repair permissions with OS X 10.4. If you have a Seagate 300GB, upgrading to Mac OS X 10.4.4 will allow it to perform faster. In fact, moving to Mac OS X 10.4.4 will usually provide more of a hard drive performance increase than moving to a 16MB cache.
How does the New Maxtor 300GB Compare?
Maxtor has released the new MaxLine III 7V300F0 300GB 3.0Gb/s SATA hard drive. It still has a 16MB cache and adds native command queuing and 9ms seek time. This hard drive has a five year warranty just like the Seagate. As a previous AMUG review compared the Seagate and Maxtor 300GB models, I thought it might be interesting to see how the performance of the new Maxtor MaxLine III 7V300F0 compared to the Seagate 300GB 16MB cache model.
In this test, the WiebeTech SilverSATA II was used in mirror mode to duplicate 51.42 GB of data from the Seagate 300GB 16MB cache hard drive to the new MaxLine III 7V300F0. The SilverSATA II is a very cool device. It can be used as a regular SATA enclosure or for duplicating SATA hard drives with or without being connected to a computer. In mirror mode the SilverSATA II performs a device copy of the hard drive inserted first to the hard drive inserted last. I simply loaded the Seagate 300GB 16MB in the top bay and turned the SilverSATA II on. After the SilverSATA II was running, the Maxtor was inserted into the bottom bay and the SilverSATA II began a device copy of the Seagate to the Maxtor with no computer connected. Using this method, any data that was on the Maxtor is erased and the Seagate data is duplicated on the Maxtor. The device copy is identical down to the same partition size. This is similar to the fast block to block level, hard disk to hard disk copy method used by the Disk Jockey. This same duplication method was used between the Seagate 8MB and the 16MB before testing. With the Maxtor MaxLine III 7V300F0 now an exact duplicate of the Seagate 16MB the same tests were completed again.
| Seagate 300 16MB | Internal A | Internal B | 1eVE4 (Slot 4) | 1VE4 (Slot 3) |
| Copy 4.25GB to RAID | 1:17 minutes | 1:18 minutes | 1:15 minutes | 1:16 minutes |
| Copy 4.25GB from RAID | 1:18 minutes | 1:19 minutes | 1:17 minutes | 1:17 minutes |
| Repair Permissions | 32 seconds | 32 seconds | 32 seconds | 32 seconds |
| Duplicate 4.25GB | 2:36 minutes | 2:36 minutes | 2:37 minutes | 2:37 minutes |
| Xbench 1.1.3 Disk/Overall | 110/214 | 109/214 | 113/215 | 112/214 |
| DiskTester 1GB Read/Write | 48.7/49.5 | 49.1/48.7 | 49.9/49.3 | 49.9/49.2 |
| - | ||||
| Maxline 300 16MB 7V300F0 | Internal A | Internal B | 1eVE4 (Slot 4) | 1VE4 (Slot 3) |
| Copy 4.25GB to RAID | 1:07 minutes | 1:07 minutes | 1:06 minutes | 1:06 minutes |
| Copy 4.25GB from RAID | 1:05 minutes | 1:05 minutes | 1:04 minutes | 1:04 minutes |
| Repair Permissions | 30 seconds | 30 seconds | 30 seconds | 31 seconds |
| Duplicate 4.25GB | 2:15 minutes | 2:15 minutes | 2:13 minutes | 2:13 minutes |
| Xbench 1.1.3 Disk/Overall | 158/233 | 158/233 | 162/234 | 161/234 |
| DiskTester 1GB Read/Write | 56.1/60.3 | 56.3/60.5 | 59.3/60.6 | 59.1/60.6 |
As you can see from the results above, The Maxtor MaxLine III 7V300F0 easily beats the 16MB Seagate 300GB SATA hard drive. The Maxtor is 10 seconds faster
copying data to the 1.1TB four drive striped RAID set (which is about 14% faster). It is even faster copying data to it from the RAID. Repairing Permissions is only a few seconds faster with the Maxtor. However, duplicating 4.25GB on the Maxtor is 15% faster. The Xbench scores are significantly better with the Maxtor and the DiskTester scores are about 20% faster using the Maxtor. Based on this information, it appears that the Maxtor MaxLine III 7V300F0 300GB beats the Seagate 7200.8 models in the 300GB SATA category.DiskTester 1GB Run Area Test
In order to see how each of the three hard drives would perform as they became full, each drive was erased and tested using a DiskTester 1GB "Run Area Test". DiskTester measures the combined performance of a volume and the Mac OS X operating system. Using the command: ./disktester run-area-test --transfer-size 131072 --iterations 3 --test-size 1024 --delta-percent 10 DriveName, puts DiskTester to work testing how the hard drive will perform when empty, 10% full 20% full and so on. A PowerMac G5 2.0GHz Dual with Mac OS X 10.4.4 was used with the hard drives mounted on a FirmTek SeriTek/1eVE4 SATA host adapter. The results for each SATA hard drive are shown below.
Maxtor MaxLine III 7V300F0
Seagate 300GB 16MB ST3300631AS
Seagate 300GB 8MB ST3300831AS
As you can see in the results above, the Maxtor 300GB stays faster as the hard drives become full. By the time the disks are 90% full the Maxtor MaxLine III 7V300F0 is almost 30% faster than the Seagate 300GB SATA models.
Size, Power & Weight
On a Macintosh the Seagate 300GB 16MB and 8MB models format to 279.25 GB each. The Maxtor 300GB MaxLine III 7V300F0 formats to 279.27 GB. According to the Kill-a-Watt electricity usage monitor, the Seagate 300GB 8MB
and 16MB SATA models use 11 watts when idle. When busy the Seagate 300GB SATA hard drives can utilize approximately 13 watts each. The Maxtor MaxLine III 7V300F0 uses 9 watts when idle and approximately 12 watts when busy copying data. The Seagate 300GB hard drives weighs 1 pound 6.4 ounces. The Maxtor 300GB weighs 1 pound 5.8 ounces. The Maxtor MaxLine III 7V300F0 formats a little bigger, weighs less and uses a little less power. Discussion
This article was designed to find out if the Seagate 7200.8 300GB 16MB SATA hard drive provided enhanced performance over the 8MB model. However, as the review progressed it became clear that the performance improvements were minimal. Once the Maxtor MaxLine III 7V300F0 was introduced to the comparison it became clear that it was the star of the show. The Maxtor 300GB model 7V300F0 performs faster, stays faster as the drive becomes full, formats 20MB larger, weighs less and uses a little less power than the Seagate 7200.8 300GB SATA hard drives.
Now that Seagate is acquiring Maxtor in a 1.9 billion deal, it will be interesting to see if Seagate products improve from the sharing of technology with Maxtor or if Maxtor technology stays with the Maxtor brand. While I have had more failures with Maxtor hard drives than Seagate models, the Maxtor MaxLine III 7V300F0 is a very nice performer that comes with a five year warranty. After looking at the results, my next SATA hard drive purchase will probably be a MaxLine III 7V300F0 300GB SATA hard drive.
Pros
Maxtor MaxLine III 7V300F0 300GB SATA
Performs better than the Seagate 300GB in these tests.
Wieghs less, uses less power.
Performs well as the drive becomes full.
Has a five year warranty.
Quiet, low heat.
Seagate 300GB 16MB SATA model ST3300631AS
Steady performer, high quality.
Five year warranty.
Quiet, low heat.
Cons
Maxtor MaxLine III 7V300F0 300GB SATA
This model is available in less local locations than the Seagate.
I have experienced more failures with Maxtor drives.
Seagate 300GB 16MB SATA model ST3300631AS
Slower overall performance than the Maxtor MaxLine III 7V300F0.
Weighs more and uses more power than the Maxtor MaxLine III 7V300F0.
Very little performance advantage over Seagate 7200.8 300GB 8MB model.
Seagate 300GB 16MB SATA gets 3 AMUGs out of 5!
These tests convinced me that the Maxtor MaxLine III 7V300F0 300GB SATA hard drive is a better performer for the PowerMac than the Seagate 7200.8 300GB 16MB SATA. I had hoped that the 16MB cache on the Seagate would provide a larger performance advantage. Instead, I found that the 16MB upgrade provides very minimal real world performance enhancement to the Seagate Barracuda 7200.8 300GB SATA hard drive. Perhaps in the future, AMUG will be able to test the Seagate Barracuda 7200.9 300GB against the Maxtor and determine how it compares. There are reports at barefeats.com that the Seagate 7200.9 cannot be seen by the PowerMac G5 built-in SATA controller unless SSC (Spread Spectrum Clocking) is turned off using a DOS utility. As such without access to a PC and the DOS program Macintosh users may have trouble with the Seagate 7200.9 SATA hard drives. I found a review of the 500GB Seagate 7200.9 at anandtech.com. That article indicates that the 7200.9 tops out at 64MB per second and ends at 30-38MB per second when full. If that data is duplicated with DiskTester on the Macintosh, the 7200.9 will measure 10% slower than the Maxtor MaxLine III 7V300F0. Stay tuned for the next adventure of AMUG Macintosh Drive Wars.
Contact Information:
Seagate Technology
920 Disc Drive
Scotts Valley, California 95066
Phone: 877-271-3285
http://www.seagate.com/contact
Copyright 2006
Arizona Macintosh Users Group, Inc. (AMUG)
Visit AMUG at www.amug.org for computer news, and resources.
JOIN AMUG!
Article Comments:
From: Rob-ART Morgan <rob-art@barefeats.com>
Subject: Re: AMUG Seagate 16MB vs 8MB Review
Date: Sat, 4 Feb 2006 09:18:22 -0800
Below are the results of the DiskTester test of the Seagate 7200.9 500GB drive.
If you want to compare the speed when full with the 300GB drives, you have to look at the 60% numbers, not the 90 or 100% result since the 7200.9 at 60% is at the 300GB mark. The "empty" times of the Maxtor beats it by a long shot. But if you compare the 7200.9 at 60% (or 300GB) to the Maxtor (or 7200.8) at 100% or 300GB, the 7200.9 is faster. Even your Seagate 7200.8 beat it while empty. Obviously this 500GB Seagate is about capacity more than ultimate speed. I used the Sonnet Tempo-X eSATA 4+4 card for testing the drive. Testing was done on a G5/2.0Ghz Power Mac (Oct 2004) with 2GB of RAM.
rob-ART