You have probably guessed already that this review will be devoted to a new generation of desktop hard disk drives from Seagate - Barracuda ATA IV. The number "4" in the name stands for the fourth generation of IDE HDDs, which are proudly called Barracuda drives. This is how the Seagate's SCSI drives high reputation helped the company to push into the market a new IDE HDD family.
We all do remember very well that the first Barracuda IDE hard disk drives were not so fast as their SCSI fellows. However, introducing more and more new models Seagate managed to reduce the performance gap little by little, so that it got every chance to satisfy its own ambitions one day. Will it happen today? Well, we'll see immediately, as the tests, which will help us make the decision are just in front of you. So, let's go!
Well, to tell the truth, we were struck dumb by the way the new Seagate Barracuda ATA IV actually looked… Even though we have already seen some pics in the reviews available on the web, the exterior of the drive puzzled us, in a way. Seagate has given up its traditional design and created an ordinary "metal brick". What made Seagate veer away from the design, which has already stood the test of time and proven pretty successful? Was it for the finer mechanics that required more protection against vibration and overheating? In fact, it's pretty hard to say, however, the general tendency is quite clear: the HDD "insights" are protected by massive case sides and metal plates.
By the way, we forgot to say why Barracuda ATA IV requires such fine mechanics: this HDD boasts 40GB per platter data density. As we remember, the platters of the same capacity are used in Seagate U Series 6 HDDs, although the spindle rotation speed of the latter is considerably smaller than that of our today's hero (5,400rpm against 7,200rpm by Barracuda ATA IV). Moreover, the heads of U6 drives are positioned much slower as well, but we will return to this discussion a bit later. The width of the data track has been reduced nearly twice compared with that of the previous generation drive, Barracuda ATA III, and since the average seek time remained unchanged, the drive construction needed to be much safer and more vibration-free. Otherwise the read/write head will need more time to calm down after faster positioning and as a result the hard disk drive will be no longer that fast.
Besides, this kind of HDD design will improve the heat dissipation into the environment, which will undoubtedly have a positive effect on the overall drive's reliability.
Well, this seems to be about all we could say on the matter. Now it's high time we introduced the new hard disk drive to you, guys:
And what have we got inside the casing? We have already told you that this hard drive features 40GB platters. The only thing we left out here is the capacity: elder models will come with the capacity of 60GB and 80GB, i.e. there will be only 2 platters used. Therefore, Barracuda ATA IV will be much quieter and "cooler" than those hard disk drives, which require 3 platters to provide the capacity of 60GB or 80GB. In fact, Barracuda ATA IV owes its quiet operation mostly to new fluid bearing engines. Their new construction together with a new casing design (which creates no resonance ;) allowed Seagate to achieve unprecedentedly low noise level of the working drives. The company claims that the single-platter drive makes only 2Bel of noise in idle mode (when the spindle is rotating but the heads do not move). Dual-platter drives show around 2.4Bel of noise in idle mode.
Besides Seagate Barracuda ATA IV HDD, we also included some very popular drives, namely: Quantum Fireball Plus AS, IBM Deskstar 60GXP, Western Digital Caviar WD600BB and Western Digital Caviar WD800BB. The drives participating in our tests featured the following capacities: 60GB (by Quantum, IBM and WD600BB) and 80GB (by Seagate and WD800BB). So, we tried to carry out the tests in more or less equal conditions for all the candidates. Of course, the drives we took a look at, belong to different generations (we couldn't ask the companies to launch their had drives simultaneously, really :), use different platters, etc. As for us, we were guided by only one single criterion when selecting the opponents to our Barracuda ATA IV: the HDD tested had to belong to the latest generation and feature the highest storage capacity available in retail today.
To make the comparison easier for you, we decided to compose a table with all the brief technical specs:
|Seagate Barracuda ATA IV |
|Western Digital Caviar |
WD600BB / WD800BB
|IBM Deskstar 60GXP |
|Quantum Fireball Plus AS|
|Spindle Rotation Speed||7,200rpm||7,200rpm||7,200rpm||7,200rpm|
|Average Seek Time (Read)||9.5ms||8.9ms||8.5ms||8.5ms|
|Average Track-to-Track |
Seek Time (Read)
|Average Full Stroke |
Seek Time (Read)
There are at least two things catching your attention in this table. First of all, Seagate claims a really good track-to-track seek time for its Barracuda ATA IV HDD. The track-to-track density by Barracuda ATA IV is very high (58,000 tracks per inch) that is why it is a really interesting question how on earth Seagate engineers managed to combine high density with a fast heads speed like that. Seagate's competitors really have to think about it.
Secondly, please, pay attention to Western Digital drives: WD600BB and WD800BB. They use three 27GB platters, however, the 60GB model was designed not by leaving out the sixth read/write head. If all the five sides of the three platters were used to the full extent, we could get a 67.5GB HDD. However, it has already become very popular to make the drive capacity look nice and beautiful, so we have a 60GB model here. How did they get it? Well, we'll talk about it later in the article…
Testbed and Methods
We tested the newcomer in the following test system:
- Intel Pentium III (Coppermine) 600MHz CPU;
- ASUS CUBX-E mainboard, bios 1007A;
- 2x128MB PC133 SDRAM by Hyundai;
- Matrox Millennium 4MB graphics card;
- Windows 98/Windows 2000 Pro.
In order to check the HDDs performance in different UDMA modes, we had to use the following controllers:
- UDMA33: the controller integrated into i440BX chipset;
- UDMA66: Promise Ultra66 controller;
- UDMA100: integrated Promise Ultra100 controller.
For the Promise controllers we used the drivers ver. 1.60 (build 33).
The disk drives were connected as Master-units to a separate IDE-channel. DMA support in Windows was enabled. We used FAT32 and NTFS file systems to format each of them as one logical drive of the maximum size with the default cluster. All the tests were run 4 times and then the average results were taken for the diagrams. The HDDs didn't rest for cooling down between the tests.
Here are the benchmarks used:
- Windows 98 WinBench 99 1.2
- Adaptec Threadmark 2.0
- Windows 2000 WinBench 99 1.2
- HDTach 2.61
- IOMeter 1999.10.20
As usual, we resorted to two tests to find out the average access time: HDTach and Winbench:
The new Seagate drive proved just excellent, taking into account that its platters are twice as dense as those of the most competitors. Moreover, the average access time shown by Barracuda ATA IV is 1ms better than that of the good old Barracuda ATA III. No doubt, it is a great success for the Seagate engineers who managed to combine high speed with high data density so wel
The fastest HDD as far as average access time goes, is still IBM Deskstar 60GXP.
Linear Read Speed
To measure the linear read speed we used a special Disk Transfer Rate benchmark from WinBench99 benchmark set.
As a rule, Seagate hard drives appear pretty fast in this test. Let's see, if Barracuda ATA IV proves this rule as well.
Well, everything turned up to our expectations! Seagate increased the data density not only by adding more tracks, but also by adding more sectors per track. Of course, this move deserves our applause, as higher data density should undoubtedly tell on the drive's performance in file and synthetic benchmarks.
We would also like to draw your attention to the way the Western Digital HDD performed in this test. The drives show almost the same read rate in the beginning, while in the end WD600BB takes the lead rushing ahead of its counterpart. Do you remember that we have already mentioned this peculiarity of the WD600BB? It is very likely that WD reduced the number of cylinders per platter by its 600BB drive that is why the platter "ends up" with more dense tracks than in case of WD800BB. As for the slight difference in linear read speed shown by WD600BB and WD800BB in the beginning, it can be explained by the measuring error as well as by WD's policy. We are not ready to make any definite statements here.
Well, let's see what Seagate Barracuda ATA IV is worth here. As usual, we would like to start with the average read speed:
As you can see, the hard disk drives are again positioned in the same order as we have just seen on the liner read speed diagram. The only difference is that the Quantum HDD yielded to both Western Digital fellows having occupied the last position.
High linear data density of the Seagate Barracuda ATA IV allowed this drive to easily beat all the others. Although we can talk about the victory only in UDMA100 mode. For some unknown reason, Barracuda ATA IV suddenly slows down in UDMA33 and UDMA66 protocols. In fact, we have already mentioned in our previous reviews that Seagate is not getting along with the UDMA33 protocols very well, however, as for the UDMA66, things have always been just perfect here.
Well, now let's pass over to the average write test:
Hm, here things seem to be even worse! Barracuda ATA IV falls to the last position in our rating, with the IBM Deskstar 60GXP wins the first prize. The most interesting thing here, however, is the second prize winner: WD600BB, which succeeded due to its "shorter" platter and hence higher average write speed than that of the WD800BB.
We will try to explain in the Conclusion section what caused such unusual behavior of Barracuda ATA IV HDD, and in the meanwhile, please, take a look at one more benchmark taken from the HDTach 2.61 set. It is Read Burst Rate, which measures how fast the data block is read from the HDD cache:
It's a pity but Seagate Barracuda ATA IV didn't prove to be an exception here. We mean that Seagate drives are, unfortunately, again in the very end and don't seem to be moving any closer to the leaders.
And the first position again belongs to the IBM baby followed very closely by Quantum Fireball Plus AS and both WD HDDs.
Well, judging by HDTach results, Seagate Barracuda ATA IV doesn't look that attractive. However, we should definitely bear in mind that this is a synthetic benchmark that is why its results can't be a determining factor when making the final conclusions. Anyway, let's proceed: it's time for WinBench to step in!
WinBench99 1.2 for Windows98
Hey, it's a brilliant start! Only 2KB/sec brought the victory to the IBM Deskstar 60GXP. Actually, we were really impressed by a small performance difference like that. However, we do not invent the numbers that is why we didn't "pull up" the results of Barracuda ATA IV… What for? Actually, any more or less sane person does understand that a small performance difference like that will not mean anything for the performance of the first three HDDs in real life. In other words, the solutions from IBM, Seagate and Quantum will perform evenly. As for the WD HDDs, they fell a little bit behind, but are still doing pretty well here.
Adaptec Threadmark 2.0
The laurels appear shared between the two WD drives, while Barracuda ATA IV was only the third rushing slightly ahead of its No. 1 competitor - IBM Deskstar 60GXP. The Quantum Fireball Plus AS occupied the last rating line.
WinBench99 1.2 for Win2000 (FAT32)
In general, the whole thing repeats the picture we have just seen in Win98 tests. IBM Deskstar 60GXP is leading the race, then comes Seagate Barracuda ATA IV, then we can see a battle between Quantum Fireball Plus AS and Western Digital pieces. Note that the drives show very small performance difference so that it can be again regarded as purely a measuring error. Unfortunately, WinBench test cannot boast providing stable results every time it is run. That is why we always undertake 4 cycles and then average the results obtained. However, who would guarantee that 4 times is enough here?...
WinBench99 1.2 for Win2000 (NTFS)
As we have already proven multiple times, the shift to a different file system may provoke a drastic change in the drive's rating. But sometimes it may not. Let's see what we've got here:
One more rule seems to be proven here: WD HDDs are usually very efficient in NTFS file system that is why here they also took the first two positions. The third position belongs to… No, not to IBM Deskstar 60GXP, as you may have expected, but to Seagate Barracuda ATA IV. The notorious IBM drive yielded even to Quantum Fireball Plus AS turning the last one in the race.
High performance of the Barracuda's heads unit lets us pin some hopes upon this drive. Even though far not everything depends on the fastness of the HDD mechanics, we cannot completely disregard its contribution. Here are the results:
Well, the results shown by the new Seagate Barracuda ATA IV are not very impressive, but they are still considerably better than those of its predecessor, Barracuda ATA III.
Now let's take a closer look at different patterns:
Even though the drive boasts advanced mechanics, it proves to be targeted solely for desktop systems.
Well, and now we have just to compare the performance of Seagate Barracuda ATA IV with that shown by other testing participants. We will compare the Sum Total value, which is calculated taking into consideration the results obtained with Light, Moderate and Heavy loads (16, 64 and 256 requests):
Heat and Noise
This time we resorted not to hardware means to measure the temperature, but to software. Since we run all Intel IOMeter tests using Promise controller, we needed a utility to read the current temperature from the HDD SMART-field, if there is any info like that there, of course. Unfortunately, not all the HDDs come out equipped with thermal sensors, but Barracuda ATA IV luckily had one :)
Well, we managed to find this utility created by one smart Russian guy. It is called DTemp and you can download it here.
According to the report the utility provides in the end, Seagate Barracuda ATA IV heated up to 42oC after 2 hours of IOMeter tests (the room temperature was around 22oC). Actually, we believe this is quite acceptable, although it would be much better to arrange some extra cooling for the device, just in case.
As for the noise, it is a really thorny subject for us still, as we can't get our hands on the special equipment for noise level measuring. That is why this time we will again rely on our personal impressions and feelings, though we bet you will not be disappointed, as we have got our hand in it already :)
So, as Seagate has promised, we didn't hear the spindle rotating. We mean that we didn't hear anything at all! As for the heads positioning, the situation is a bit different. If you move your ear as close as 10cm to the HDD running in fast seek mode with the PC case open, then you may hear the heads moving. But as soon as you close the PC case, the power supply unit fan and the CPU cooler will be the only noise makers.
Well, the three-day marathon with the Seagate Barracuda ATA IV is finally over. Reluctantly, we have to admit that we had to put Seagate Barracuda ATA IV on the same plane with IBM Deskstar 60GXP. Judging by the benchmarks results, Barracuda ATA IV has always been on the first or second position in Business and High-End WinBench tests. And they are exactly the ones playing the most important role when we choose a HDD for work with Windows applications.
IOMeter tests showed that Barracuda ATA IV firmware is optimized for "non-server" tasks.
Adaptec Threadmark proved that the drive firmware is also optimized for streaming operation.
The main disappointment was the HDTach benchmark: the performance of Seagate Barracuda ATA IV appeared to be dependent on the current UDMA mode. We suspect that it was because of too great firmware optimization for Windows applications. It looks as if the caching algorithm of Barracuda drive were incompatible with the data pack size of 1MB used in HDTach benchmark. Its cache seems to be too multi-segment…
In fact, we wouldn't be surprised to find out that the cache line size of Barracuda ATA IV depended on the UDMA interface. In other words, the higher is the protocol speed, the larger line firmware selects.
However, it may be the benchmark, which should be blamed for low performance results. To check if this supposition makes sense, we also tested the drives with the ATTO Benchmark and we didn't discover any problems by Seagate Barracuda ATA IV HDD when the interface got changed. In the nearest future we will complete the analysis of these results, so that you have a chance to check them, too. If the results obtained in ATTO Benchmark prove credible enough, we will even think of including them in our standard testing set.
All in all, Seagate launched a really dangerous beast combining brilliant performance and high reliability with quite reasonable price (due to a more optimal construction). The new design as well as noiseless operation make this solution a really attractive buy.
That is why if you are about to buy a new large IDE HDD in the nearest future, we would advise you to think about getting a Seagate Barracuda ATA IV. It is really worth your consideration at least.