About a year ago we got acquainted with the hard disk drive from the new Western Digital family called WD Protege. It is interesting that Western Digital first implemented 20Gb platters in its value HDD solutions and only then introduced them in new WDxxxBB hard disk drives with 7,200rpm spindle rotation speed.

The first acquaintance with this HDD was pretty instructive, as this hard drive easily outperformed not only the drives of the same category, but also some solutions with 7,200rpm rotation speed. A bit later, this HDD decided to pit its speed against that of the opponents, which included a higher class solution from Western Digital, WD200BB (see our HDD Roundup: 7 Hard Disk Drives with 20GB Platters). Surprisingly, our hero turned out not too much slower than the latter.

A year has passed since then: all the storage solutions manufacturers worked really hard all this time, so that finally 20GB platters have turned out no wonder any more, and the storage capacity of the disk drives doubled due to introduction of 40GB platters. And then, suddenly, we got hold of WD protege HDD with 40GB storage capacity (this has become our favourite size :)

Testing Participants

To make an illustrative and exciting competition we selected the following hard drives to take part in our race (with the fastest 5,400rpm HDDs among them):

• Two hard disk drives with new 40GB platters: Seagate U6 and Maxtor D-540X-4K;
• Fujitsu MPG AT-E (for its achievements under Windows 98);
• WD400AB, which is positioned by Western Digital as a higher class solution, though we have already seen a Protege outperforming Caviar. Will the story repeat?

Well, let's take a closer look at WD400EB:

  

The hard disk drive looks just the same as the predecessors: the typical reverse side and the lid shape remained the same (it's hard to mix Protege up with anything else).

* - We keep and keep saying that the spindle rotation speed of WDxxxAB hard drives is a bit higher than what is stated on the Western Digital official web-site, but the company seems to be unwilling to believe us :(
** - One more evident absurdity. If the drive has two heads, then why does it need two platters? It seems as if the WD specialists got mistaken somewhere. Remember about this hitch.
*** - This number is absolutely correct!

So, the claimed parameters for WD400EB are a bit different from the WD200EB we have already looked at (in particular, the average seek time and full stroke time are different). This can be explained by a normal difference between a single- and dual-platter HDD model, however, we can't forget about that strange situation with the number of heads and platters in the table above. Maybe WD400EB has only one platter? Then it should be a 40GB one!

Anyway, let's try to guess during the tests session.

Testbed and Methods

We tested the newcomer in the following test system:

• Intel Pentium III (Coppermine) 600MHz CPU;
• ASUS CUBX-E mainboard, bios 1007A;
• 2 x 128MB 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 on ASUS CUBX-E mainboard.

For 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

Performance

Average Access Time

To tests the average access time let's resort to the good old HDTach and WinBench99 benchmarks:

Well, we can't say anything definite taking into account only access time rates. WD400EB is slower than Maxtor D540-4K, but faster than Seagate U6. And the fastest mechanics belongs to Fujitsu MPG with 20GB platters and WD400AB with 30GB platters.

Sustained Linear Read Speed

And this benchmark will help us to understand how high is per platter data density of WD400EB HDD and if it uses the same platters as WD200EB solution:

As we see, the platters (or platter?) of WD400EB are very dense that is why reading from the platter in the beginning of the hard disk drive (on the densest tracks) and in the end of it is higher than by other hard drives.

When we tested WD200EB, we found out that its DTR: Beginning made 25.6MB/sec, while by WD400EB it reached 39MB/sec. From the previous benchmarks we know that despite more platters used, the read speed doesn't grow, which implies that the performance growth has to do with the platter changes!

Can the read speed from the platter of a 5,400rpm HDD make 39MB/sec (even not all the 7,200rpm HDDs can reach read speed like that)? Yes, it can! We came across this peculiarity when we tested Maxtor D531X and D541X hard drives. And this review is another proof of that.

The second curious coincidence is the fact that Maxtor D541X, which showed impressive performance like that, used a 40GB platter. Do you feel that the rope is already getting round the neck? :)

This test showed that WD Protege 400EB is an indisputable leader in the read speed among 5,400rpm HDDs, although we still can't answer the question about the number of platters used in it.

HDTach 2.61

This benchmark will help us to check the average read and write speed and the maximum read speed from the buffer of the hard disk drive.

The only thing that put us on our guard was the average access time depending on the UDMA-protocol. This is most likely to be just a coincidence, but it is a very interesting coincidence then, we should admit.

Judging by the read speed from the buffer, the electronics of both WD hard drives is identical.

The platters however, are totally different. The average read speed by WD400EB is over 11MB/sec higher than that of WD400AB. The gap between our hero and its nearest competitor, Maxtor D540X-4K makes over 4MB/sec. Yeah, this is a really cool "Low-End" hard disk drive...

In the write speed test the tremendous advantage of WD400EB over its competitors has suddenly vanished. Though it nevertheless managed to win the first prize, having left Maxtor D540X-4K 1MB/sec behind.

WinBench99 1.2 for Windows98

Let's see how WD Protege will behave in Windows98:

At first glance, the results seem just brilliant. However, we suggest comparing them with the results obtained for other testing participants before making any conclusions:

Beautiful! In UDMA100 WD400EB managed to outperform Maxtor D540X-4K HDD and to become a leader again.

In High-End benchmarks the competitors were fighting with impressive enthusiasm. Three HDDs were trying to win the victory, and Maxtor D540X-4K succeeded. The second prize belonged to Fujitsu MPG AT-E (that was one of the reasons for us to include it into these tests), and WD400EB had to put up with third prize only.

Adaptec Threadmark 2.0

This benchmark deserves being called one of WD's favourites, as all WD hard disk drives usually show very high results in it. The main idea of this benchmark looks as follows: the test opens a few files for reading and writing and shifts between them now and then. This way it checks whether the HDD suits for multi-thread operation. The results represent the file processing speed in MB/sec.

Well, once again WD hard drives proved their perfection in this benchmark: the difference in performance shown by WD400AB and WD400EB can be explained by 11-time difference in spindle rotation speeds.

WinBench99 1.2 for Win2000 (FAT32)

The results show that WD400EB will be ahead of all the other 5,400rpm HDDs.

In Business Disk WinMark our hero won the second prize following its counterpart, WD400AB, and Maxtor D540X-4K now stays only the third.

In High-End Disk WinMark WD400EB suddenly beat all the rivals. It appeared over 2MB/sec faster than its closest competitor, Maxtor D540X-4K, which is more than worthy for a Low-End HDD.

WinBench99 1.2 for Win2000 (NTFS)

Judging by the previous reviews of ours, Western Digital hard disk drives work more than fine in NTFS. What do you think, will WD400EB continue this good tradition?

We'll find it out right away! :)

To tell the truth, the result appeared somewhat unexpected: it came the third :(

At the same time, WD400AB still saved the face of Western Digital by winning the first place in this test, so we shouldn't doubt about the good family traditions of WD solutions. Now we pass over to the next diagram:

Well, we are pleased to see that there are some predictable things left in this world. WD400EB defeated Maxtor D540X-4K and proudly returned the title of the fastest HDD in Win2000 NTFS. Though this time they will have to share it between the two drives :)

Intel IOMeter

Now we will have a look at Intel IOMeter results:

Of all the results provided by Intel IOMeter benchmarks set (by the way, in the table above there is only one percent of all the numbers this tests gives out), we are interested only in Total I/O, that is the amount of read/write requests processed by the hard disk drive per second.

This is exactly where the faster spindle rotation speed of WD400AB played a really important role. It appeared simply unattainable for the rest of the racers.

WD400EB HDD performs a bit worse in case of lower workload than Maxtor D540X-4K (no wonder, since WD400EB features slower actuator). However, as the number of commands in a line increases, it manages to catch up with the competitor showing perfect firmware optimization.

And in this pattern, Maxtor's solution turned out the fastest of all. However, WD400EB didn't fall far behind it (we saw almost the same situation in WinBench tests). The remarkable thing about it is the low results shown by WD400AB! We wonder what cooled it down so greatly then?

But in DataBase pattern WD400AB again got at the head, leaving the rest of the group far behind. WD400EB and Maxtor D540X-4K again ran nearly neck and neck, with Maxtor HDD being a little bit ahead.

Temperature

To measure the temperature of the working HDDs we resorted to DTemp utility. The room temperature stayed around 22^oC, and the measurements were made after more than 2.5 hours of IOMeter tests running.

The working temperature for WD400EB made 36 degrees Centigrade, which corresponds to the typical temperature of contemporary single-platter 5,400rpm HDDs.

Conclusion

WD400EB HDD performed just perfectly in WinBench in both: Windows98 and Win2000. In IOMeter benchmarks it wasn't that fast, but it proved at least not at all worse than any other good 5,400rpm hard drives.

This way, we can state that we have now got a new leader among desktop hard disk drives with 5,400rpm spindle rotation speed. But will it stay there for long? After the model for the Value market segment, Western Digital will also launch a new HDD for the mainstream sector (will it be WD800AB, by any chance?) that is why the leader may change again. It is quite possible that Seagate will also launch something new to replace its U6 solution (which loses more and more with every new test session looking very pale against the background of competitors like our today's hero).

Well, what helped WD400EB to win this race? We have two possible answers to this question: firmware and data density. Though the correct answer seems to be "a combination of good firmware with high data density". :)

As for the per platter data density by WD400EB, it is worth dwelling on. As we have already mentioned, we saw the 39MB/sec read rate by a HDD with 5,400rpm spindle rotation speed, although for WD hard drives this data density is not typical. Though the situation has started changing little by little lately. The newest Western Digital hard disk drives, WD1000BB and WD1200BB, show 43MB/sec and 45MB/sec DTR (Beginning) respectively (note that the last HDD is built of 3 40GB platters).

It is quite possible that WD400EB is based not on 2 platters, as the specs list on the company site states, but on one 40GB platter with very high data density on the first tracks, untypical for 5,400rpm HDDs.

Does it increase the reliability of the HDD? It's hard to say. On the one hand, it allows reducing the number of platters/heads used, but on the other, we all know very well what this hunting for records may lead to.

In conclusion we would like to discuss one more thing with you. Sometimes when reading different reviews we come across some complaints about not very beautifully shaped linear read speed graphs accompanied by some explanations about "bad platter surface quality". As we have already taken the whole crazy lot of graphs like that from the entire bunch of HDDs working on different UDMA-controllers, we have to admit that statements like that are somewhat "irresponsible".

Pretty often we saw that fixed size of Windows swap-file or updated drivers for the UDMA-controller made a really magic effect on the platter quality!

But there should always be exceptions to any rule (which prove the rules, as the proverb says). Today we will show you three linear read graphs for WD400EB HDD with one very clear peculiarity:

UDMA100: Promise Ultra100


UDMA66: Promise Ultra66


UDMA33: i440BX

Note that the local "problem" is very closely connected with some particular place on the HDD and can be observed for all UDMA-protocols.

We believe that only in this case we can speak about some relocated sectors or groups of sectors.