1TB and 5400 tpm: Western Caviar Green vs Samsung EcoGreen - BeHardware
>> Hard Disks

Written by Marc Prieur

Published on January 23, 2009

URL: http://www.behardware.com/art/lire/750/

Page 1


For a long time we hadn't seen any 5400 rpms but they have now made something of a come back. Western were the first to signal the return with its Caviar GP range launched in 2007, when it chose to make a virtue out of what had previously been slightly shameful.

However, the first tests showed that performances weren’t that far behind the 7200 rpm models and this with a significant advantage in terms of lower sound levels and energy consumption.

Not necessarily cheaper than their 7200 rpm equivalents, these 5400 rpm drives are an alternative with advantages and disadvantages that we’re going to analyse by looking at the ranges of two manufacturers, Western Digital and Samsung.

The influence of rotation speed
Rotation speed has a direct impact on performance, both in terms of throughput and access times. In terms of throughput, it’s relatively simple: 4 rotations at 7200 rpm take the same time as 3 at 5400 rpm, a net gain of 33% in terms of throughput in theory, on same-density platters. Of course, although at the moment 5400 rpm drives use comparable densities to 7200 rpm drives, in time the 5400 rpm drive may well become denser, which would counterbalance the gain in throughput.

Rotation speed also has an impact on hard drive access times. To gain access to data, first the reading head has to get itself onto the right track on the platter, something that is not linked to rotation speed: this is what manufacturers usually call access time on their spec sheet.

However the head still needs to access the right piece of data, which might be further down the track. So you have to wait ¼, or ½ or whatever turns for the read head to be above the piece of data it’s looking for. This is what is known as latency, and the value given normally corresponds to ½ turn: 5.55 ms at 5400 rpm, 4.17 ms at 7200 rpm, 3 ms at 10 000 rpm.

Therefore, if all else is equal, a 5400 rpm drive will have a shortfall of around 1.38 ms in terms of access time in practice. Not a huge amount obviously, but to be taken into account all the same.
Perfect for some types of use
5400 rpm drives are a good choice for several types of use. The first, of course is for a machine that doesn’t need the highest performance but a significant storage capacity and good environmental spec, such as a Home Cinema PC. But they will also serve well in a NAS.

They also have their place within a high-performance PC, as a secondary storage facility. You could for example use a VelociRaptor, or even an SSD, for your everyday applications and a 5400 rpm for your data: your holiday Blu-ray and the 100 daily photos of your youngest can easily be stored here!

Page 2
The drives, the test

The drives
For this test, we used 5 1 TB disk drives:

- Samsung SpinPoint F1 HD103UJ (7200 rpm, 32 MB cache)
- Samsung EcoGreen HD103UI (5400 rpm, 32 MB cache)
- WD Caviar Green WD10EACS (5400 rpm, 16 MB cache)
- WD Caviar Green WD10EADS (5400 rpm, 32 MB cache)
- WD Caviar Black WD1001FALS (7200 rpm, 32 MB cache)

The 5400 rpm models are then systematically placed opposite the 7200 rpm versions from the same manufacturer. All these drives use 3 platters with a density of 333 GB each, including the WD10EACS. At first, this model used 4 platters of 250 GB per platter, but Western changed it on the quiet, not distinguishing between both models: a habit the constructor has long since adopted, in contrast to the others who do not generally mix surface densities on products with the same retail reference!
Test protocol
Various measurements were carried out during this test. First of all we were interested in the “synthetic” performances of these drives: cache and sequential speeds and average acces time. Next, were more practical tests, first of all involving an applicative performance index based on PC Mark Vantage and then a server load type simulation of files with IOMeter. This was followed by an evaluation of writing, reading, close (on the same partition) and far copying (on a partition which starts on 50% of the drive) with various groups of files.

These files are composed as follows:

- Large: 6 files (on average 2.2 GB) totalling 13.2 GB
-Medim sized: 7.96 GB of 10,480 files (each averaging 796 KB)
-Small sized: 2.86 GB of 68,184 files (each averaging 44 KB)

The source or target of reading or writing on the drive was a RAID of two Raptor 150 GB drives so as to make sure we weren't limited. This type of measurement is worthwhile because, while the sequential speed gives us an idea of the performance in copying large files, things can be different with smaller ones.

All measurements were taken without noise level reduction but we did also carry out some measuremenst afterwards with this option activated. The test machine was based on an X38 chipset mounted on an ASUSTeK P5E motherboard while Serial ATA ports were configured in the bios in AHCI (Advanced Host Controller Interface) so that NCQ could be used.

In addition to performance we also measured temperature after 2 hours of intensive use, as well as energy consumption. Noise pollution of each drive was also evaluated.

Page 3
Synthetic readings

Synthetic readings

We start with the cache speed, measured with the h2bench Core Test. Performances shouldn’t vary between 5400 and 7200 rpm versions here as what is being measured is the access to chip memory integrated in the PCB. Although the Samsung cache is slightly faster, the difference is small.

Also taken with the help of h2bench, access time shows that 5400 rpm drives are not as far behind as all that. The difference of 1.4 to 1.5 ms is to be expected given what we said above about the latency differences of 1.4 ms: the mechanism in terms of reading heads is the same. Whatever the rotation speed, note that the mechanism used by Western is slightly faster than Samsung. Activation of AAM has a much more significant impact with Western – we’ll see what impact it has on noise levels later.

h2bench allowed us to measure the read and write speeds of the drives during sequential access.

The difference between 7200 and 5400 rpm versions is obvious here, with an advantage of ¼ with Western and 1/3 with Samsung. In both reading and writing, the SpinPoint F1 is on average the faster of the two 7200 rpms, while the Western 5400 rpms are slightly ahead of the EcoGreens.

This graph shows speeds according to how far in on the drive you are on 1 TB drives. There isn’t a notable difference between the drives, speeds gradually falling off.

Page 4
PC Mark Vantage

PC Mark Vantage
We now move on to less synthetic tests, starting with an index of hard drive performances in PC Mark Vantage. FutureMark reproduces a set of reading/writing tasks on the drive, namely a Vista start-up, loading of applications (Word, Photoshop, IE, Outlook), the manipulation of multimedia files (photo, video, music), games (loading of Alan Wake) and disk scan with Windows Defender.

Without noise level reduction, the Western Black does best, and performances are not far off the VelociRaptor in fact. 5400 rpm drives are significantly behind, but the 32 MB cache does have a positive impact, as you can see if you compare the two Caviar Greens. Western is ahead for both the 7200 and 5400 rpms.

With these sorts of scores the 5400 rpms really are doing well for themselves and even if they are not as fast as the 7200 rpm drives, in practice the difference is very limited: 9 and 10.5 seconds to launch Photoshop CS4 for the SpinPoint F1 and the EcoGreen respectively.

You can see moreover that it is the Caviar Black that is worst affected by activation of AAM.

Page 5
Management of files

File copying
This brings us to file copying. We measure reading and writing speeds of various groups of files, as well as close and far copying. These groups were composed of:

- A collection of large files: 6 files (on average 2.2 GB) totalling 13.2 GB
-Medim sized: 7.96 GB of 10,480 files (each averaging 796 KB)
-Small sized: 2.86 GB of 68,184 files (each averaging 44 KB)

The source or target in reading or writing on the drive is a RAID of two Raptor 150 GB drives, so as not to be limited by this.

The first surprise was that although the 7200 rpms have an advantage for reading large files, once you drop down to medium-sized or smaller files, performances are very close. Activation of AAM doesn’t really have a negative impact on speeds.

For writing, things are a bit different, with the 5400 rpms performing less well both for large and medium-sized files. Overall, the Samsungs are faster than the Westerns, so much so that for writing small and medium-sized files the EcoGreen is in front of both the Caviar Black and even the VelociRaptor! Here again the impact of AAM is virtually zero.

Several things are worth noting. Whatever the files, the Samsungs are faster, with a clear advantage with the 7200 rpm version in the case of medium-sized and large files. With the Westerns the performance difference due to rotation speed is lower and the 32 MB version is strangely notably slower with large files than the 16 MB. Again with the Westerns, AAM has a significant impact on performance.

Far copying is the most taxing for the drives when manipulating single-user files. Samsungs advantage is less here and once again the Western Green 32 MB and even the Western Black give strange results with large files.

Page 6
I/O Meter

IOMeter is used to simulate the load in a multi-user environment by using a server type file load comprised of 80% reading and 20% writing all in a 100% random manner on the drive. In this type of situation, NCQ can be particularly useful because of multiple concurrent commands. In this test, we measured performances expressed in inputs/outputs per second (IO/s) with 1, 2, 4, 8, 16, 32, 64 and 128 simultaneous commands.

The more simultaneous commands there are, the greater the advantage to Western over Samsung. Although with just a single command the SpinPoint F1 is faster than the Western Green, the Westerns have an advantage afterwards.

With AAM, performances are clearly down, especially the Westerns, which allows the Samsungs to catch up a bit. Although without AAM activated the cache between the 16 and 32 MB versions made a difference with the Caviar Greens, here they give equivalent results. Note that with noise level reduction activated, the SpinPoint was fastest, then overtaken by the Westerns which manage NCQ better.

Page 7
Energy consumption, noise, temperature

Energy consumption
Here is the data for energy consumption for each of the drives that we were able to read with a hook-on ammeter, by reading voltages used on the 5V and 12V lines by each of the models.

There is a clear difference between the 5400 and 7200 rpm models and between the 5400 rpms where the Western does much better than the Samsung at rest, with the in load scores pretty comparable. For the 7200 rpms, the Caviar Black consumes less at rest but more in load.
Next we measured temperatures of the drives after 2 hours of intensive use in IOMeter. These measurements were taken with the HDD outside of the casing, without the fan, each of the drives slightly raised. The temperature was measured in two places using an infrared thermometer, above the middle of the drive and at the hottest point on the left side. We also read the temperature on the internal drive sensor (SMART). As usual the SMART sensor on the SpinPoint F1 under-evaluated the temperatures, something we already know from our previous tests on this model.

Once again you can see the difference between the 5400 and 7200 rpm drives. The Samsung EcoGreen heats a little less (1°C) than the Western Caviar Green, and the same goes for the SpinPoint F1 compared to the Caviar Black with a clear difference (2 to 3°C).
Noise pollution
To take this we placed each drive so as to avoid all vibrations (seeing as they were placed on the desk). A sonometer was placed 10 cm above the drive so as to measure noise pollution, the PC used being without any fan so as to isolate the sound of the drive:

The Caviar Black is too noisy during rotation and access, with levels that compromise it for anyone looking for a quiet PC. The SpinPoint F1 does much better but doesn’t compare to what we got with the 5400 rpms. Although with less of a lead, Samsung still gives the best results with the EcoGreen.

So as to have an idea of the sort of noise emitted by these drives we wanted to record them as usual. Unfortunately, our recording methods are not sensitive enough to pick up the low sound levels of the 5400 rpms and we couldn’t record the levels during rotation or access with AAM activated.

Page 8

At a time when environmental factors are becoming more and more important, the new lease of life for the 5400 rpm is not a surprise. As we saw during the different tests, they do allow you to reduce energy consumption significantly, as well as temperture and noise levels while maintaining pretty good performance levels.

Of course the 7200 rpm drives are still a bit more responsive when functioning as a system drive, although not as good as SSDs which however have low capacity: combining an SSD with a 5400 rpm within the same machine would seem to give the ideal high performance but quiet solution. These 5400 rpms will also have a place within a PCHC or an NAS.

Our preference is for Samsung’s EcoGreen over Western Digital’s Caviar Green, as it has lower sound levels and better write performance. The Caviar Greens do nevertheless have the advantage for simultaneous access, which might be an advantage in a shared NAS. For pure storage, the 32 MB version has nothing on the 16 MB version, and there is really no advantage unless you’re going to use the Caviar Green as the system disk.

As for the 7200 rpm models, the 1TB Western Digital Caviar Black outperforms the Samsung SpinPoint F1 overall. However it heats up and is noisier, which means many will reject it out of hand.

To finish with, we do wish to say that there will no doubt be a lot of new 5400 rpm models this year, with platter density going from 333 to 500 GB, whether Western or Samsung. This ought to make the 5400 rpm solution worth an even closer look!

Copyright © 1997-2014 BeHardware. All rights reserved.