Tech-Hounds.com

Hitachi Deskstar 7K160

SATA II hard disk drive
7200 rpm, 160 GB

Evaluating hard drive in relation to games and of course, game performance is pretty difficult, but not impossible. It's just a question of common sense - when is the hard drive used by games? The answer of course is when the game is reading data like when it loads a level, and saves data, like when you're saving a save game (or the game saves an autosave / checkpoints). There is another possibility, texture / object loads during gameplay, but that's much harder to predict and reproduce. They are also signs that the system doesn't have enough RAM so the operating system has to swap memory from RAM to virtual memory. As our article, "The Quest of Lag Free Gaming" Part 1 and Part 2 showed, in these instances, it's much better to add more RAM than it is to get a faster hard drive.

From our experience with that article, we saw indications that hard drive performance does matter in the two example  instances - loads and writes, particularly loads. So, having a faster hard drive in addition to a bigger one is one of the concerns of many users and gamers today. A bigger, faster hard drive will hold more game installations, downloaded patches and even ISO images for virtual drives and also allows the system to access these files much faster - contributing to a better gaming experience though not necessarily as quantifiable as frame rate.   

There is a relatively cheap way to have both capacity and higher transfer rate. Most motherboards and chipsets come with built in RAID 0 / 1 support these days. So, in addition to testing single drive performance, gamers will often pay attention to RAID 0 (stripe) performance. Today, we're going to run these tests on the Hitachi Deskstar 7K160. For informations purposes, we will be comparing the results with our mainstay hard drive of choice, the Maxtor DiamondMax 10. For RAID 0 array test, we left the default NForce RAID IDE settings at their default settings (optimal).

Overview

We usually don't run synthetic benchmarks, but in this case we feel it's important to pinpoint areas we want to focus on. Since we're focusing mainly on gaming performance, we're going to focus on the Hitachi Deskstar 7K160's read performance. In that regard, it would be very helpful to know the factors involved such as transfer rates, access time and CPU utilizations. Since measuring these aspects of storages is nearly impossible (or at least very hard to do) with real world games, thus the use of synthetic benchmarks. Rest assured though, these synthetic benchmarks are not the main influencing factor in our evaluation. We will still rely on games for that purpose. Let's take a look at the results. 

HD Tach



The Hitachi Deskstar 7K160 is definitely faster than our DiamondMax 10. Having SATA II means burst speed are much higher, in fact it's almost twice what our DiamondMax 10 can offer. Sequential read and write are also significantly higher and that's a good thing. With fairly about the same access time and CPU utilization, no doubt you'll appreciate  the additional speed and space. Overall, the Hitachi Deskstar is the better drive here.



With RAID 0 (stripe), you practically almost double your hard drives' transfer rate for both read and write. Burst speed jumped so high, that the array of two Hitachi Deskstar can offer transfer rate slightly higher than what you get from a single SCSI Ultra 320. Our two DiamondMax 10 also got a healthy dose of additional transfer rate, but it's still far from the levels of the two Hitachi Deskstar. However, using a RAID array is not without some disadvantages. According to this test, CPU utilization are much higher - jumping from 2 - 4 % to around 13 %.

However, there seems to be something quite odd about the two Hitachi Deskstar RAID 0 array results. Read transfer rate didn't kick in until we get to around 125 GB and write transfer rates are really odd, to say the least. As you can see, we don't have such a problem with our DiamondMax 10 drives. This is probably why both the average read and write speed from these two arrays are very close, in fact write is better with the DiamondMax 10 array.



Looking closer at the results, even with this odd behavior, the Hitachi Deskstar RAID 0 array is still able to offer read higher and close to the two DiamondMax stripe array. That explains the still higher average read from the Hitachi Deskstar array results - 116 MB/s compared to 97 MB/s. Slighty lower access time no doubt also contributed to this fact.

IOMeter

With such an odd showing of results from HDTach, we needed another synthetic benchmark to examine these drives. We used four configurations for IOMeter test - 100 % Sequential Read / Write and !00% Random Read / Write. This should give us an idea of what peak performance these drives can offer. The results are in the tables below. the first group of results are for reads and the second are for writes.

Sequential -100%
Read	Read IO/s	Read MB/s
Deskstar array	3063.15	95.72
Diamondmax array	6639.32	103.74
Deskstar single	2325.19	72.66
Diamondmax single	1807.57	56.49
Write	Write IO/s	Write MB/s
Deskstar array	3738.3	116.82
Diamondmax array	5679.23	88.74
Deskstar single	2314.38	72.32
Diamondmax single	1828.36	57.14

Sequential -100%
Read	Average Response Time	Maximum Response Time	% CPU Utilization
Deskstar array	0.33	16.85	8.81
Diamondmax array	0.15	10.12	13.18
Deskstar single	0.43	17.06	3.44
Diamondmax single	0.55	167.67	4.42
Write	Average Response Time	Maximum Response Time	% CPU Utilization
Deskstar array	0.27	1.9	7.01
Diamondmax array	0.18	8.18	9.2
Deskstar single	0.43	10.27	3.85
Diamondmax single	0.55	17.16	2.97

You'll no doubt notice the results from IOMeter are different than HDTach save for CPU utilization. Results are generally higher here, but it's relatively pretty close. As a single drive, the Hitachi Deskstar is again the better drive by a significant margin. Both in average read and write, the Hitachi Deskstar 7K160 is faster by almost 50% than DiamondMax 10. No doubt, the drive's support for SATA II allows it to maintain the same CPU utilization and average response time, with lower maximum response time. That in means higher I/O operations and that means higher transfer rate for both read and write. However, once we use them in a stripe array, the same odd thing we saw with HDTach rears its head again. The Hitachi Deskstar is actually beaten in average read speed, though it's faster in average write.

Random – 100%
Read	Read IO/s	Read MB/s
Deskstar array	69.2	2.16
Diamondmax array	68.73	1.07
Deskstar single	71.94	2.25
Diamondmax single	67.14	2.1
Write	Write IO/s	Write MB/s
Deskstar array	159.87	5
Diamondmax array	164.38	2.57
Deskstar single	112.33	3.51
Diamondmax single	96.46	3.01

Random – 100%
Read	Average Response Time	Maximum Response Time	% CPU Utilization
Deskstar array	14.45	25.68	0.48
Diamondmax array	14.55	27.09	0.26
Deskstar single	13.9	29.69	0.47
Diamondmax single	14.89	30.65	0.32
Write	Average Response Time	Maximum Response Time	% CPU Utilization
Deskstar array	6.25	116.54	0.58
Diamondmax array	6.08	36.93	0.36
Deskstar single	8.9	269.15	0.57
Diamondmax single	10.37	37.22	0.28

In this synthetic 100 % random test, we're not really looking at transfer rate - we're focusing on I/O operations and access times. The Hitachi Deskstar manages a slightly higher I/O operations per second, mostly because of lower average response times. It does have a higher CPU utilization in both random read and write - almost twice in write in fact. It also has a very high maximum response time with random reads, but that's not that bad since it still manage a very respective lower average response time overall. In a RAID 0 configuration, the Hitachi Deskstar and Maxtor DiamondMax 10 are evenly matched, though the DiamondMax 10 have an advantage because of its lower CPU utilization.

These results are a little confusing to say the least. However, there's is a trend we can spot from these results. Overall, as a single drive the Hitachi Deskstar 7K160 is fast and definitely better than our current staple hard drive, the Maxtor DiamondMax 10. All the statistics are better - transfer rates, access time and CPU utilization for both read and write. Put two of these drives into a RAID array, you still have better read performance overall, but not necessarily so for write operations.

Performance

To measure real world performance, we choose to measure the time it takes for a game to load a level. Although measuring save game performance is valuable from this article standpoint, it's very difficult to do and in games like Quake 4, saving a game involves more than just the hard drive because the system have to make a thumbnail shot of your current view. We choose three games for this test - F.E.A.R, Quake 4 and Serious Sam II. Because of the way the game works (caching some elements in RAM and virtual memory), we ran test for both first time load and reloads.

We chose both F.E.A.R's and Quake 4's opening cinematic simply because these levels are generally much larger than the average levels for both games. For Serious Sam II, we choose the final Mental Institution level for the same reason. Measurements are taken with FRAPS, from the time the game starts loading the level to the very beginning of the cinematic. Since we're measuring load times, the measurements are not frame per second, but load time (in msec).

We'd like to thank Tagan for supplying the additional power supply for this article.

Our test setup
AMD Athlon 64 3500+ socket 939
2 x 1024 MB Kingston KVR 3-3-3 PC3200 DDR-SDRAM
MSI K8N NForce 4 SLI motherboard
Leadtek WinFast GeForce PX7900GS 256 MB
Maxtor DiamondMaxPlus9 80 GBs Serial ATA 8 MB buffer
ASUS E-616 DVD-ROM
Tagan TG530-U15 530 watts ATX/BTX power supply

Windows XP Professional with Service Pack 2 installed
NVIDIA Forceware 84.21 reference driver
NVIDIA NForce 4 6.66 reference driver
Creative SoundBlaster Live! 24 bit 5.12.1.512 driver.
DirectX 9.0c

The graphs are pretty self explanatory, but in case you can't see the text, they're arranged in groups of three colors: green for first test run,  blue for the second test run and red for the third test run. Results in black are the average of three runs. The purpose of including all three runs is to see how much variability these test have.  Results are in milliseconds, for example 25821 really means 25.821 seconds. 

The results:

No doubt the results that's more representative to real world performance are the first run results. The game spent most of its time fetching data for the first time from the hard drive. There are some write operations involved, since it's pretty likely not all the data is written to RAM but instead stored in a cache on the hard drive. The higher read transfer rates from the Hitachi Deskstar 7K160 is paying dividends here. Using a stripe array gives us a shorter load time, though not by much (4 seconds).

Consecutive loads pretty much confirms the trend we see with first run loads. Though there are more variations between runs, its obvious the Hitachi Deskstar 7K160 is still much faster than our Maxtor DiamondMax 10, both as a single drive and in a RAID 0 array. With the Hitachi Deskstar 7K160, we basically spent no more than 20 seconds, while with the DiamondMax 10 we spent more than 20 seconds.


With Quake 4, we again see first time loads are much shorter in duration with the Hitachi Deskstar 7K160 than our DiamondMax 10 drives. That applies when we ran these hard drives as single drives or in a RAID 0 array.  Unlike F.E.A.R, we only save 2 seconds overall with a stripe array and with such a long load, it's actually a moot point. Results with consecutive loads on both drives, as a single drive and in a stripe array are pretty much in the same, so it's obvious the hard drive is not much of factor here.


Serious Sam II is probably an example of a wrinkle in the Hitachi Deskstar 7K160 rapport. Though results with a single drive configuration is pretty much the same with the Maxtor DiamondMax 10, using two Deskstar 7K160 in a array will actually get you longer loading times in this game, with this particular level. Though that might be just the twitch in our fingers, it's unlikely so since even the shortest results with the Hitachi Deskstar 7K160 RAID 0 array is almost the same as the longest results with other configurations.

With these results, we can see that hard drive performance is still a factor, even when you're playing games. With more graphically intensive games and environments, the amount of data used will grow and the slower your hard drive, the longer you have to wait when loading a level. True, not all games will behave the same way - F.E.A.R is definitely the most sensitive to hard drive performance out of the three games we use in this test.  It's actually sensitive enough for RAID 0 (stripe) to matter.

Conclusion:

Hard drives are important, they are the mainstay of storage for computers everywhere for as long as they are computers. Though many changes have occurred during the last few years, the most important factors regarding hard drive performance is relatively constant - transfer rate, access time and CPU utilization. The Hitachi Deskstar 7K160 has a healthy helping in those department, enough for gamers who want shorter load times and overall better storage performance.

You might say, does 2 - 4 seconds shorter load times matter? Well, that actually depends on your hard drive usage. It may not be if you are the non-quintessential gamer who follows developer guidelines and not run anything else in the background, that repeatedly reads and writes to the hard drive. However, we now live in a world where we are constantly connected. When we play games, it's very likely we will also be listening to songs, downloading some files and the all important instant messaging clients. Like any other software and the game you're playing, these background software is accessing the hard drive. In this type of situation, hard drive statistics like burst speed, access time and CPU utilization are small yet important factors. Having a better hard drives in regard to those statistics might just help.

Go to top