Samsung F1s are the top performing 7200RPM desktop drives of the day. We put the 750G and 1TB models through their paces in SPCR’s acoustics chamber to find out whether they can take the low noise crown.
June 10, 2008 by Mike Chin
Products
|
Samsung F1_3D (Model HD753LJ)
750GB 7,200 RPM Desktop Hard DriveSamsung F1_3D (Model HD103UJ) 1000GB 7,200 RPM Desktop Hard Drive |
Sample Supplier
|
|
Market Price
|
US$110~160 and US$170~299 |
Samsung, along with Western Digital, has been the leader for quiet hard drives in recent years, so a new line from this brand is always greeted with some anticipation at SPCR. The F1 series hard drives have been available (at least in Canada and the US) for a few months. The newest of the company’s 7200 rpm desktop hard drives, the F1 series boasts an industry-leading 334GB maximum capacity per platter, which promises very high performance. This was 33% higher areal density than its closest competitors when the line was first released. Now, WD has nearly caught up with a couple of new models in their SE16 Caviar line that offer 320GB per platter.
Many capacities are offered in the F1 line: 160GB, 250GB, 320GB, 500GB, 640GB, 750GB and a terabyte. It seem unlikely that 334GB platters are used in all models. As flagship drives, the 750GB and 1000GB models offer 32mb cache, compared to 16mb or 8mb for the smaller capacity drives. Both are said to be 3-platter drives, perhaps with the 750GB utilizing just a portion of one platter, although there’s some question whether the 750GB models have three 250GB platters.
Despite repeated requests for samples of the new models, Samsung is no longer making samples available to reviewers in North America. Samsung Canada has not provided any HDD samples for about a year. Our F1 samples came by courtesy of Anitec Computers, a Vancouver Canada retailer and sponsor of SPCR. The 750GB came first, a couple of months ago, and then we waited for the terabyte model to include it in the same review. The latter has been in high demand, with short supply. One amazing thing about waiting: The price dropped dramatically over these few months, and even the terabyte model is now being sold by Anitec for just $175!
Samsung F1- 750GB/32mb cache model — New and improved?
The terabyte model looks identical except for the label.
Samsung F1 series hard drives
(from Samsung’s product overview) |
|
FEATURE & BRIEF | COMMENT |
Max. 334GB Formatted Capacity per Disk | Higher areal density means faster performance, all other things being equal. This is the highest for now, but this sector is constantly changing. |
Improved recording stability over temperature with PMR | Interesting claim for Perpendicular Magnetic Recording technology, which is quickly becoming standard. |
Advanced dynamic Flying-On-Demand (FOD) control for best data integrity | Here’s what FOD is said to be: control technology ensures enhanced read/write sensitivity by dynamically lowering the fly-height in read or write mode while maintaining a consistent and higher fly-height across the disk in standard mode. |
Intelligent compensation of external disturbance | Hmmm… Like letting the answer machine get the phone when you’re having dinner?? |
SATA Native Command Queuing Feature | Fairly common these days, and not that useful for desktop PCs. |
Staggerd Spin-up Support | Nice for multiple HDD setups. |
Improved performance with dual-ARM based firmware | ?? |
Automatic Acoustic Management Feature, NoiseGuardâ„¢, SilentSeekâ„¢ | Obviously, the proof of noise reduction features is in the listening and acoustic measurements in our review. |
Early performance-oriented reviews of the 1000GB Samsung F1 gave it very high marks, with transfer rates surpassing every other desktop drive, including the famed 10,000 RPM Raptor drives from WD. (See review at The Tech Report) The question remains whether the 750GB model uses three 250GB platters (which means performance would be as good as most of the competition but not up to that of the 1,000GB F1) or three 334GB platters with portions disabled. Without direct access to technical people at Samsung, this is not an easy issue to settle. The only definitive way we know of is to compare the throughput of the 750GB and 1,000GB models. If the two drives measure the same (or very close) then it’s safe to say they have the same areal density. If the 750GB drive is slower, then it would be using 250GB platters. This is why we waited for the 1TB model, to compare the performance of the two.
SPECIFICATIONS
The specifications below are specific to models that we examined. Capacity,
cache size, platter number, interface, and performance can vary from model
to model. Don’t assume they’re the same in a given line. Acoustics and power dissipation
also vary depending on the number of platters in the drive; smaller capacity
drives tend to have fewer platters, and tend to produce less noise and use less
power.
Specifications: Samsung F1 750 GB & 1000 GB hard drives (from Samsung’s product overview) |
||
Capacity |
750GB
|
1000GB
|
Cache |
32MB
|
32MB
|
Disks / Heads |
3 / 6
|
3 / 6
|
Interface | SATA 3Gb/s | SATA 3Gb/s |
Spindle Rotation Speed |
7,200 RPM
|
7,200 RPM
|
Data Transfer: Media to/from Buffer (Max) | 175 MB/sec | 175 MB/sec |
Data Transfer: Buffer to/from Host (Max) | 300 MB/sec | 300 MB/sec |
Average Seek |
8.9 ms
|
8.9 ms
|
Average Latency |
4.17 ms
|
4.17 ms
|
Weight |
640 g
|
640 g
|
Power Requirements: Idle / Seek (avg.) |
7.7 / 8.6 W
|
7.7 / 8.6 W
|
Acoustics: Idle / Seek |
2.7 / 2.9 bels
|
2.7 / 2.9 bels
|
TEST RESULTS
Our sample was tested according to our standard
hard drive testing methodology. The test drive sample was compared against many other drives. Our methodology focuses specifically on
HDD noise, and great effort is taken to ensure it is comprehensively measured
and described. Performance is not tested, for reasons discussed in detail in
the methodology article.
For comprehensive HDD performance testing results,
we recommend Storage
Review, who have established a long reputation as the specialist in
this field. At the time of writing, they had not yet reviewed any of the Samsung F1 series. As mentioned earlier, The Tech Report ran a thorough performance analysis on the terabyte flagship in February.
Two forms of hard drive noise are measured:
- Airborne acoustics
- Vibration-induced noise.
These types of noise impact the subjective
perception of hard drive noise di fferently depending on how and where the drive
is mounted.
Both forms of noise are evaluated objectively and
subjectively. Both the subjective and objective analyses are essential to understanding
the acoustics of the drives. Airborne acoustics are measured using a professional
caliber SLM. Measurements are taken at a distance of one meter above the top
of the drive using an A-weighted filter. Vibration noise is rated on a scale
of 1-10 by comparing against our standard reference drives.
A final caveat: As with most reviews, our comments
are relevant to the sample we tested. Your sample may not be identical. There
are always some sample variances, and manufacturers also make changes without
telling everyone.
Ambient conditions at time of testing were 18 dBA and 20°C.
DRIVE NOISE EVALUATION
|
||||
Drive
Mfg date firmware version |
Vibration
1-10 (10 = no vibration) |
Activity State
|
Airborne Acoustics |
Measured Power |
Samsung F1 3D
HD753LJ 750GB February 2008 firmware 1AA01109 |
6
|
Idle
|
21 dBA@1m |
6.9 W |
Seek (AAM)
|
22 dBA@1m |
8.9 W |
||
Seek (Normal)
|
22 dBA@1m |
10.2 W |
||
Samsung F1 3D
HD103UJ 1TB February 2008 firmware 1AA01109 |
5
|
Idle
|
21 dBA@1m |
7.0 W |
Seek (AAM)
|
22 dBA@1m |
8.8 W |
||
Seek (Normal)
|
22 dBA@1m |
9.9 W |
||
REFERENCE DRIVES*
|
||||
Seagate Barracuda
7200.11 1TB ST31000340AS December 2007 firmware SD15 |
3
|
Idle
|
23~24 dBA@1m |
7.8 W |
Seek (Normal)
|
24~25 dBA@1m |
11.6 W |
||
WDl Caviar SE16 500GB WD5000KS
March 2006 firmware 07.02E07 |
4
|
Idle
|
21 dBA@1m |
8.5 W |
Seek (AAM)
|
21-22 dBA@1m
|
8.6 W
|
||
Seek (Normal)
|
23 dBA@1m |
10.7 W |
||
Western Digital Caviar Green Power 750GB WD7500AACS
October 2007 firmware 01.01B01 |
7
|
Idle
|
19~20 dBA@1m |
3.3~5.9 W |
Seek (AAM)
|
19~20 dBA@1m
|
5.4 W
|
||
Seek (Normal)
|
21 dBA@1m |
6.7 W |
||
Hitachi Deskstar 7K1000 1TB HDS721010KLA330
June 2007 firmware GKAOA70M |
5
|
Low RPM Idle
|
20-21 dBA@1m
|
3.6 W
|
Idle
|
24 dBA@1m |
8.5 W 6.2 W (unloaded) |
||
Seek (AAM)
|
24 dBA@1m
|
11.7 W
|
||
Seek (Normal)
|
26-27 dBA@1m |
15.4 W |
||
Seagate Barracuda IV
40GB – ST340016A – firmware 3.10 |
6
|
Idle
|
20 dBA@1m |
6.7 W |
Seek (AAM)
|
23 dBA@1m |
11.3 W |
||
Seek (Normal)
|
25-26 dBA@1m |
11.6 W |
||
Samsung Spinpoint P80 (Nidec motor) 80GB
June 04 firmware TK100-24 |
4
|
Idle
|
21 dBA@1m |
6.3 W |
Seek (AAM)
|
23-24 dBA@1m |
8.3 W |
||
Seek (Normal)
|
25-26 dBA@1m |
9.1 W |
||
* These drives are references in the sense of previously tested and known entitities to which newly tested products can be compared and contrasted against.
|
The F1 750GB drive is the quietest Samsung we’ve encountered so far, surpassing all previous samples by at least a small margin. It reaffirms Samsung’s quiet leadership among 7200 RPM drives, very slightly edging the WD SE16 500GB sample we tested two years ago for acoustics, and going a notch better for vibration.
At idle, the sample measured at about the 20~21 dBA@1m level of the very quietest 7,200 RPM drives. The increased noise in seek is very slight, but audible. Turning Automatic Acoustic Management on had a very minor effect on seek noise, changing it slightly in quality but not in measured SPL. It was difficult to judge which was the quieter; the potential reduction in performance is not worth the possible improved noise here.
The 1000GB model was just about identical, judging qualitativiely as well as by measurements. Only the vibration was slightly higher, and this is probably well within sample variance. Both of these samples will benefit from soft mounting in a quiet system, like most hard drives.
The F1s don’t quite reach the quiet levels of the WD Green Power drives, which have the natural advantage of a lower 90Hz fundamental frequency because of their lower 5400 RPM spindle speed. Human hearing acuity drops as frequency drops, so even if the GP drives emit the same amount of acoustic energy, we simply don’t hear it as well as the higher 120Hz fundamental tone of 7200 RPM drives. You do pay a performance price because of that slower spindle speed.
Our measured maximum power was a bit higher than the 8.6W claimed. We were recording the highest peaks; the average power during seek was probably as claimed. Idle power was a bit lower than other 7,200 RPM drives of similar capacity. These Samsungs have only three platters, compared to the 4 and 5 platters of similar capacity drives we’ve reviewed in the past, which helps keep the power consumption down. The Hitachi Feature Tool, used to change acoustic settings and obtain other information about the drives, had no trouble correctly reading the 32 MB cache of the drives, an issue we saw with the 1TB Seagate Barracuda 7200.11.
PERFORMANCE: 750GB vs 1000GB
We mentioned that some users have been asking whether the 750GB and 1000GB models of the F1 have the same areal density. Based on our HD Tach test results, the answer is no. It would appear that the 750GB model does use 250GB platters instead of the 333GB/platter of the terabyte model.
The sequential read of the terabyte drive is 120 MB/s to 60 MB/s. The 750GB drive’s sequential read is 97 MB/s to 50 MB/s. Average read rates, 750 GB = 77.4 MB/s – 1,000 GB = 93 MB/s. Thats a 20% difference at the start, the end, and the average, which makes it nearer to the theoretical performance difference of 33% that can be attributed to the platter density/drive capacity 250 + 33% = 333 …… 750 + 33% = 1000. The drive controller may actually be bottlenecking the 333GB platter drive, as it should theoretically be 33% faster than the 250GB platter drive, when it is only 20% faster.
If they are short stroked platters, they would have the same performance at the start, but a higher average as the 750GB mark on the 1TB drive shows a read rate of 80 MB/s. Likewise if Samsung used a short stroke design that cut the front off of the platters and not the back, then the back end of the drive would have the same read rate as the 1TB model, 60 MB/s and not 50 MB/s.
Note: Thanks to AndyB for help with this analysis.
HD Tach results for the 750 GB model. Click on image to enlarge.
HD Tach results for the 1000 GB model. Click on image to enlarge.
AUDIO RECORDINGS
Audio recordings were made of the drives and are presented here
in MP3 format. The recordings below contains 5 seconds of ambient noise, and
10 seconds of idle noise followed by 10 seconds of seek noise with AAM enabled
and 10 seconds without.
Keep in mind that the recordings paint only part of the acoustic
picture; vibration noise is not recorded, and drives often sound different depending
on the angle from which they are heard.
These recordings were made
with a high resolution, studio quality, digital recording system, then
converted to LAME 128kbps encoded MP3s. We’ve listened long and hard
to ensure there is no audible degradation from the original WAV files
to these MP3s. They represent a quick snapshot of what we heard during
the review.
Most recordings are made
from a distance of one meter. The recording is intended to give you
an idea of how the subject of this review sound in actual use — one
meter is a reasonable typical distance between a computer or computer
component and your ear. The recording contains stretches of ambient
noise that you can use to judge the relative loudness of the subject.
For best results, set your volume control so that the ambient noise
is just barely audible. Be aware that very quiet subjects may not be
audible — if we couldn’t hear it from one meter, chances are we couldn’t
record it either!
Occasionally, we may include
an extra recording from a distance of one foot. This recording is designed
to bring out the fine details of the noise — typically when the
source is so quiet that it is very close to ambient when heard from
one meter. Use this recording with caution! Although more detailed,
it does not represent how the subject sounds in actual use. It is best
to listen to this recording after you have listened to the one meter
recording.
More details about how
we make these recordings can be found in our short article: Audio
Recording Methods Revised.
Only one of the F1 drive samples were recorded since they sounded identical acoustically.
- Samsung F1 750 GB (HD753LJ) — Idle: 20~21 / Seek:
22 dBA@1m / AAM: 22 dBA@1m— One
Meter
Comparatives:
- Seagate Barracuda 7200.11 ST31000340AS — Idle: 23~24 / Seek:
24~25 dBA@1m — One
Meter - Western Digital Caviar Green Power WD7500AACS — Idle: 19~20 /
AAM: 19~20 / Seek: 21 dBA@1m — One
Meter - Hitachi Global Storage Deskstar 7K1000 HDS721010KLA330 — Idle:
24 / AAM: 24 / Seek: 26~27 dBA@1m — One
Meter - Hitachi Global Storage Deskstar 7K1000 HDS721010KLA330 — Low
RPM Mode: 20~21 dBA@1m — One
Meter - Samsung T Series HD400LJ — Idle: 20 / AAM: 21 Seek: 22-23 dBA@1m
— One Meter - Samsung P80 SP0802N (Nidec) — Idle: 21 / AAM: 23-24 Seek:
25-26 dBA@1m — One
Meter - Seagate Barracuda IV ST340016A — Idle: 20 / AAM: 23 Seek:
25-26 dBA@1m — One
Meter
CONCLUSIONS
The Samsung F1 750GB and 1000GB drives are the quietest 7200 RPM drives currently available, by a small margin. Their overall noise signature is quite benign, with very low tonality; the mostly broadband sound easily blends in with other noise in a PC. The other high capacity 7200 RPM drives we’ve tested don’t come anywhere close to this level of quiet. That the Samsung 1TB F1 drive should also set new highs in performance is icing on the cake. Only the 5400 RPM WD Green Power drives offer better acoustics, but with nowhere near the performance.
The vibration level on both samples is quite modest, lower than in previous Samsung samples, which have often been a bit on the high side. These drives are quiet enough and vibrate little enough that elastic string suspension is not a mandatory recommendation, except perhaps for silent PC diehards. A soft grommet mounting will probably provide good enough isolation to keep vibrations from adding much of their typical 120 Hz low pitched hum.
The pricing of these Samsung drives is remarkable: They’re amazingly good value, considering that they have huge capacity, the best 7200 RPM performance (for the 1TB model), and the lowest noise. The only 3.5″ drive that’s clearly quieter is the 5400RPM 1TB Green Power by WD, and while it has good throughput, its random access speed is far worse, and it is still selling typically for $230~$300. You might find one on sale, for about $200, rock bottom. In contrast, the 1TB Samsung F1 can be found quite easily for under $200, as low as $172. (So our sample provider Anitec’s current CA$176 price is a very good deal.) The 750GB is much cheaper, typically $100~$120.
Western Digital’s 320GB/platter SE16 Caviar series is currently limited to 320GB and 640GB models. The latter has already been reviewed by The Tech Report as a viable competitor for the Samsung F1 series for both performance and noise. We’ve examined the 320GB single-platter model (and found some disturbing variations in samples) and we currently have a sample of the 2-platter 640GB model on the test bench. Presumably, the WD 640GB model should outperform the Samsung F1 750GB. This WD can be found for under $100; hopefully, WD will soon produce a 1TB 3-platter model version to compete with the Samsung 1TB F1.
In the meanwhile, the Samsung 1TB F1 looks like the winning combination of high performance, low noise and affordable high capacity on the market today.
Many thanks to Anitec Computers of Vancouver Canada
for the Samsung F1 hard drive samples.
* * *
SPCR Articles of Related Interest:
SPCR’s Hard Drive Testing Methodology
SPCR’s Recommended Hard Drives
WD Green Power: A New Benchmark in HDD Acoustics
& Power
The Terabyte Landmark: Hitachi Deskstar 7K1000
Seagate Barracuda 7200.11 Terabyte drive
WD Caviar SE16 single platter 320GB drive
* * *