Review: Samsung Spinpoint F1

Silencing hard drives, optical drives and other storage devices

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
LodeHacker
Posts: 628
Joined: Sat Dec 13, 2008 1:25 pm
Location: Finland

Review: Samsung Spinpoint F1

Post by LodeHacker » Mon Feb 09, 2009 1:22 am

DISCLAIMER: This is by no means a professional quality review. Every following bit of information is based on my own experience with said product for 5 hours. Rather than a review call it "first impressions" if you like. Nothing should be taken seriously here and all information is provided as is. Thanks you for understanding.

Samsung Spinpoint F1 HD322HJ

Drive details
320GB capacity (single platter)
16MB cache (great cache algorithms!)
7200RPM speed (great I/O performance!)
3.5" size (normal desktop drive)


1. Holding the Samsung in my hands for the first time
This is my first Samsung hard disk drive and my expectations are great I/O performance and quiet operation. I have a Samsung optical DVD drive, which is one of the fastest to deliver 22X DVD recording speed -- I expect the hard disk drive to be as astonishing as well. Previously I had only Seagate and Western Digital and the occasional Maxtor. Holding the Samsung for the first time in my hands and all I can say... it feels special. Let's get this baby installed, shall we?

2. Testbed and operating environment
Not going to take pictures of my own room for the very reason that it's always messy here! Testbed is my favorite PC consisting of an Intel Core 2 Duo E6300 and NVIDIA GeForce 6800 GS installed in an ASUS P5K WS. The Samsung is going to be installed externally -- using a passthrough SATA cable going to the motherboard and in-between is the Akasa AllInOne multi-functional panel (first revision). Operating environment has an ambient temperature of 22C. Ambient noise is loud, my sister listens to music at loud levels and the TV is very often on, but when everyone else is sleeping and I'm still hacking through the night it's very quiet and that's when I hear my PC components very sensitively.

3. Installation and temperature readings
Installation was a breeze and why shouldn't it be? Although OEM packed the Samsung came with a very nice quick reference sheet showing visually how the hard disk drive should be installed. Cute little pictures of a character posing as a hard disk drive are also included -- small plus for Samsung! Anyway, connect SATA and power cable... hit the power button and off we go! The hard disk drive has been running for 5 minutes copying a big 50GB file (dummy file filled with random bytes). SMART temperature reading is 35C. I tried to stream a DVD quality movie from the hard disk drive and getting halfway through the film (Scary Movie 3) the drive reported 34C temperature. Now the drive is idling and it's running at 33C (was idling at least 30 minutes). Impressive, but SMART data should not be taken seriously so let me add subjective information on the temperatures: when the hard disk drive was copying that big 50GB file it felt just a tiny bit warmer and when idling the drive is only warm, but not hot.

4. *Cough* FINALLY *cough* I/O performance
hdparm 8.9 64-bit, command used hdparm -tT /dev/[s|h]d[a-z]

Samsung Spinpoint F1 HD322HJ
Timing cached reads: 2646 MB in 2.00 seconds = 1323.56 MB/sec
Timing buffered disk reads: 338 MB in 3.00 seconds = 112.62 MB/sec

Seagate Barracuda 7200.11 ST3500320AS
Timing cached reads: 2564 MB in 2.00 seconds = 1282.39 MB/sec
Timing buffered disk reads: 314 MB in 3.00 seconds = 104.56 MB/sec

The Samsung is around 10MB/s faster than my Barracuda. Although it seems like a small difference, I did see a great boost when copying multiple small files and boot up was around 2 seconds faster (measured with a stopwatch).

5. NNNOOOIIISSSEEE!!!
Actually, it IS quiet! Never heard so quiet seeks before! Compared to my Barracuda this is almost inaudible. There's only one complaint: for a single platter hard disk drive I expected it to vibrate less. Much less actually. My Barracuda has two platters, but still vibrates a hair less than the Samsung. Will get a Nexus Drive-A-Way and see how it eliminates vibration (not yet ready to take the plunge and mod my case to allow hard disk drive suspension). Best of all there's no high pitched whine and nothing else but airborne acoustics. I saw immediately a great change in the noise level of my PC when I swapped the Barracuda for the Samsung!

6. Enough bullshit! Hot or not?!
The hard disk drive is HOT! Not as in very warm but you know if you listen to an euphoric trance track you get that hot feeling... *dreaming* -- I mean, it is really a great drive! I've had it only for around 5 hours now and I am overly impressed. Yes the Samsung did deliver excellent I/O performance, but I am a little but disappointed regarding the vibration. I will see how that changes once I put it into a Nexus Drive-A-Way. All in all a great hard disk drive!

7. (Extra) Direct Audio Sampling
Some of you know that I do immense audio recording and use a technique of what is a revised sampling method. Don't have the time to explain it in detail but in essence direct audio sampling reads audio data directly off from the hard disk drive without a buffer in the middle. The sampler translates the audio into a usable format (decodes a special lossless codec) and then fills a quick 2MB buffer. From this buffer the sampler directs audio into the mixer and from there to my audio interface. Using intelligent algorithms the sampler starts to buffer audio data with the highest sample size that is in the range of 2 seconds. Using this special sampling method it is possible to achieve latencies as low as 0ms with dedicated hardware. My top result was 1ms with a Seagate Cheetah 10K.7 SCSI hard disk drive. Using my current Barracuda I can achieve non-crackling audio at 3ms. With an earlier Western Digital it was even higher. Now I got a whole different picture of 7200RPM hard disk drives though, because the Samsung was able to throw in power that allowed 1.8ms recording, getting very near to the speed of the Cheetah and almost twice as fast as the Barracuda! Great, I will definitely put multiple of these in a RAID configuration for pure audio recording and processing.

LodeHacker
Posts: 628
Joined: Sat Dec 13, 2008 1:25 pm
Location: Finland

Post by LodeHacker » Wed Feb 11, 2009 6:33 am

Got the Nexus Drive-A-Way and installed it in my case so that it is vertically placed next to the PCI slots. Airborne acoustics and seeks = GONE. Vibration on the other hand did not change and the drive is now vibrating enough to get the case vibrating as well. I will at some point get a SATA to eSATA adapter so I can have the HDD outside the case and so minimize all noise. Regarding temperatures, when the HDD was naked it was around 3~4C hotter than when in the Nexus Drive-A-Way. Inside the Nexus enclosure the HDD is idling at 30C. Great and sturdy enclosure, I say it is a GREAT product! Only con is that it does not really minimize vibration. I need to find a way to get the HDD mounted inside the enclosure so that it does not vibrate and all my wishes will come true.

P.S. I got the Titanium version, EXTREMELY sleek! Like I said I will make it external because it looks very sexy, much better than other HDD enclosures!

~El~Jefe~
Friend of SPCR
Posts: 2887
Joined: Mon Feb 28, 2005 4:21 pm
Location: New York City zzzz
Contact:

Post by ~El~Jefe~ » Mon Feb 16, 2009 9:23 pm

interesting review, neat

samuelmorris
Posts: 168
Joined: Thu Mar 22, 2007 1:00 pm
Location: York, UK

Post by samuelmorris » Wed Feb 18, 2009 11:11 am

A Spinpoint F1 in a Drive-a-Way is unstoppable for airborne and seek, that's from personal experience. The vibrations are basically transferred directly though, and F1s vibrate a lot. The only way I found to stop that was to use the drive-a-ways externally, sitting on a towel. Heat didn't seem an issue, but it's hardly portable.

Aris
Posts: 2299
Joined: Mon Dec 15, 2003 10:29 am
Location: Bellevue, Nebraska
Contact:

Post by Aris » Wed Feb 18, 2009 1:16 pm

Drive enclosures do almost nothing for vibrations. So while seeks may be muted a bit and you wont hear any "whoosh" sound, the vibration from hard mounting the drive will likely be just as loud as it was without the enclosure.

Suspension really is the best way to do it. Either that or get a drive with less vibration like a 2.5" or SSD.

You could also place it on some soft foam at the bottom of your case, and then Tie it down with some elastic cords. It would be a bit more secure than a dangling drive and still eliminate vibrations.

Wibla
Friend of SPCR
Posts: 779
Joined: Sun Jun 03, 2007 12:03 am
Location: Norway

Post by Wibla » Wed Feb 18, 2009 1:27 pm

HDparm is at best a lousy benchmarking tool, at worst its totally useless.

Retest with bonnie++ or IOmeter please.

Aris
Posts: 2299
Joined: Mon Dec 15, 2003 10:29 am
Location: Bellevue, Nebraska
Contact:

Post by Aris » Wed Feb 18, 2009 2:19 pm

Wibla wrote:...or IOmeter please.
http://faq.storagereview.com/tiki-index ... Benchmarks

"Weaknesses: Does not take locality effects into account and thus is totally useless for assessing single-user performance. "

http://www.storagereview.com/articles/2 ... page=0%2C3

"IOMeter unfortunately lacks one key setting: "Restrict xx% of access to within yyyyy sectors." Such a parameter would permit, say, 50% of accesses to occur within 8 megabytes of the last request... effectively simulating locality. As it stands, however, IOMeter is tilted towards accesses that span the entire drive without regards to locality- a pattern that much more closely simulates multi-user server environments rather than a single-user desktop/workstation."


A "Content Creation Winstone" suite would be best. Storagereview.com uses a modified winstone suite to test with.

LodeHacker
Posts: 628
Joined: Sat Dec 13, 2008 1:25 pm
Location: Finland

Post by LodeHacker » Wed Feb 18, 2009 3:12 pm

Funny story I sold the Samsung, bought the WD6400AAKS. The latter makes more noise so I sold it. Now I got the Sharkoon HDD Vibe Fixer which is similar to NoVibes and I will buy next week the exact same Samsung again.

Btw hdparm results were there just to show off numbers. If you calculate, 112 - 104 = 8, far from my mentioned 10MB. The 10MB comes from a file copy test I made. I monitored HDD access with gkrellm and copied a single 100GB file from an older Barracuda. At all times the Samsung was 10MB faster than the Barracuda mentioned in my review. That's true experience and measurements. Why you need still bonnie++?

At any case the Drive-A-Way is awesome but I sold it. Got a USB enclosure instead so I can use my HDD externally as well without hacking myself a SATA to eSATA adapter ( which would work with the Drive-A-Way for a complete sexy eSATA compatible enclosure ).

LodeHacker
Posts: 628
Joined: Sat Dec 13, 2008 1:25 pm
Location: Finland

Post by LodeHacker » Wed Feb 18, 2009 3:22 pm

By the way for the insterested the WD6400AAKS was the worst HDD I ever used in my life for audio production! Lousy 4.2ms in direct audio sampling and with AAM enabled it's 6ms :shock:

Wibla
Friend of SPCR
Posts: 779
Joined: Sun Jun 03, 2007 12:03 am
Location: Norway

Post by Wibla » Wed Feb 18, 2009 3:57 pm

Aris wrote:
Wibla wrote:...or IOmeter please.
http://faq.storagereview.com/tiki-index ... Benchmarks

"Weaknesses: Does not take locality effects into account and thus is totally useless for assessing single-user performance. "

http://www.storagereview.com/articles/2 ... page=0%2C3

"IOMeter unfortunately lacks one key setting: "Restrict xx% of access to within yyyyy sectors." Such a parameter would permit, say, 50% of accesses to occur within 8 megabytes of the last request... effectively simulating locality. As it stands, however, IOMeter is tilted towards accesses that span the entire drive without regards to locality- a pattern that much more closely simulates multi-user server environments rather than a single-user desktop/workstation."


A "Content Creation Winstone" suite would be best. Storagereview.com uses a modified winstone suite to test with.
IOmeter isnt perfect, as with most other tools.. but it gives you an idea. try using hdparm on a raid array, you will get totally false numbers...

limee
Posts: 126
Joined: Fri Mar 28, 2003 7:12 pm
Location: Ontario, Canada

Post by limee » Sat Feb 28, 2009 9:11 pm

Nice review and it mirrors my experience. I agree with the previous comment about suspension being needed for 3.5" drives. Having a front fan blowing on it works wonders for the temps, and it doesn't even have to be a lot of air, so any quiet fan bumped down until you can't hear it (still spinning of course) should do.

bimbobo
Posts: 5
Joined: Mon Mar 09, 2009 11:46 am
Location: Denmark

Post by bimbobo » Tue Mar 10, 2009 9:03 am

samuelmorris wrote:A Spinpoint F1 in a Drive-a-Way is unstoppable for airborne and seek, that's from personal experience. The vibrations are basically transferred directly though, and F1s vibrate a lot. The only way I found to stop that was to use the drive-a-ways externally, sitting on a towel. Heat didn't seem an issue, but it's hardly portable.
I made an home made case for the Samsung F1 320GB and now it's so silent that I have to check the "activity lamp" in my desktop to be sure it's working..
I will post the picture of the very rudemental but supereffective case in the week-end.
No other sound silencer did it better and I tried 3 of them.
Aris wrote: You could also place it on some soft foam at the bottom of your case, and then Tie it down with some elastic cords. It would be a bit more secure than a dangling drive and still eliminate vibrations.
That's almost what I did... it's excellent and costs ZERO
Regards

P.S. temp is between 28-38C

Post Reply