Hitachi 2TB 5K3000 - Testing & Info

aphasia

[H]ard|Gawd
Joined
Jan 15, 2011
Messages
1,137
i'm planning on purchasing 3 (to start) hitachi 2TB 5K3000 (HDS5C3020ALA632 / 5xxx rpm) HD's to install into my NAS soon but i have some questions.

-do these drive use 512b or 4K sectors?
-i want to test these drives in my PC first before plugging into my NAS. what software & tests should i do?
 
Last edited:
yeah on the website it says;
Sector size (variable, Bytes/sector): 512
though not sure what it means by 'variable'

anyway i've purchased them today at local distributor for $101 AUD each.
guess i'll find out the hard way if there is a problem / performance hit when i plug into my NAS (thecus n5200br pro).
probably just install 1 drive as JBOD and do a quick read/write over the network. should tell me if there is a issue.

any thoughts on software/tests to perform on drives?
 
yeah i also just got a couple on a deal at newegg (actually found this thread searching for the model number, lol) guess i'll run some tests when i get em, too. (tomorrow)
 

yeah, you saw that too huh. was waiting for these drives to hit australia.

i'm going to run the usual battery of benches (HD Tune/HD Tach, ATTO, crystal etc). what i was after was a hard disk stress test. something to do a surface scan, check for bad sectors etc.

don't want to stick these drives in RAID only to pull one or more b/c of a defective drive.
 
Yea, I saw them last week been meaning to get my hands on one.

Would love to see your feed back. Pictures if you can too.

*waits :)
 
i'll be running a bunch of benches and tests tonight.

also been reading up on hitachi's DFT V4.16(Drive Fitness Test). i'll probably run that on each of the drives as well. only issue there is i don't have a CD/DVD drive to run the util in a DOS environment. will try to run a bootable image off a USB flash drive. if there are any suggestions on other programs to use to test/stress drive, please chime in.

only have 1 picture atm. will post more later.
hitachi%205K3000%202TB%20(in%20packaging).jpg
 
more photos of physical drives;

just your 'run-of-the-mill' hard drive tops + labels
hitachi%204%20(front)crop.JPG


bit of weirdness here, notice the different coloured PCB on drive to the left...
hitachi%205%20(back)crop.JPG
 
ok on to basic info on the hitachi drives in windows;

yep, all disks present and accounted for.
disk%20drives%20(explorer).PNG



hmmm... slight discrepancy on formatted size on disk '0' (D:\) compared to disk '2' (E:\) & '3' (G:\)...
hitachi%20-Disk%20Management.PNG



wtf, i'm missing 8228864 bytes or 8 MB... it's like disk '0' (D:\) doubled up on the 'unallocated' space at the beginning of the drive.
hitachi%20-Local%20Disk%20Properties.PNG



partition wizard shows the 'unallocated space' is the same across the 3 hitachi drives.
hitachi%20-Partition%20Wizard.PNG


now i'm confused.
i've finished the benches, bit of variance there as well. not sure if the results fall outside the margin of error though. will post shortly.
 
ATTO benches...
-testing on gigabyte EP41-UD3L (ICH7/R), pentium E5400 2.7GHz, 2GB RAM, Win7.


not a lot in them btw 512bytes to 16KB. though there is a bit of a difference at peak after 32KB. drive '0' (D:\) with its slightly shorter capacity, hits 140MB/s for both read & write. the other 2 driver peak at 130'ish. difference of 5MB to 10MB.

despite the issue of missing capacity on first drive, results not bad for a cheap, slow spinning drive.

ATTO%20hitachi%20drive%20-D.PNG
ATTO%20hitachi%20drive%20-E.PNG
ATTO%20hitachi%20drive%20-G.PNG
 
That's pretty good... Faster than most greens on the market! WD, Samsung etc....

Did you notice the drive getting hotter and noisy?
 
CrystalDiskMark benches...

Crystal shows the same difference as ATTO. the shorter drive 'D' is a little quicker than 'E' & 'G'.
CrystalDiskMark%20hitachi%20drive%20-D.PNG
CrystalDiskMark%20hitachi%20drive%20-E.PNG
CrystalDiskMark%20hitachi%20drive%20-G.PNG


That's pretty good... Faster than most greens on the market! WD, Samsung etc....

Did you notice the drive getting hotter and noisy?

the next lot of pics should answer that question;

HDTune%20hitachi%20drive%20-D.png
HDTune%20hitachi%20drive%20-E.png
HDTune%20hitachi%20drive%20-G.png


wasn't paying much attention to proper cooling for this test. the 3 hitachi disks are all jammed into the same drive rack next to a WD 150G Raptor with little to no space between each. they are sitting at the bottom of my old lian li PC70 case (closed).

noise wise i couldn't hear a thing. to be honest if they were making much sound, it was drowned out by an old Zalman CPU HS fan. i did a full sector scan via HD tune on the hitchi disk that was slightly smaller (no errors) & i couldn't hear a thing. again CPU fan was the loudest thing coming from the case.
 
was 'drive 0' the one with the diff colour pcb? i have a batch of 10 sitting in front of me and one of the pcbs is slightly lighter than the others, not quite as different as yours, though.
 
The access time test is limited to 1TB maximum. As a result, the 15.7 ms access time is deceptively low for this particular drive. Most of these drives have 19 ms access times when properly tested. I came to the conclusion that the freeware version of HDTune does not properly support hard drives larger than 1TB.
 
E4g1e is correct. HDTune 2.55 short-strokes to 1TB when doing an access test.

Here are some comparisons of real access times when testing the full 2TB.

5000-6000 RPM
2TB WD20EARS = ~17ms
2TB HD204UI = ~18ms
2TB 5K3000 = ~19ms

7200 RPM
2TB WD2001 FASS/FAEX = ~12ms
2TB 7K2000 = ~13ms
 
Pretty interesting. I have four drives that I have been running for the past few weeks and have seen consistent 140MB/s across all of them.

Hitachi 5K3000 2TB review.

i'd blame the delta on his "ICH7/R" controller. benches on these drives are very consistent with a decent, newer controller. motherboard SATA is hardly an objective test unless we're talking the newer Intel offerings.
 
was 'drive 0' the one with the diff colour pcb? i have a batch of 10 sitting in front of me and one of the pcbs is slightly lighter than the others, not quite as different as yours, though.

no it was drive 3 (G:\), so the slowest disk in the test.
also all disks have same firmware

agree with odditory about regarding testing on ICH7/R, thats why i listed it in the first place. could of tested on my current rig with a ICH9R, but not enough room / couldn't be arsed.
 
E4g1e is correct. HDTune 2.55 short-strokes to 1TB when doing an access test.

Here are some comparisons of real access times when testing the full 2TB.

5000-6000 RPM
2TB WD20EARS = ~17ms
2TB HD204UI = ~18ms
2TB 5K3000 = ~19ms

7200 RPM
2TB WD2001 FASS/FAEX = ~12ms
2TB 7K2000 = ~13ms

Will probably adjust that one. Frankly though, when using any 3.5" SATA drive you are really doing sequential transfers at this point. Random I/O you mostly just try caching. Access times are a bit less exciting at this point for this reason, unless you are using them as boot drives (where actually 50% slower means something).

I tend to like having benchmarks anyone can run for free... but it is very valid that the short stroking gives an incomplete picture.
 
Thanks for the tests. I just ordered 4 of these for a new WHS build that I am working on. I was able to get one of them for $49.99 as part of a combo deal on newegg ;)
 
i'm jealous, i could've bought 6 drives with the same money and the favourable exchange rate (AUD to US). even if i could purchase stuff from newegg etc, probably wouldn't due to the length of time to return a bad drive and i'd be worried the drive wouldn't be packed correctly for such a long trip.
 
My four drives arrived from the Egg on Thursday. Three of them seem flawless so far, but one is making quite a bit of noise. All three PCBs were identical, and they were packaged very nicely.

I may RMA the noisy drive since its kind of annoying, but the other three seem to be working out nicely.
 
My four drives arrived from the Egg on Thursday. Three of them seem flawless so far, but one is making quite a bit of noise. All three PCBs were identical, and they were packaged very nicely.
Did they have a couple of deep scratches on their tops? I don't think it was due to handling by Newegg because the 2TB was bulk packed but the 1TB was retail boxed.
 
just a follow-up on the size discrepancy.

i email hitachi support, they replied 4 or 5 days later.
by then i'd already build my array in my NAS and just ignored the small size discrepancy. i haven't had an issue with RAID since the array build back in late feb.

anyway he's the copy of the email (they advised i could used the Feature Tool to reset drive back to manufacturing defaults which i didn't bother to do)

==========================
Dear Sir/Madam,

Thank you for contacting Hitachi Asia Pacific Technical Support Centre.

I am sorry for the late reply.

According to our record, the 3 drives are built around the same time and have the same firmware code. May I know how do you find that the drive F30E6J7D is missing 8228864 bytes? Which screen were you looking at?

This drive should have 3,907,029,168 sectors or 2,000,398,934,016 bytes.

We do have a free utility called Feature Tool that you can use to view the capacity of the drive, and also use it to change the capacity back to factory default. Do you want to try that? Feature Tool can be downloaded from this URL:
http://www.hitachigst.com/support/downloads/#FeatureTool

Thank you.

Regards, TechSup
#############################################
We support Hard Disk Drives sold through Distribution channel only.
Hitachi Global Storage Technologies
日立环球存储科技
Tel : (+65) 6840 9595 Fax : (+65) 6746 2076
Email: [email protected]
Tollfree : 1800 001 685 (Australia), 10800 650 0226 (North China), 10800 265 0226 (South China), 800 933 683 (Hong Kong), 00080 0650 1244 (India), 0018 0365 7854 (Indonesia), 1800 80 6016 (Malaysia), 1027 1800 6507 157(Philippines), 0800 44 6449 (New Zealand), 1800 8409595 (Singapore), 0080 165 1647 (Taiwan), 0018 0065 6450 (Thailand)
Website: http://www.HitachiGST.com
==========================
 
Hmm, interesting! I'm thinking of buying 4 of these for a NAS but I don't like to buy from China when I have the opportunity not to. That said, I imagine WDs are made in China too, eh?
 
Back
Top