Approve the Cookies
This website uses cookies to improve your user experience. By using this site, you agree to our use of cookies and our Privacy Policy.
OK
Forums  •   • New posts  •   • RTAT  •   • 'Best of'  •   • Gallery  •   • Gear
Guest
Forums  •   • New posts  •   • RTAT  •   • 'Best of'  •   • Gallery  •   • Gear
Register to forums    Log in

 
FORUMS General Gear Talk Computers 
Thread started 15 Oct 2011 (Saturday) 19:45
Search threadPrev/next
sponsored links (only for non-logged)

Hard drive vs SSD performance comparison

 
tim
Light Bringer
Avatar
51,010 posts
Likes: 375
Joined Nov 2004
Location: Wellington, New Zealand
     
Oct 15, 2011 19:45 |  #1

Out of interest I put my drives through their paces. I found exactly what I expected, though it's interesting to see numbers.

My SSD drives are OWC 3G, not their latest version. My Seagate and Western Digital drives are 1-2 years old.

What I found:
- SSDs have better read throughput than hard drives, though hard drives are only 30-50% behind for reading large chunks of data
- Hard drives have better write throughput than SSDs
- All drives slow down with lots of random reads and writes, but SSDs are about 50
times faster than hard drives
- Larger SSDs are faster than smaller SSDs, presumably because they have more flash chips and can parallelise the data access over them.


HOSTED PHOTO
please log in to view hosted photos in full size.


Professional wedding photographer, solution architect and general technical guy with multiple Amazon Web Services certifications.
Read all my FAQs (wedding, printing, lighting, books, etc)

  
  LOG IN TO REPLY
zombieman
Junior Member
27 posts
Joined May 2005
     
Oct 16, 2011 05:27 |  #2

And here's a nice, detailed performance comparison specifically for Lightroom:
http://www.computer-darkroom.com …be-lightroom-performance/ (external link)




  
  LOG IN TO REPLY
tim
THREAD ­ STARTER
Light Bringer
Avatar
51,010 posts
Likes: 375
Joined Nov 2004
Location: Wellington, New Zealand
     
Oct 16, 2011 05:58 |  #3

I did a bridge version here.


Professional wedding photographer, solution architect and general technical guy with multiple Amazon Web Services certifications.
Read all my FAQs (wedding, printing, lighting, books, etc)

  
  LOG IN TO REPLY
hollis_f
Cream of the Crop
Avatar
10,649 posts
Gallery: 1 photo
Likes: 85
Joined Jul 2007
Location: Sussex, UK
     
Oct 16, 2011 07:31 |  #4

Yuck! Those benchmarks are pretty slow.

Here's my Intel SSD

IMAGE: http://www.frankhollis.com/temp/SSD%20G3.jpg

Frank Hollis - Retired mass spectroscopist
Give a man a fish and he'll eat for a day. Teach a man to fish and he'll complain about the withdrawal of his free fish entitlement.
Gear Website (external link)

  
  LOG IN TO REPLY
solara
Senior Member
620 posts
Joined Feb 2010
     
Oct 16, 2011 10:20 |  #5

Wow, you really hammered your SSD running 5x 1000mb. No need to use such big chunks for testing. The 5x 100mb used above is a better idea.

I'd let your drive idle overnight for a few days for garbage collection to kick in so you can regain some of your write speeds. Oh, and write speeds will depend heavily on whether a particular SSD has saturated all of it's NAND.

I'm assuming the OWC uses a Sandforce controller - they perform poorly with CrystalDiskMark since it uses uncompressible data.


5D III, 7D | 17-55 f/2.8 | 16-35 f/4 | 24-105 f/4 | 85 f/1.8 | 135 f/2 | 70-200 f/4 IS | 580EX II | YN-560 | Manfrotto 190XPROB+498RC2

  
  LOG IN TO REPLY
tim
THREAD ­ STARTER
Light Bringer
Avatar
51,010 posts
Likes: 375
Joined Nov 2004
Location: Wellington, New Zealand
     
Oct 16, 2011 16:06 |  #6

The Intel one does look a bunch faster. I just used the default test, which was 5GB. My SSDs are a generation old, of the Sandforce controller. How old is the Intel one, and how expensive was it?

I very rarely write to my SSDs, other than the page file sitting on one of them, but I have 16GB of RAM so I doubt it's used much.

Do the sandforce controllers compress data before they write them? That'd be a smart thing to do, just a bit more processing power could increase performance a fair bit for some data.


Professional wedding photographer, solution architect and general technical guy with multiple Amazon Web Services certifications.
Read all my FAQs (wedding, printing, lighting, books, etc)

  
  LOG IN TO REPLY
hihohito
Member
225 posts
Joined Oct 2010
     
Oct 17, 2011 03:14 |  #7

hollis_f wrote in post #13256795 (external link)
Yuck! Those benchmarks are pretty slow.

Here's my Intel SSD

QUOTED IMAGE

How fast will your intel ssd be with 1000mb files? It looks like you used 100mb files.




  
  LOG IN TO REPLY
tim
THREAD ­ STARTER
Light Bringer
Avatar
51,010 posts
Likes: 375
Joined Nov 2004
Location: Wellington, New Zealand
     
Oct 17, 2011 03:23 |  #8

hihohito wrote in post #13260432 (external link)
How fast will your intel ssd be with 1000mb files? It looks like you used 100mb files.

I just ran a test with 100MB set. Random reads of small blocks was 50% faster, but otherwise not much changed. I don't think that setting makes much difference for SSDs, it'd probably make more difference for hard drives, which have higher latency.


Professional wedding photographer, solution architect and general technical guy with multiple Amazon Web Services certifications.
Read all my FAQs (wedding, printing, lighting, books, etc)

  
  LOG IN TO REPLY
hollis_f
Cream of the Crop
Avatar
10,649 posts
Gallery: 1 photo
Likes: 85
Joined Jul 2007
Location: Sussex, UK
     
Oct 17, 2011 04:13 |  #9

hihohito wrote in post #13260432 (external link)
How fast will your intel ssd be with 1000mb files? It looks like you used 100mb files.

Those figures aren't file sizes - they're total amounts of data transferred. It makes very little difference to the numbers if one uses 100MB instead of 1000MB - but it does run the test 10 times faster.


Frank Hollis - Retired mass spectroscopist
Give a man a fish and he'll eat for a day. Teach a man to fish and he'll complain about the withdrawal of his free fish entitlement.
Gear Website (external link)

  
  LOG IN TO REPLY
solara
Senior Member
620 posts
Joined Feb 2010
     
Oct 17, 2011 11:12 |  #10

And using smaller test blocks will reduce wear on the SSD.

And yes, Sandforce drives compress the data before writing them to the NAND - that's how they're able to have such good write speeds on typical/average compressible data.


5D III, 7D | 17-55 f/2.8 | 16-35 f/4 | 24-105 f/4 | 85 f/1.8 | 135 f/2 | 70-200 f/4 IS | 580EX II | YN-560 | Manfrotto 190XPROB+498RC2

  
  LOG IN TO REPLY
sponsored links (only for non-logged)

4,111 views & 0 likes for this thread, 5 members have posted to it.
Hard drive vs SSD performance comparison
FORUMS General Gear Talk Computers 
AAA
x 1600
y 1600

Jump to forum...   •  Rules   •  Forums   •  New posts   •  RTAT   •  'Best of'   •  Gallery   •  Gear   •  Reviews   •  Member list   •  Polls   •  Image rules   •  Search   •  Password reset   •  Home

Not a member yet?
Register to forums
Registered members may log in to forums and access all the features: full search, image upload, follow forums, own gear list and ratings, likes, more forums, private messaging, thread follow, notifications, own gallery, all settings, view hosted photos, own reviews, see more and do more... and all is free. Don't be a stranger - register now and start posting!


COOKIES DISCLAIMER: This website uses cookies to improve your user experience. By using this site, you agree to our use of cookies and to our privacy policy.
Privacy policy and cookie usage info.


POWERED BY AMASS forum software 2.58forum software
version 2.58 /
code and design
by Pekka Saarinen ©
for photography-on-the.net

Latest registered member is semonsters
1534 guests, 131 members online
Simultaneous users record so far is 15,144, that happened on Nov 22, 2018

Photography-on-the.net Digital Photography Forums is the website for photographers and all who love great photos, camera and post processing techniques, gear talk, discussion and sharing. Professionals, hobbyists, newbies and those who don't even own a camera -- all are welcome regardless of skill, favourite brand, gear, gender or age. Registering and usage is free.