Overclock.net › Forums › Components › Hard Drives & Storage › SSD › AS SSD benchmark
New Posts  All Forums:Forum Nav:

AS SSD benchmark

post #1 of 25
Thread Starter 
Ok I found a cool trick out to boost SSD performance just thought I share my findings.



All I did was compress the drive its a 4x 64 Gb Raid 0 with crucial c300s.

Read speed is insane you lose alot of write due to nature of the compressing but who the hell writes to drive that much for people like me who mostly use main drive to read IE play games or surf web and such stuff its perfect.

Basicly it hink it works as since the files are compressed there is less to read and since reading a compressed file takes little effort on the CPU its faster then if the file was bigger.
post #2 of 25
4x C300's?
Four?
As in one less than 5?
    
CPUMotherboardGraphicsRAM
Xeon w3680 Gigabyte X58A-UD3R GTX 295 Engineering Samples 2X Corsair XMS3 
Hard DriveHard DriveCoolingOS
Sandisk SSD Seagate Corsair H50 W7 Ult 
MonitorMonitorKeyboardCase
Viewsonic VA1912wb Viewsonic VA912b K350 CM HAF 
Mouse
Logitech M570 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Xeon w3680 Gigabyte X58A-UD3R GTX 295 Engineering Samples 2X Corsair XMS3 
Hard DriveHard DriveCoolingOS
Sandisk SSD Seagate Corsair H50 W7 Ult 
MonitorMonitorKeyboardCase
Viewsonic VA1912wb Viewsonic VA912b K350 CM HAF 
Mouse
Logitech M570 
  hide details  
Reply
post #3 of 25
Thread Starter 
Yes 4 ? I know its not 3 and defo not 5 why is that bad or something ?
post #4 of 25
Quote:
Originally Posted by SSD Twinkle Toes View Post
Yes 4 ? I know its not 3 and defo not 5 why is that bad or something ?
No. That's unusually good. I was like 600MB~ R/W what the eff how is that so high-FOUR SSDS?!?!?!
    
CPUMotherboardGraphicsRAM
Xeon w3680 Gigabyte X58A-UD3R GTX 295 Engineering Samples 2X Corsair XMS3 
Hard DriveHard DriveCoolingOS
Sandisk SSD Seagate Corsair H50 W7 Ult 
MonitorMonitorKeyboardCase
Viewsonic VA1912wb Viewsonic VA912b K350 CM HAF 
Mouse
Logitech M570 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Xeon w3680 Gigabyte X58A-UD3R GTX 295 Engineering Samples 2X Corsair XMS3 
Hard DriveHard DriveCoolingOS
Sandisk SSD Seagate Corsair H50 W7 Ult 
MonitorMonitorKeyboardCase
Viewsonic VA1912wb Viewsonic VA912b K350 CM HAF 
Mouse
Logitech M570 
  hide details  
Reply
post #5 of 25
Thread Starter 
Yeah I was surprised as well I felt it was to good info to not share tis why I joined the forums and posted it up.

I am new to this SSD stuff and was just messing around with settings.
I am not new to OCing mind as I have i7950 setup water cooled got my 950 to 4500 so far pretty stable.

like I said tho this is no good if you use the SSDs to write alot IE saving big files but then agian why would you on 64 Gb C300s which was made for reading.
post #6 of 25
in the long run this will actually slow down your usage. look to your CPU usage during these tests, or when you are loading things. It has to now uncompress all of that info on the fly.
Its a neat trick, but not good for 24/7 usage as it will eat cpu cycles for breakfast
post #7 of 25
Thread Starter 
Quote:
Originally Posted by Computurd View Post
in the long run this will actually slow down your usage. look to your CPU usage during these tests, or when you are loading things. It has to now uncompress all of that info on the fly.
Its a neat trick, but not good for 24/7 usage as it will eat cpu cycles for breakfast
I did think of that so i looked at my CPU usage over 2 tests and it don't even use up 14% of 1 core on my i7950 writing and reading.

I had same idea but it really isn't very stressful on the new i7s.

O and btw this isn't a trick it really improves the speed of SSDs reading data.

And it 100% depends on processor speed as i done a test on my i7 over clocked and on defult clocks and its a big gap with my i7 on defult speed its about 400 mb/s and like above was on 4.5 GHz OC 600 mb/s read.

So why run a really powerful CPU if you not going to use its power to boost ya SSD performance.I tried it in gaming as well and I get no CPU lagg even in CPU intensive games like Left 4 dead.

O and I havent even mentioned fact it gives you about 35-40% extra space on drive ?
Edited by SSD Twinkle Toes - 3/8/11 at 8:33pm
post #8 of 25
Magic! Obviously.
I found compression helps insanely with the read's and enables max bandwidth(whatever that may be de fault or other ...) for the writes. With newer hardware(cpu's) shouldn't negatively effect anything.

I'm going to get a 2nd new hd and try my current ssd in my older system to see how well this scales across sata types asap.

As a note, cpu useage before compression and some minor mods to some things was reported as -1%(obviously something is a miss) and with compression during read/write is reported as 1%.
Edited by kenolak - 3/8/11 at 8:36pm
Synthkart
(13 items)
 
  
CPUMotherboardGraphicsRAM
1055T Asus Onboard x4200 2x2GB GeIL DDR 1333 (9-9-9) 
Hard DriveOptical DriveOSMonitor
Seagate 320GB Sata3.0 Liteon iHAS124 Win7 x64 Some old crt 
KeyboardPowerCaseMouse
digital Raidmax-450K Some cheap raidmax Labtec 
Mouse Pad
  hide details  
Reply
Synthkart
(13 items)
 
  
CPUMotherboardGraphicsRAM
1055T Asus Onboard x4200 2x2GB GeIL DDR 1333 (9-9-9) 
Hard DriveOptical DriveOSMonitor
Seagate 320GB Sata3.0 Liteon iHAS124 Win7 x64 Some old crt 
KeyboardPowerCaseMouse
digital Raidmax-450K Some cheap raidmax Labtec 
Mouse Pad
  hide details  
Reply
post #9 of 25
Thread Starter 
Quote:
Originally Posted by kenolak View Post
Magic! Obviously.
I found compression helps insanely with the read's and enables max bandwidth(whatever that may be de fault or other ...) for the writes. With newer hardware(cpu's) shouldn't negatively effect anything.

I'm going to get a 2nd new hd and try my current ssd in my older system to see how well this scales across sata types asap.

As a note, cpu useage before compression and some minor mods to some things was reported as -1%(obviously something is a miss) and with compression during read/write is reported as 1%.
I agree be nice to see how those tests come out with hard drive speed/CPU speed/Ram.
I am guessing slower CPU more its slows down the SSD from reading and writing in compressed state.

I am sure tho my write times wouldn't be so low if it was on a better SSD for writing as C300s are not designed for writing be nice to see some results from someone with a 300+ write and read drive.
post #10 of 25
it would be nice to see a latency measurement before and after the compression. If in fact you are adding latency due to the computing overhead, then in reality you havent gained anything. in real world scenario large sequentials arent going to do much for you.
Also, would be interesting to see how this effects degradation of the array over a period of time.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: SSD
Overclock.net › Forums › Components › Hard Drives & Storage › SSD › AS SSD benchmark