Overclock.net › Forums › Components › Hard Drives & Storage › Myth? Larger capacity C: drive yields worse performance?
New Posts  All Forums:Forum Nav:

Myth? Larger capacity C: drive yields worse performance? - Page 2

post #11 of 24
Try to google some benchmarks. Also google "inside the hard drive" or similar.
Edited by larsch - 1/8/12 at 12:33pm
post #12 of 24
go right ahead, post back if you find something interesting
post #13 of 24
Oh well, suit yourself.

I can add a bit about your calculation. The sectors are not numbered contiguously, they are interleaved. That means you cannot calculate it like you think.
Edited by larsch - 1/8/12 at 1:09pm
post #14 of 24
If you don't believe this than find some more reviews. The higher the density the hard drives paters(usually the larger the drive) the faster the hard drive is.

http://www.anandtech.com/show/3981/western-digital-caviar-green-3tb-and-my-book-essential-3tb-drives-reviewed/5
Edited by rancor - 1/8/12 at 1:19pm
post #15 of 24
More capacity doesn't mean worse performance. At the moment I have a 500 GB Seagate Barracuda (7200 RPM) SATA II and a 120 GB Hitachi something or other (5400 RPM) SATA II and the Hitachi has worse performance than the Seagate.

This myth was probably started because SSDs (when the capacity options were 16 GB, 32 GB and 64 GB) were a lot faster than the 750 GB - 1 TB HDDs.
    
CPUMotherboardGraphicsRAM
AMD Ryzen R5 1600 Asus PRIME B350 PLUS  AMD Radeon HD7950 16GB Corsair Vengence (2x8GB) 
Hard DriveHard DriveHard DriveOS
1TB WD Blue 500GB WD Blue 120GB Hitachi Windows 10 Pro 
MonitorMonitorKeyboardPower
LG 32LD450 Dell Ducky DK9008 OCN Edition Corsair TX650v2 
Case
Fractal Design Core 3000 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
AMD Ryzen R5 1600 Asus PRIME B350 PLUS  AMD Radeon HD7950 16GB Corsair Vengence (2x8GB) 
Hard DriveHard DriveHard DriveOS
1TB WD Blue 500GB WD Blue 120GB Hitachi Windows 10 Pro 
MonitorMonitorKeyboardPower
LG 32LD450 Dell Ducky DK9008 OCN Edition Corsair TX650v2 
Case
Fractal Design Core 3000 
  hide details  
Reply
post #16 of 24
Quote:
Originally Posted by andrews2547 View Post

More capacity doesn't mean worse performance. At the moment I have a 500 GB Seagate Barracuda (7200 RPM) SATA II and a 120 GB Hitachi something or other (5400 RPM) SATA II and the Hitachi has worse performance than the Seagate.
This myth was probably started because SSDs (when the capacity options were 16 GB, 32 GB and 64 GB) were a lot faster than the 750 GB - 1 TB HDDs.

So what you're telling me right now is that your 7200 RPM drive is faster than your 5400 RPM drive.

More breaking news at 11.
The Fortress
(15 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 930 @ 4.0 GHz 1.28V Gigabyte X58A-UD3R Sapphire Radeon HD 5850 1GB 850MHz Core, 4800MH... Kingston 6GB 1528MHz 7-7-7-20-1T 
Hard DriveOptical DriveCoolingOS
WD Caviar Black 1TB LG WH10LS30 10X Blu-ray Burner CoolerMaster Hyper 212+ Windows 7 Ultimate 64-bit 
MonitorKeyboardPowerCase
Acer X233Hbd 23" 1080p Logitech Illuminated Keyboard BFG GS-650 Silverstone FT01 
MouseMouse PadAudio
Razer Naga Molten Edition Cloth Pad Philips 2.1 
  hide details  
Reply
The Fortress
(15 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 930 @ 4.0 GHz 1.28V Gigabyte X58A-UD3R Sapphire Radeon HD 5850 1GB 850MHz Core, 4800MH... Kingston 6GB 1528MHz 7-7-7-20-1T 
Hard DriveOptical DriveCoolingOS
WD Caviar Black 1TB LG WH10LS30 10X Blu-ray Burner CoolerMaster Hyper 212+ Windows 7 Ultimate 64-bit 
MonitorKeyboardPowerCase
Acer X233Hbd 23" 1080p Logitech Illuminated Keyboard BFG GS-650 Silverstone FT01 
MouseMouse PadAudio
Razer Naga Molten Edition Cloth Pad Philips 2.1 
  hide details  
Reply
post #17 of 24
As someone else said, higher-density platters are faster, all else the same (ie 3x 160gb platters vs. 1x 500gb platter, the 500gb platter HDD is faster, period). Also, the "start" of the disk(really the outside) is faster by a little bit. If you run HDTune on a clean hard drive (just with OS), you'll see the outside is ~10% faster than the inside.

OP, you really should partition off C:\ so it either holds just the OS, the OS + programs, or OS + programs + games. 2nd partition for whichever 2 of the 3 (if any) aren't on C:\, make that logical. 3rd partition for data. There's no reason you should have videos, music, etc. all on C: if you have enough space to partition the drive.
 
Lanbox Lite
(16 items)
 
 
CPUGraphicsRAMHard Drive
i3-2310M Intel HD 3000 8GB DDR3 Samsung 830 64GB SSD 
Hard DriveOSMonitorCase
640GB Hitachi HDD Windows 7 13.3" LCD Magnesium Alloy, 3.2lbs 
CPUMotherboardGraphicsRAM
Phenom II X4 955 BE Gigabyte GA-MA78GPM-UD2H MSI Hawk R5770 3x2GB G.Skill DDR2 800 4-4-4-12 
Hard DriveHard DriveOptical DriveCooling
1TB Samsung Spinpoint F3 250 GB WD Caviar Black Samsung 20X DVD-R/RW Thermaltake MaxOrb 
CoolingOSMonitorPower
Noctua NF-B9-1600 Windows 7 Pro 64-bit BenQ E2420HD, 24" 1920x1080 TT Purepower 500W 
Case
TT Lanbox Lite 
  hide details  
Reply
 
Lanbox Lite
(16 items)
 
 
CPUGraphicsRAMHard Drive
i3-2310M Intel HD 3000 8GB DDR3 Samsung 830 64GB SSD 
Hard DriveOSMonitorCase
640GB Hitachi HDD Windows 7 13.3" LCD Magnesium Alloy, 3.2lbs 
CPUMotherboardGraphicsRAM
Phenom II X4 955 BE Gigabyte GA-MA78GPM-UD2H MSI Hawk R5770 3x2GB G.Skill DDR2 800 4-4-4-12 
Hard DriveHard DriveOptical DriveCooling
1TB Samsung Spinpoint F3 250 GB WD Caviar Black Samsung 20X DVD-R/RW Thermaltake MaxOrb 
CoolingOSMonitorPower
Noctua NF-B9-1600 Windows 7 Pro 64-bit BenQ E2420HD, 24" 1920x1080 TT Purepower 500W 
Case
TT Lanbox Lite 
  hide details  
Reply
post #18 of 24
Quote:
Originally Posted by Detroitsoldier View Post

So what you're telling me right now is that your 7200 RPM drive is faster than your 5400 RPM drive.
More breaking news at 11.

Yes. I was just saying it's not the capacity that determines the speed of the drive it's the RPM that matters. If it capacity does determine the speed then the 5400 RPM would have been faster than my 7200 RPM.
    
CPUMotherboardGraphicsRAM
AMD Ryzen R5 1600 Asus PRIME B350 PLUS  AMD Radeon HD7950 16GB Corsair Vengence (2x8GB) 
Hard DriveHard DriveHard DriveOS
1TB WD Blue 500GB WD Blue 120GB Hitachi Windows 10 Pro 
MonitorMonitorKeyboardPower
LG 32LD450 Dell Ducky DK9008 OCN Edition Corsair TX650v2 
Case
Fractal Design Core 3000 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
AMD Ryzen R5 1600 Asus PRIME B350 PLUS  AMD Radeon HD7950 16GB Corsair Vengence (2x8GB) 
Hard DriveHard DriveHard DriveOS
1TB WD Blue 500GB WD Blue 120GB Hitachi Windows 10 Pro 
MonitorMonitorKeyboardPower
LG 32LD450 Dell Ducky DK9008 OCN Edition Corsair TX650v2 
Case
Fractal Design Core 3000 
  hide details  
Reply
post #19 of 24
Quote:
Originally Posted by hazarada View Post

i actually believed that physical position of data on a drive makes a difference when i was like 12, surely enough i did test it with a game database and guess what - no difference whatsoever. whatever changed your loading times wasn't related to that. if you employed just a little bit of logic you could figure out why
You probably tested it on a game that was CPU limited instead. Some games can be put on SSDs and yet load exactly the same speed. Or if it was a multiplayer game, a large part of it is downloading data from the server. In TF2 my biggest loadtime jump came from upgrading from 3mbit to 6mbit.
Quote:
Originally Posted by larsch View Post

Quote:
Originally Posted by hazarada View Post

i actually believed that physical position of data on a drive makes a difference when i was like 12, surely enough i did test it with a game database and guess what - no difference whatsoever.
But there is a difference. There are more sectors per track at the outer edge of the platter, so the hard drive can potentially read more data per rotation, if data is stored there.
Correct.
Quote:
Originally Posted by hazarada View Post

if that were true then there would be a 3x speed difference between the outside and inside of the platter, obviously that is not the case, the heads still read at the same speed. Also if that were the case 2tb 5400rpm drive would be 5 times faster than a 146gb 15000rpm drive because more sectors in a revolution - again this isn't true.
What you're failing to grasp is every read or write involves two things - the seek to the right sector, and also the reading or writing part. (determined by sequential performance) Three times the sequential performance does not equal three times the speed. Modern drives can read 1MB in about 8ms, but the seek may take another 7-20ms depending on how fast the drive seeks. (So 15-28ms for reading a 1MB texture is probably pretty common) Now, if you put that texture at the end of the drive, sequential performance could drop to around half, doubling the time to read that 1MB to 16ms. That gives you a range of ~21-36ms. (I chose ~21ms because if you examine the charts below, the Velociraptor's sequential performance does not deteriorate as fast - more on that later.)

Given two drives with equal seek times (lets say 13ms), one reading 40MB/sec and the other 120MB/sec, the slower drive will take ~38ms to read that 1MB texture, while the faster drive will take ~21.5ms. Three times the sequential performance only provides a 76% performance increase when dealing with large (1MB) reads/writes.

Many reads/writes are smaller, so the difference will be less - but with games larger reads are very common. It's highly likely that you chose a game that is limited by other factors. Many games compress their data and require beefy CPUs to unpack everything. And as mentioned above, multiplayer game loadtimes may be determined more by your connection speed.

10000RPM: http://img201.imageshack.us/img201/2614/wdraptorhdtune.jpg
7200RPM: http://img221.imageshack.us/img221/3350/hdtuneproahcibenchmarkvne.png
5400RPM: http://img109.imageshack.us/img109/5864/hdtuneproahcibenchmarkv.png

wdraptorhdtune.jpg
hdtuneproahcibenchmarkvne.png
hdtuneproahcibenchmarkv.png

Now, if you divide the minimum and maximum sequential speeds for each of the above graphs, you get a ratio - for the Velociraptor it's about 1.7, for the WD Black it's about 2.0, and for the WD Green it's about 2.4. The max speed equals the min speed multiplied by those numbers. Why is there such a difference between drive types?

It's nothing to do with the RPM. RPM only affects seek/access times. It's because of an unadvertised factor - the higher RPM drives are using less surface area from each platter... My WD Green is using 667GB platters, which is why performance falls so much as it reaches the the end. My WD Black 640GB is actually using less dense platters, but only part of them. Consider it a factory short-stroke, of sorts. By using only part of two ~400GB platters, they keep access times lower and sequential speeds don't drop as far. In the case of that Raptor, the short-stroke is even more extreme. I've heard that if you open up a Velociraptor, you'll see that they're using thinner platters, which resemble donuts or rings rather than a traditional platter. While most desktop drives might have platters that are about ~3cm from edge to inside (arbitrary numbers), a Raptor will probably be closer to ~1-1.5cm. (again, not exact/real measurements, but the physical platter sizes are very different)

Quote:
Originally Posted by andrews2547 View Post

More capacity doesn't mean worse performance. At the moment I have a 500 GB Seagate Barracuda (7200 RPM) SATA II and a 120 GB Hitachi something or other (5400 RPM) SATA II and the Hitachi has worse performance than the Seagate.
What we're saying is more capacity means better performance (other factors being the same) - but using less of the available capacity enhances performance even more. If you can restrict reads/writes to a certain area (through smart partitioning, defragmenting, etc.), you get better performance out of your drive.

It's just part of the nature of hard drives. Actually, SSDs are the same way - they need some reserve space for their garbage collection and wear levelling algorithms. For HDDs performance crumbles because seek times go up and sequential performance goes down, but the same rule applies to all drives - they perform worse when full. If you must a hard drive, sort your data so the head only has to move around within a small section/ring of the platter when doing a task. As mentioned in previous posts, the easiest way to do this is keep all your executable data (OS, games, apps, etc.) as close to the start of the drive as possible.
Quote:
Originally Posted by jrbroad77 View Post

As someone else said, higher-density platters are faster, all else the same (ie 3x 160gb platters vs. 1x 500gb platter, the 500gb platter HDD is faster, period). Also, the "start" of the disk(really the outside) is faster by a little bit. If you run HDTune on a clean hard drive (just with OS), you'll see the outside is ~10% faster than the inside.

OP, you really should partition off C:\ so it either holds just the OS, the OS + programs, or OS + programs + games. 2nd partition for whichever 2 of the 3 (if any) aren't on C:\, make that logical. 3rd partition for data. There's no reason you should have videos, music, etc. all on C: if you have enough space to partition the drive.
There are exceptions to the rule. WD has a habit of making their Green drives with very high density platters, yet they have average sequential speeds - the controllers selected seem to be capped at a certain speed... Hitachi on the other hand has extremely fast 5400RPM drives - their benchmarks show much more jaggy graphs, indicating no such cap. The difference is about 120MB/sec (WD) vs 160MB/sec. (Hitachi)
Quote:
Originally Posted by andrews2547 View Post

Yes. I was just saying it's not the capacity that determines the speed of the drive it's the RPM that matters. If it capacity does determine the speed then the 5400 RPM would have been faster than my 7200 RPM.
Again, smaller does not make it faster. Using a lower percentage of the available capacity does. It has real effects on the distance the head seeks, where files are positioned on the platters, etc.
Edited by Kramy - 1/8/12 at 3:35pm
     
CPUMotherboardGraphicsRAM
FX-8350 @ 4.6ghz 1.425v Sabertooth 990FX Zotac GTX 1070 Amp! 32GB (4x8GB) Kingston Fury 1866mhz CL10 1T 1.5v 
Hard DriveHard DriveHard DriveHard Drive
Hitachi 2TB - HDS723020BLA64 Crucial M500 960GB - CT960M500SSD1 WD 4TB Black - WD4001FAEX WD 4TB Black - WD4001FAEX 
Hard DriveHard DriveHard DriveHard Drive
Seagate 2TB - ST2000DM001 WD 640GB Black - WD6401AALS Seagate 6TB - ST6000DM001 Micron M500 - MTFDDAK480MAV 
Optical DriveOptical DriveOptical DriveOptical Drive
Pioneer DVR-115DBK Samsung SH-S243D/BEBE LG GH22NS90 Lite-On EBAU108 External DVD 
CoolingCoolingCoolingOS
Noctua NH-D15 Fractal Design 140mm Fan - FD-FAN-140 Fractal Design 120mm Fan - FD-FAN-120 Windows 7 
OSMonitorMonitorKeyboard
Ubuntu 14.04 LTS Samsung 2343BWX 23" (2048x1152) BenQ GW2765HT 27" IPS LCD Monitor (2560x1440) Genius Keyboard KB-G235 PA 
PowerCaseMouseMouse Pad
Seasonic Platinum 1000w PSU - SS-1000XP Fractal Design Define R5 Steelseries Rival 6500 dcpi Mousepad + Desk 
AudioAudioOtherOther
Onboard Realtek® ALC 892 Cyber Acoustics CA-2992 2.0 Speakers Razer Krait 1600 DPI nGear G-C1601 Card Reader 
OtherOther
Pisen Card Reader Gigabyte M7 Thor 
CPUMotherboardGraphicsRAM
VIA Eden C7 1200mhz Jetway J7F4K1G2ES-LF VIA Onboard 512MB PC2-4200 
Hard DriveOptical DriveCoolingOS
A-Data FlashDrive None Fanless Linux 
MonitorKeyboardPowerCase
Headless None 65w Power Brick Antec ISK300-65 
MouseAudio
None Onboard VIA 
CPUMotherboardGraphicsRAM
Barton XP-M 2400+ FJQ4C Asus A7N8X-E Deluxe 256MB BFG 7800GS OC 2x512MB Samsung PC3200 
Hard DriveOSMonitorKeyboard
Raptor 74GB/Seagate 120GB Windows 2000 Pro 60hz 17" CRT Brandless PS2 Keyboard 
PowerMouse
Antec Neo HE 550w Logitech G5 
  hide details  
Reply
     
CPUMotherboardGraphicsRAM
FX-8350 @ 4.6ghz 1.425v Sabertooth 990FX Zotac GTX 1070 Amp! 32GB (4x8GB) Kingston Fury 1866mhz CL10 1T 1.5v 
Hard DriveHard DriveHard DriveHard Drive
Hitachi 2TB - HDS723020BLA64 Crucial M500 960GB - CT960M500SSD1 WD 4TB Black - WD4001FAEX WD 4TB Black - WD4001FAEX 
Hard DriveHard DriveHard DriveHard Drive
Seagate 2TB - ST2000DM001 WD 640GB Black - WD6401AALS Seagate 6TB - ST6000DM001 Micron M500 - MTFDDAK480MAV 
Optical DriveOptical DriveOptical DriveOptical Drive
Pioneer DVR-115DBK Samsung SH-S243D/BEBE LG GH22NS90 Lite-On EBAU108 External DVD 
CoolingCoolingCoolingOS
Noctua NH-D15 Fractal Design 140mm Fan - FD-FAN-140 Fractal Design 120mm Fan - FD-FAN-120 Windows 7 
OSMonitorMonitorKeyboard
Ubuntu 14.04 LTS Samsung 2343BWX 23" (2048x1152) BenQ GW2765HT 27" IPS LCD Monitor (2560x1440) Genius Keyboard KB-G235 PA 
PowerCaseMouseMouse Pad
Seasonic Platinum 1000w PSU - SS-1000XP Fractal Design Define R5 Steelseries Rival 6500 dcpi Mousepad + Desk 
AudioAudioOtherOther
Onboard Realtek® ALC 892 Cyber Acoustics CA-2992 2.0 Speakers Razer Krait 1600 DPI nGear G-C1601 Card Reader 
OtherOther
Pisen Card Reader Gigabyte M7 Thor 
CPUMotherboardGraphicsRAM
VIA Eden C7 1200mhz Jetway J7F4K1G2ES-LF VIA Onboard 512MB PC2-4200 
Hard DriveOptical DriveCoolingOS
A-Data FlashDrive None Fanless Linux 
MonitorKeyboardPowerCase
Headless None 65w Power Brick Antec ISK300-65 
MouseAudio
None Onboard VIA 
CPUMotherboardGraphicsRAM
Barton XP-M 2400+ FJQ4C Asus A7N8X-E Deluxe 256MB BFG 7800GS OC 2x512MB Samsung PC3200 
Hard DriveOSMonitorKeyboard
Raptor 74GB/Seagate 120GB Windows 2000 Pro 60hz 17" CRT Brandless PS2 Keyboard 
PowerMouse
Antec Neo HE 550w Logitech G5 
  hide details  
Reply
post #20 of 24
nvm lol.

research research research...
Edited by RedShades - 1/8/12 at 4:36pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hard Drives & Storage
Overclock.net › Forums › Components › Hard Drives & Storage › Myth? Larger capacity C: drive yields worse performance?