Installing OS on first partition?

Peter780

Junior Member
Jul 31, 2013
12
0
61
Is it true that it is better to install the operating system on the first partition of the hard disk because the access to it will be faster that way and the OS will perform better?
 
Dec 10, 2005
28,726
13,892
136
I had heard this as well at one point. I would imagine that with current high-capacity hdds, it doesn't really matter. If you want speed, pick up a small ssd for your OS and put your data on the hard drive.
 

Ketchup

Elite Member
Sep 1, 2002
14,559
248
106
Yes, for the best performance it is best to install your OS on the first partition on a mechanical drive. The sectors are closer together on the inside of the disk, allowing faster access to the OS files, your files, and the page file. This will also ensure that your OS is all on the same platter, if your drive has more than one.
 

Billb2

Diamond Member
Mar 25, 2005
3,035
70
86
Whoa there. All sectors are the same size. Data is read/written fastest on the outer edge of a platter since it's speed is faster (as in distance/time)
 

sm625

Diamond Member
May 6, 2011
8,172
137
106
It is correct to say the sectors are closer together, since you can fit more sectors per unit of length with respect to the radius.
 

blankslate

Diamond Member
Jun 16, 2008
8,797
572
126
Here are a couple of links that provide more information...

Turns out that the outer edge is the area of the HDD that is faster according to these links
http://recoverymonkey.org/2012/07/24/what-is-hard-disk-short-stroking/
http://searchsolidstatestorage.techtarget.com/definition/Short-Stroking

Here is a guide to partitioning your HDD to maximize it's performance.

It looks like that it's an option only if you really have much more time than money though.

http://www.pcworld.com/article/255224/how_to_partition_your_hard_drive_to_optimize_performance.html

Hope these help with your question.
 

Ketchup

Elite Member
Sep 1, 2002
14,559
248
106
Thanks for the info blankslate. Can't remember who I heard the myth about the inside.

Peter780, I hope this answers you question.
 

Gintaras

Golden Member
Dec 28, 2000
1,892
1
71
It looks like that it's an option only if you really have much more time than money though.

http://www.pcworld.com/article/255224/how_to_partition_your_hard_drive_to_optimize_performance.html

Hope these help with your question.
Overall score 1.63%

Starting applications 2.54%

Importing pictures 3.66%

Video editing 1.18%

Gaming 0.75%

Windows Media Center 0.25%

Windows Defender 4.06%

Adding music 0.00%

"much faster"...have you noticed? parts of milliseconds - A lifetime?
 

Gintaras

Golden Member
Dec 28, 2000
1,892
1
71
Whoa there. All sectors are the same size. Data is read/written fastest on the outer edge of a platter since it's speed is faster (as in distance/time)

All platter spins at the same speed - either 5400 0r 7200 or 10000 or 15000...there no parts of platter spinning faster or slower...

If a human lived 100,000 years instead of 100, he might be able to notice...

Scientists are like priests in religion - trying to convince you about some things...

When I play 33 1/3 rpm Deep Purple - Machine Head, on first song of either side Ritchie Blackimore isn't playing faster than on the last song of each side...

Well..according some posts - he might...but who ever noticed that...just like on HDDs - Hudreds of parts of MILLISECONDS....
 

NutBucket

Lifer
Aug 30, 2000
27,151
635
126
The difference is constant linear velocity vs constant angular velocity. HDs and LPs have constant angular velocity whereas a CD has constant linear velocity; the disc slows down as the outer portions of the disc are accessed.
 

boochi

Senior member
May 21, 2011
983
0
0
All platter spins at the same speed - either 5400 0r 7200 or 10000 or 15000...there no parts of platter spinning faster or slower...

Very true although, you could say the linear velocity would be greater in the same sense as a CD or DVD write speed increases as it approaches the outer rim.

edit: beaten to the punch by a parrot, argh....
 
Dec 10, 2005
28,726
13,892
136
Scientists are like priests in religion - trying to convince you about some things...

Except for the fact that scientists present evidence to back up their claims while priests do not. Otherwise, they are exactly the same /sarcasm.
 

blankslate

Diamond Member
Jun 16, 2008
8,797
572
126
Overall score 1.63%

Starting applications 2.54%

Importing pictures 3.66%

Video editing 1.18%

Gaming 0.75%

Windows Media Center 0.25%

Windows Defender 4.06%

Adding music 0.00%

"much faster"...have you noticed? parts of milliseconds - A lifetime?

Ok. but I said it's an "if you have much more time that money."

sure there's an improvement that you can get for free (aside from the time investment takes to run HD tach for benchmarks on the drive and then partition it accordingly), but if you are willing to spend some money you can get an even greater improvement.

Say an SSD or perhaps the Seagate hybrid drives which put the most accessed data on NAND memory in addition to the spinning platters.

Nothing you have pointed out contradicts my previous statement.

That's what the phrase much more time that money meant... if you don't have the money to upgrade the storage on the system then some time spent can net some gains.

Maybe I should just respond in short very simple sentences from now on....
 
Last edited:

Cerb

Elite Member
Aug 26, 2000
17,484
33
86
Solution: http://www.newegg.com/Product/Produc...82E16820171667

You won't need an Intel H/Z/etc. chipset, and it's pretty cheap. OT: it's also a nice little drive for *n*x appliances.

HDDs split the platter up into sections, to get similar density across the platter.
http://www.dewassoc.com/kbase/hard_drives/hard_drive_glossary_5.htm#ZDR

The outer edge is still spinning fastest, and transfers the fastest, but when you're waiting on the HDD, you're mostly waiting on the arm, which is why those application benches got so little added performance.
 

Ketchup

Elite Member
Sep 1, 2002
14,559
248
106
Solution: http://www.newegg.com/Product/Produc...82E16820171667

You won't need an Intel H/Z/etc. chipset, and it's pretty cheap. OT: it's also a nice little drive for *n*x appliances.

HDDs split the platter up into sections, to get similar density across the platter.
http://www.dewassoc.com/kbase/hard_drives/hard_drive_glossary_5.htm#ZDR

The outer edge is still spinning fastest, and transfers the fastest, but when you're waiting on the HDD, you're mostly waiting on the arm, which is why those application benches got so little added performance.

That ReadyCache is pretty cool. I was actually surprised by the amount of positive reviews. I just didn't really think that many people would like it so well.

I looked at the Sandisk page, but I still don't know exactly what it does. Do you know how those work Cerb? I thought maybe it was just taking the page file and Superfetch, but that's just a guess.
 

Cerb

Elite Member
Aug 26, 2000
17,484
33
86
That ReadyCache is pretty cool. I was actually surprised by the amount of positive reviews. I just didn't really think that many people would like it so well.

I looked at the Sandisk page, but I still don't know exactly what it does. Do you know how those work Cerb? I thought maybe it was just taking the page file and Superfetch, but that's just a guess.
The software is ExpressCache, a competitor to Intel's SRT (often seen on Lenovo, Asus, and Samsung notebooks), that isn't limited to certain Intel chipsets. I haven't used the software with that drive, myself, though. The SSD works fine as a plain OS drive, formatted EXT4 :).
 

ashetos

Senior member
Jul 23, 2013
254
14
76
I'm pretty sure that all platters are used for sequential disk access, in a way that you cannot place your partition only in one platter. It should be such that, the higher the LBA is, the lower the throughput is.

Please, correct me if I'm wrong.
 

thewhat

Member
May 9, 2010
186
6
76
Overall score 1.63%

Starting applications 2.54%

Importing pictures 3.66%

Video editing 1.18%

Gaming 0.75%

Windows Media Center 0.25%

Windows Defender 4.06%

Adding music 0.00%

"much faster"...have you noticed? parts of milliseconds - A lifetime?
A couple of points:
1. they used a very large first partition (200GB)
2. they most likely did the comparison with fresh installs, meaning most programs were installed toward the beginning of the 1TB partition anyway

With a more reasonably sized first partition and with more data on the drive (and more fragmentation) you could very well see a significantly bigger improvement.

To me, setting up a smaller partition for the OS and programs is a no-brainer. All it takes is 10 seconds when you're setting up a drive for the first time. There's no need to run any benchmarks and whatnot.
And it has other benefits besides speed improvement.
 
Last edited:

Ketchup

Elite Member
Sep 1, 2002
14,559
248
106
If you are looking at hard drives from a speed perspective, here is what I do, and it works pretty well. Note I am referring to only having one drive here.

1. If all I have is one copy of Windows and no virtual machines, everything goes on one partition.
2. Use Auslogics Disk Defrag once a month or so with the "Defrag and Optimize" option. It will push data as close to the "outside" as possible.
3. If I do run Virtual Machines, each one gets its own drive letter. The VM (and all my other data) gets much less fragmented that way.
 

blankslate

Diamond Member
Jun 16, 2008
8,797
572
126
A couple of points:

And it has other benefits besides speed improvement.

The main one for me is being able to reinstall the OS without having to move the data to another drive first. Unless I really managed to download some nasty malware. *knock on wood* that hasn't happened so far...
 

Billb2

Diamond Member
Mar 25, 2005
3,035
70
86
All platter spins at the same speed - either 5400 0r 7200 or 10000 or 15000...there no parts of platter spinning faster or slower...

You didn't understand my answer.

Yes, all part of the platter spin at the same speed, So, yes, the angular velocity is the same anywhere on the platter.

But the absolute speed at any particular radial distance from the center of rotation will be faster the larger the radius is.

So, under the head, the platter is moving past at differing speeds depending on the location of the head.

Don't ya think so?
 
Last edited: