Performance Bet

[tratrie]

Riddle me this...

I have a bet going with a friend of mine over performance. Here is the scenario:

Let's say you have 2 drives, A & B.
A & B use the same interface, the same number of heads, the same everything...EXCEPT Drive A is 200GB in size and Drive B is 50GB in size.

Now would one give you better performance than the other?

I'm saying Drive A would perform better due to the increased areal density.

He says Drive B would perform better because Drive A wouldn't have enough heads.

So what's the answer, or are we both wrong?
Remember, EVERYTHING is the same EXCEPT size.

Tanks!

 

[camieabz]

Probably about even.

The larger capacity hard drive <b>would</b> however be hotter, noisier and therefore in theory last less time than the smaller drive. If that's got anything to do with the question. :lol:

I'd go for the one you can afford everytime.

<b><font color=blue>~ <A HREF="http://www.btvillarin.com/phpBB/viewtopic.php?t=324" target="_new">My System Specs</A> ~<font color=blue></b>

 

[basmic]

Why not run HDTach 2.61 <A HREF="http://www.tcdlabs.com/hdtach.htm" target="_new">http://www.tcdlabs.com/hdtach.htm</A>?

Run both drives in both system setups - I'd be interested in the results you get back, if you feel like publishing them.

What brand/model are the hard-drives? What are the system setups (CPU, motherboard, interface (ATA-33, 66, 100.....), etc)?

====
Basmic

 

[Ncogneto]

Let's say you have 2 drives, A & B.
A & B use the same interface, the same number of heads, the same everything...EXCEPT Drive A is 200GB in size and Drive B is 50GB in size.

Now would one give you better performance than the other?

I'm saying Drive A would perform better due to the increased areal density.

Your question cannot be answered as such because it contradicts itself. Everything cannot be the same otherwise they would have the same amount of storage. Both you and your freinds are taking positions that make the original statement false to begin with.

In your case you are claiming an increase in areal density which would be a difference.

In your friends case he is claiming that one drive has more platters( I am guessing do to the fact of his head statement)than the other again making the original statement false.

So to settel the arguement you need to further clarify the two drives and there composition.

It's not what they tell you, its what they don't tell you!<P ID="edit"><FONT SIZE=-1><EM>Edited by ncogneto on 10/12/02 08:39 PM.</EM></FONT></P>

 

[HammerBot]

He did write that everything, except size is the same. That means number of platters heads etc. is identical but the data density on the platters is higher.
Given these circumstances I would expect that the drive with the higher data density is the fastest, since the 'data-frequency' seen by the heads is higher.

 

[Ncogneto]

He did write that everything, except size is the same

That contradicts with this:

That means number of platters heads etc. is identical but the data density on the platters is higher.

Perhaps better stated it should read everything but data density is the same. Furthermore to jump from a 50 gig drive to a 200 gig drive while keeping the numbers of platters and heads the same would be nearly impossible, making the 200 gig drives data density 4x that of the 50 gig drive or making it use huge platters ( something akin to the old quatum bigfoot series).

It's not what they tell you, its what they don't tell you!

 

[tratrie]

I guess I should have mentioned that this is theoretical, not reality. It was simply a hypothetical question.

But if I had to clarify then-

Both drives are the same physical size
Both drives have the same number of platters
Both drives have the same number of heads
Both drives spin at the same speed
Both drives have the same amount of cache
Both drives have the same seek times
Both drives have the same access times
Both drives use the same interface type
(Did I leave anything out?)

The difference is Drive A is 200GB, Drive B is 50GB.

So I guess, the question would be better stated by asking - Does a significant increase in areal density provide increased performance?

Sorry for any confusion.

 

[Ncogneto]

So I guess, the question would be better stated by asking - Does a significant increase in areal density provide increased performance?

Most certainly, yes.

It's not what they tell you, its what they don't tell you!

 

[lhgpoobaa]

Does an increase in data density help?
Oh yes!

Think about it for a moment.
You have a drive head, and beneath it spins the drive, so the head "see's" a certain area or distance around the platter every second. As both drives spin at the same speed they both see the same physical area or distance.
Double the platter density and theoretically the drive head will see twice the amount of data travel underneath.

Allthough there are lots of other factors in play that ensures that the relationship is not 1:1. But there IS a difference.

Thats why i would reccomend a 120Gb drive using 2 60Gb platters than one using 3 40Gb platter, as well as 1 less platter giving less heat and noise.

<b>Plaid will NEVER go out of style! </b>

 

[BoBBoB]

Agreeded

 

[Ncogneto]

Thats why i would reccomend a 120Gb drive using 2 60Gb platters than one using 3 40Gb platter, as well as 1 less platter giving less heat and noise.

Actually this is not quite correct. Adding platters is one way to offset a smaller areal density. While increasing areal density increases performance (STR) when comparing disks with an identical platter count, it becomes a bit more complicated when comparing disks with different platter counts.

It's not what they tell you, its what they don't tell you!

 

[Napoleon]

Furthermore, there are two factors at play in the areal density, namely TPI (Tracks Per Inch, in the direction of platter radius) and BPI (Bits Per Inch, in the track circumference). Areal density is typically expressed as TPI * BPI, yielding the "bits per square inch" areal density. The BPI and zoning determine the amount of data per rotation.

With the platter count thing, I assume that having a huge stack of platters would allow a longer transfer before there's any need to seek to the adjacent cylinder and possibly miss a full rotation in the process?


<font color=red><b><i>You want WHAT on the [-peep-] CEILING?!</i></b></font color=red> -Michelangelo