Edit: 80GB drive vs. 200GB

[imported_Mitchell]

I recommend atleast 2 drives in a system. moving between drives is always faster then moving on the same drive.

just last night I DL'd World of Warcraft beta (2.6 GIGS) in order to unzip it at a reasonable speed I unzipped it from one drive to another, much faster.

It also helps having a nice fast drive for bootup and then anotehr drive to run programs from (games) as it's faster to a degree.

Your choice but I would do 1 80 gig SATA Python as a boot drive and a 250 SATA Maxtor or WD as the data drive etc.

Ultimately it is your choice though.

 

[VirtualLarry]

Well, interestingly, I DO just happen to have an 80GB WD JB drive as my primary OS/boot drive, and a 160GB WD JB as a secondary for data, and recently-added third 250GB Maxtor DM+9 added for storage.

The 80GB is an -ETA0 drive, so it is 80GB/platter, and has decent transfer rates. However, seek times have become worse, what used to be 13-14ms are now 15-16ms, as measured by HD tach. I don't know if that's simply because it is the boot/OS drive that contains the pagefile/registry or what. The 160GB is a -DUA0, which is I believe, 60GB/platter. Therefore, obviously lower STR, but seek times are decent. The Maxtor is the most recent drive, and probably has ~83GB/platter, as it is of fairly recent mfg. STR is very decent, seeks a little slower than the 160GB, and much noisier. Overall, all of the drives are quiet at idle. The WD's are BB motors, the Maxtor is a FDB.

I would definately say that having seperate boot and data drives are worth it. I can rip DVDs to one drive, download at full speed from another, and still have enough spare disk bandwidth on the boot drive for the pagefile, apps, tempfiles, etc. Considering that my WDs are over a year and about a year old, respectively, I think that you could get even better performance by going with a newer drive, like a 100 or 133GB/platter Seagate (7200.8), which seem to have impressive STR scores, as well a good 5-year warranty. Some of the newest 7200 drives can even compete with the earlier first-gen 36GB 10K Raptors in performance, which is itself notable. I'm kind of a "WD guy" myself, the Maxtor I picked up only because the price was so good. Whatever drive you get, make sure to "exercise" it quite a bit before using it to store anything important. The Maxtor sure needed it, and I find it to be a good rule-of-thumb regardless.

(Oh yes, having seperate drives for doing Ghost 2003 disk-to-disk image backups also makes them fly. I use a Promise Ultra100 TX2 controller, which allows Ghost (and any DOS program, really) to use UDMA speeds in DOS. Some controllers like the SI/CMD ones don't do that, unless you add a special driver to your Ghost boot disk.)

 

[Arcanedeath]

Assuming both the 80 and the 200gb drive have the same platter density and are made by the same manufacturer then their should be no noticible difference in performance, in most cases though larger drives have higher density platters so have to seek less for the same amount of data that a smaller drive would thus improving performance, so in some cases larger drives may perform better than smaller drives but it all depends on the drives you are getting. Check out Storagereview.com 's Drive index to compare the drives you are purchasing w/ your other options or for more info on what your looking for.

 

[Pariah]

Originally posted by: Tostada

Originally posted by: Chainzsaw
A 200GB HD IS faster than an 80GB one.
The reason? More platters, much more dense.
Only 250GB+ drives can compete with a 74GB Raptor.

But for response times, the Raptor stomps hands down.

You say your getting 2GB of ram, yet you can't afford at least a 120GB HD?

I saw for what your doing, cut down to 1.5 or even 1GB of ram and get more HD space.

Possibly dual 200's.

But, in the end it's your choice.

GL

Where are you coming up with this crap?

How exactly does more platters mean "much more dense"? More platters just means more platters. Do you even know what density is?

I have no idea what he was trying to say, but you're not correct either. For a given size of data, the larger capacity will be faster. How much? Depends on the drive, the descrepancy in platter count and what the drives are being used for. Simple example, (these are ballpark numbers that don't mirror real world drives but still accurately portray the concept being explained):

Say you have one 80GB drives (1 platter) and one 240GB drive (3 platters). Now, lets say you have 80GB of storage you need to store. 80 GB's would completely fill the 1 platter drive, yet only fill 1/3 of the larger drive. This means that the smaller drive might have the search the entire platter surface for data, while the larger drive would never have to search more than 1/3 of the platter surface (if you made an 80GB partition). A 2/3's reduction in average seek time (6ms for 7200RPM ATA drive) would improve average access time by more than moving to a 10K SCSI drive. The 2nd part is STR which degrades as you move towards the center of the drive. So while the 80th GB on the 1 platter drive is reading at around 35MB/s, on the 3 platter drive, that same piece of data would be read around 55MB/s. So using a larger capacity drive could yield a significant performance boost depending on the application (50% boost in STR, almost 50% reduction in average access time for this example). Again, these are oversimplified ideal scenario numbers to portray a point, but it is false to blindly claim that different capacity drives with the same areal density will perform the same. You may not notice the difference for what you do, but in some situations, the performance difference can be pretty dramatic.

It should also be noted that drives get a lot slower as they get fuller, and also become much more fragmented when almost full, so by having a larger drive, just having more free space on it, can boost your performance.

 

[Zucarita9000]

It should also be noted that drives get a lot slower as they get fuller, and also become much more fragmented when almost full, so by having a larger drive, just having more free space on it, can boost your performance.

Yes. That's true. The 80GB OS/Apps drive will not be completelly filled up. I currently have around 10GB of data between my OS and apps, so there'r gonna be more than enough free space.

I'm currently using a system with two drives, and I do notice a performace difference between this and other one-hdd sytems, wich is why I decided to put two drives in my new rig. It will also help in video capturing and editing.

Don't forget that this computer will get 2GB of RAM, wich means that I expect a lot less disk access for virtual memory and scratching.

 

[Tostada]

Originally posted by: Chainzsaw

What is your problem, you seem to have a chip on your shoulder, actually, make that a parrot...
I meant by much more dense, is that it has more platters in the same amount of space.
But, usually HD makers with newer generations use LESS platters and use platters that are much more dense (100GB/platter).
So yes, I do know about density. Next time before you start blasting off your nails, know what the hell your talking about.

I don't have a chip on my shoulder. Your post was 100% incorrect. You don't know what you're talking about and you should stop making things up. Let's look again at the satements you have made:

Originally posted by: Chainzsaw
A 200GB HD IS faster than an 80GB one.
The reason? More platters, much more dense.
Only 250GB+ drives can compete with a 74GB Raptor.

These statements are absolutely wrong. You are wrong in every way. There's no getting around it. More platters does not mean something is more dense; it actually means the opposite in many situations. 250+ GB drives still can't really compete with a Raptor 74, and the simple fact is that a modern 80GB drive is just as fast as a modern 250GB drive because they both use the same platters. You just don't get it. An 80GB drive (one 82GB platter) performs almost exactly the same as the 250GB model (three 82GB platters), because they have the same density, same RPM, same seek time, same everything except number of platters.

 

[Tostada]

Originally posted by: Pariah
I have no idea what he was trying to say, but you're not correct either. For a given size of data, the larger capacity will be faster. How much? Depends on the drive, the descrepancy in platter count and what the drives are being used for. Simple example, (these are ballpark numbers that don't mirror real world drives but still accurately portray the concept being explained):

Say you have one 80GB drives (1 platter) and one 240GB drive (3 platters). Now, lets say you have 80GB of storage you need to store. 80 GB's would completely fill the 1 platter drive, yet only fill 1/3 of the larger drive. This means that the smaller drive might have the search the entire platter surface for data, while the larger drive would never have to search more than 1/3 of the platter surface (if you made an 80GB partition). A 2/3's reduction in average seek time (6ms for 7200RPM ATA drive) would improve average access time by more than moving to a 10K SCSI drive. The 2nd part is STR which degrades as you move towards the center of the drive. So while the 80th GB on the 1 platter drive is reading at around 35MB/s, on the 3 platter drive, that same piece of data would be read around 55MB/s. So using a larger capacity drive could yield a significant performance boost depending on the application (50% boost in STR, almost 50% reduction in average access time for this example). Again, these are oversimplified ideal scenario numbers to portray a point, but it is false to blindly claim that different capacity drives with the same areal density will perform the same. You may not notice the difference for what you do, but in some situations, the performance difference can be pretty dramatic.

It should also be noted that drives get a lot slower as they get fuller, and also become much more fragmented when almost full, so by having a larger drive, just having more free space on it, can boost your performance.

You are wrong. Many things you say are correct, but not your conclusion.

In comparing similar drives with the same density, you're suggesting that a larger drive would be faster because for a given amount of data the head will have to travel farther on a single platter than it would with multiple platters. This is only correct if both drives contained the same amount of data and were totally defragmented. Your logic is valid. Essentially you are arguing that a 250GB drive will be faster than an 80GB drive in certain appliations if you make sure you always leave most of the drive empty, but this is largely irrelevant, because if you buy a 250GB drive it would be ridiculous to only ever use 80GB of it. You are also assuming the worst-case scenario in which the 80GB drive is completely full and has to seek from one end all the way to the opposite end. This is almost never the case, and benchmarks will show you that the impact of this is negligable.

The same is true for sequential transfer rate. You are arguing that the larger drive is better because you get more data in the area with a higher transfer rate, but how is this relevant? Are you going to partition off the drive and then never use most of it?

I can appreciate the valid theoretical points you bring up, but the bottom line is that drives from the same line have almost identical performance regardless of platter count. The numbers speak for themselves.

StorageReview only has benchmarks from the original WD800JB with two 40GB platters, but if you want to compare a modern 3-platter to a modern 5-platter drive, look at the Hitachi 7K400 vs. the 7K250. They have almost the exact same numbers. The 3-platter wins the server suite by about 2% and the 5-platter wins the workstation suite by about 2%. This is very impressive, because the 7K400 is a newer drive, and updates are usually made to improve performance at least a little bit, but the benchmarks are still split.

The simple fact is that drives with less platters have slightly better access times because they're not moving as many heads around. They're also usually cooler because they're not spinning as much mass. They're usually quieter, but minor differences in noise is more about the level of acoustic management. Drives with less platters are also more reliable because of the same reasons, and because there are less heads to possibly fail.

In the end, this is also mostly negligable. The bottom line is that they will perform almost identically.

 

[0roo0roo]

storagereview.com

and yup, can't compare unless same density and platters/drive/cache etc. u can't just say in general one would be faster based on its size.

 

[VirtualLarry]

Originally posted by: Pariah
I have no idea what he was trying to say, but you're not correct either. For a given size of data, the larger capacity will be faster. How much? Depends on the drive, the descrepancy in platter count and what the drives are being used for. Simple example, (these are ballpark numbers that don't mirror real world drives but still accurately portray the concept being explained):

Say you have one 80GB drives (1 platter) and one 240GB drive (3 platters). Now, lets say you have 80GB of storage you need to store. 80 GB's would completely fill the 1 platter drive, yet only fill 1/3 of the larger drive. This means that the smaller drive might have the search the entire platter surface for data, while the larger drive would never have to search more than 1/3 of the platter surface (if you made an 80GB partition). A 2/3's reduction in average seek time (6ms for 7200RPM ATA drive) would improve average access time by more than moving to a 10K SCSI drive. The 2nd part is STR which degrades as you move towards the center of the drive. So while the 80th GB on the 1 platter drive is reading at around 35MB/s, on the 3 platter drive, that same piece of data would be read around 55MB/s. So using a larger capacity drive could yield a significant performance boost depending on the application (50% boost in STR, almost 50% reduction in average access time for this example). Again, these are oversimplified ideal scenario numbers to portray a point, but it is false to blindly claim that different capacity drives with the same areal density will perform the same. You may not notice the difference for what you do, but in some situations, the performance difference can be pretty dramatic.

But that performance advantage from "short-stroking" the drive, only applies if you never fill it with more than 80GB. For normal usage of a 200GB, that's a pretty unlikely assumption.

Plus, you also have to factor in things like the greater inertia from the larger mass of the actuator in a drive with more platters, and thus more heads. Also, the head-settling time when seeking, is often slightly greater on drives with more platters, and they also suffer from slightly more vibration, as I understand it.

So those factors tend to cancel out any percieved advantage from a larger drive, all other things being equal, IMHO.

Originally posted by: Pariah
It should also be noted that drives get a lot slower as they get fuller, and also become much more fragmented when almost full, so by having a larger drive, just having more free space on it, can boost your performance.

It's pretty-well known that if you short-stroke a drive, you can obtain performance that almost reaches the next-higher level of drive performance, in terms of "RPM class". IOW, short-stroke a fast 7200 RPM drive to obtain STRs and reduced seek times comparable almost to a 10K RPM drive. But most consumers purchase drives for storage space, not performance, so that theoretical advantage rarely applies.

I try to make the best of both worlds - divvy up the drive space with several partitions, and put the most performance-sensitive applications at the beginning, and use the rest of the drive for lesser-used "bulk storage" purposes.

 

[Chainzsaw]

Originally posted by: Tostada

Originally posted by: Chainzsaw

What is your problem, you seem to have a chip on your shoulder, actually, make that a parrot...
I meant by much more dense, is that it has more platters in the same amount of space.
But, usually HD makers with newer generations use LESS platters and use platters that are much more dense (100GB/platter).
So yes, I do know about density. Next time before you start blasting off your nails, know what the hell your talking about.

I don't have a chip on my shoulder. Your post was 100% incorrect. You don't know what you're talking about and you should stop making things up. Let's look again at the satements you have made:

Originally posted by: Chainzsaw
A 200GB HD IS faster than an 80GB one.
The reason? More platters, much more dense.
Only 250GB+ drives can compete with a 74GB Raptor.

These statements are absolutely wrong. You are wrong in every way. There's no getting around it. More platters does not mean something is more dense; it actually means the opposite in many situations. 250+ GB drives still can't really compete with a Raptor 74, and the simple fact is that a modern 80GB drive is just as fast as a modern 250GB drive because they both use the same platters. You just don't get it. An 80GB drive (one 82GB platter) performs almost exactly the same as the 250GB model (three 82GB platters), because they have the same density, same RPM, same seek time, same everything except number of platters.

I'm going to call you and Idiot and anarchist for being closed minded. I am not in every way wrong, you are stupid to think that a Raptor owns everything, sorry to hurt your ego (actually, im not) but newer bigger drives (250+) will pwn the raptor 74. It may not be in seek time, but R/W it will get owned.

Once you grow up beyond your 15-year old mind, maybe people will take you seriously.
Fool

 

[VirtualLarry]

Originally posted by: Chainzsaw
I'm going to call you and Idiot and anarchist for being closed minded.

LOL. I didn't know that there was any such direct association between the two. If anything, I would have thought that "closed minded" would have equated with "conformist", not "anarchist". Where do these people come from, anyways? Is this what four more years of Bush will bring? Endless n00bs?

Originally posted by: Chainzsaw
Once you grow up beyond your 15-year old mind, maybe people will take you seriously.

Let me say this - if Tostada is in fact 15 years old, which I highly doubt, then I have to say that he is obviously one heck of an intelligent young person.

You, on the other hand... no comment.

 

[Tostada]

Chainzsaw:

So am I to assume that your official position is that StorageReview.com is wrong about everything? The benchmarks are bogus? You know more about hard drives than Eugene?

You're right. I am closed-minded. I don't believe you.

 

[Chainzsaw]

"LOL. I didn't know that there was any such direct association between the two. If anything, I would have thought that "closed minded" would have equated with "conformist", not "anarchist"."

Being closed minded doesn't necessarily mean it will equate to conformist as you have stated. It all depends on how you APPLY it.


"Where do these people come from, anyways? Is this what four more years of Bush will bring? Endless n00bs?""

Where do which people come from anyways? Are you referring to me? What in the world has brought you to think Bush will "bring" more noobs?

"Let me say this - if Tostada is in fact 15 years old, which I highly doubt, then I have to say that he is obviously one heck of an intelligent young person. "

You and Tostada seem to share an unsettling resemblence, which is not a good thing. /stating the obvious

"You, on the other hand... no comment. "

I thought you were a logical person, but after reading what you have said, well, it brings me to think about what I just said above.

Masquerading as multiple people is not a cool thing, and it's only usually done to boost ones ego.

BTW Tostada, are you saying ALL HD's use ONLY 80GB HD's? Do you not think manufactueres would want to use LESS platters to bring in MORE profit?

Oh well, it's like trying to teach a wolf, of course he won't listen...

 

[Chainzsaw]

LOL Tostada, you don't have to believe me.

I'm done with this thread anyways, it's a huge waste of time.

 

[Pariah]

Originally posted by: Tostada

Originally posted by: Pariah
I have no idea what he was trying to say, but you're not correct either. For a given size of data, the larger capacity will be faster. How much? Depends on the drive, the descrepancy in platter count and what the drives are being used for. Simple example, (these are ballpark numbers that don't mirror real world drives but still accurately portray the concept being explained):

Say you have one 80GB drives (1 platter) and one 240GB drive (3 platters). Now, lets say you have 80GB of storage you need to store. 80 GB's would completely fill the 1 platter drive, yet only fill 1/3 of the larger drive. This means that the smaller drive might have the search the entire platter surface for data, while the larger drive would never have to search more than 1/3 of the platter surface (if you made an 80GB partition). A 2/3's reduction in average seek time (6ms for 7200RPM ATA drive) would improve average access time by more than moving to a 10K SCSI drive. The 2nd part is STR which degrades as you move towards the center of the drive. So while the 80th GB on the 1 platter drive is reading at around 35MB/s, on the 3 platter drive, that same piece of data would be read around 55MB/s. So using a larger capacity drive could yield a significant performance boost depending on the application (50% boost in STR, almost 50% reduction in average access time for this example). Again, these are oversimplified ideal scenario numbers to portray a point, but it is false to blindly claim that different capacity drives with the same areal density will perform the same. You may not notice the difference for what you do, but in some situations, the performance difference can be pretty dramatic.

It should also be noted that drives get a lot slower as they get fuller, and also become much more fragmented when almost full, so by having a larger drive, just having more free space on it, can boost your performance.

You are wrong. Many things you say are correct, but not your conclusion.

In comparing similar drives with the same density, you're suggesting that a larger drive would be faster because for a given amount of data the head will have to travel farther on a single platter than it would with multiple platters. This is only correct if both drives contained the same amount of data and were totally defragmented. Your logic is valid. Essentially you are arguing that a 250GB drive will be faster than an 80GB drive in certain appliations if you make sure you always leave most of the drive empty, but this is largely irrelevant, because if you buy a 250GB drive it would be ridiculous to only ever use 80GB of it. You are also assuming the worst-case scenario in which the 80GB drive is completely full and has to seek from one end all the way to the opposite end. This is almost never the case, and benchmarks will show you that the impact of this is negligable.

The same is true for sequential transfer rate. You are arguing that the larger drive is better because you get more data in the area with a higher transfer rate, but how is this relevant? Are you going to partition off the drive and then never use most of it?

I can appreciate the valid theoretical points you bring up, but the bottom line is that drives from the same line have almost identical performance regardless of platter count. The numbers speak for themselves.

StorageReview only has benchmarks from the original WD800JB with two 40GB platters, but if you want to compare a modern 3-platter to a modern 5-platter drive, look at the Hitachi 7K400 vs. the 7K250. They have almost the exact same numbers. The 3-platter wins the server suite by about 2% and the 5-platter wins the workstation suite by about 2%. This is very impressive, because the 7K400 is a newer drive, and updates are usually made to improve performance at least a little bit, but the benchmarks are still split.

The simple fact is that drives with less platters have slightly better access times because they're not moving as many heads around. They're also usually cooler because they're not spinning as much mass. They're usually quieter, but minor differences in noise is more about the level of acoustic management. Drives with less platters are also more reliable because of the same reasons, and because there are less heads to possibly fail.

In the end, this is also mostly negligable. The bottom line is that they will perform almost identically.

How many people have 250GB of frequently accessed data? Any home user? You're too close minded. You don't have to leave the drive empty to benefit from shortstroking. Years ago people came up with a way to "partition" drives. And by using this secret method you can force the drive to remain within a certain area of the drive for only frequently accessed data, while creating one or more additional partitions on the rest of the drive for infrequently used data. That way you can guarantee your data doesn't drift all over the platters and remains on the fastest part of the drive for important data, yet you can still use the whole drive. Pretty amazing, huh?

 

[desy]

OK from storage view for the reading challanged.
In the same family of drives you get a 5% improvement in speed for every doubling of a drive so an 80 is 5 better than a 40
40
80 - 5% better
160 - 10% better than the 40, 5% better than the 80

So marginal improvement, but 10% on your slowest component would be noticible IMHO

 

[Tostada]

Originally posted by: Chainzsaw
BTW Tostada, are you saying ALL HD's use ONLY 80GB HD's? Do you not think manufactueres would want to use LESS platters to bring in MORE profit?

What are you trying to say here? Obviously you're having trouble with the English language.

And yes, it is definitely a fact that manufacturers want to use less platters to make more profit. That's why the WD800JB now uses a single 80GB platter, whereas it used to use two 40GB platters, and (at least the non-special edition version) used three 27GB platters before that. Because it uses an 80GB platter, it performs almost identically to other WD drives that use 80GB platters.

 

[Tostada]

Originally posted by: desy
OK from storage view for the reading challanged.
In the same family of drives you get a 5% improvement in speed for every doubling of a drive so an 80 is 5 better than a 40
40
80 - 5% better
160 - 10% better than the 40, 5% better than the 80

So marginal improvement, but 10% on your slowest component would be noticible IMHO

Where did you come up with these numbers, and why do they contradict the facts seen in benchmarks?

 

[Tostada]

Originally posted by: Pariah
How many people have 250GB of frequently accessed data? Any home user? You're too close minded. You don't have to leave the drive empty to benefit from shortstroking. Years ago people came up with a way to "partition" drives. And by using this secret method you can force the drive to remain within a certain area of the drive for only frequently accessed data, while creating one or more additional partitions on the rest of the drive for infrequently used data. That way you can guarantee your data doesn't drift all over the platters and remains on the fastest part of the drive for important data, yet you can still use the whole drive. Pretty amazing, huh?

You can argue all you want, and theoretically you have a point, but the benchmarks prove you wrong.

Yes, you can partition off the beginning of your 250GB drive, but it just doesn't matter. Most of the benchmarks focus on the beginning of the drive, and they prove that it doesn't make much of a difference.

I don't know what else to say. It is unfortunate that multiple people here are giving out bad information with nothing to back it up. Being an old member with almost 5000 posts, you should be particularly ashamed. It's pathetic for you to call me "closed minded" for looking at the hard numbers. Even if I were to ignore the benchmarks, there are just as many arguments in favor of a lower platter count, like improved access times, improved reliablilty, and less heat.

 

[desy]

Its not so much the drive is faster it isn't, but usage affects it too
If you take a bigger size of the same model drive, you inevitably end up with more of the space in the faster part of the platters. Those numbers represent equal amounts of data on progressivly larger drives.
Sorry I forgot to qualify that, data as a % of space is not worth mentioning.

This from a drive designer from Maxtor on storage review

"
The answer is "yes & no".

The recording densities are the same for each of the 3 capacity points, the additional capacity is achieved by providing more tracks/cylinder. Since the amount of data on a single track is the same for a given cylinder, regardless of the number of platters, the "on-track" (meaning, no seeks or head switches) transfer rates will be identical regardless of the number of platters. *However*, seek times are NOT the same for all 3 capacity points. The lower capacity drive has fewer heads, so it has a lower inertia rotary actuator, so it can generally be made to seek faster. Since *real* data transfers usually have a need to seek, the lower-capacity drives *can* deliver higher performance by virtue of their faster seek times (the 36G A10KIV has an average seek time that's about 10% faster than the 146G version).

You might be tempted to buy a lower capacity drive because of this, but like so many other things in life, there's a catch. The length (circumference) of a data track varies as a function of the radius (that 2piR thing). This means that inner radius tracks have less data than outer radius tracks because they're shorter. Since transfer rate is a function of how much data is recorded on a track, inner tracks by necessity have lower transfer rates. As a result, in some regards a higher capacity drive can perform better than a lower capacity drive.When you transfer a 36GB file to/from a 36GB drive, you use the whole disk, and are on the slowest inner tracks by the time you get to the end of the file. If you transfer the same 36GB file to a 146GB drive, you're still at least 1/4 of the way in from the outer radius at the end of file (it's actually better than that, since area varies by the *square* of radius). The transfer rates at the outer radii are much higher than at the inner radii. The 146G drive performs better in this transfer than the 36G drive.

Storage Review (and many other bechmarkers) choose the highest-capacity designs to review in an attempt to provide the most even comparison between different suppliers products. However, you can usually count on the lower-capacity designs to perform better in workloads that involve predominantly random seeks because they usually have faster move times. The higher capacity designs provide better transfer rates as a function of GBs-used-on-the-drive (but not % of capacity) because of their higher data density"

 

[Tostada]

http://www.storagereview.com/a...109Renaissance_12.html

3. All drives within a given family perform the same, regardless of size. We're not going to accept complete credit for this one. When taken in proper context, this statement stands. The problem with this assertion in a literal sense is that data is addressed absolutely, not relatively. For example, a 10 GB partition occupies only one-tenth of a 100 GB drive but a full half of a 20 gig unit. When performing a full stroke across the partition, the 100 GB disk in reality performs a relatively short seek while the poor 20 GB drive has to pull out a half-stroke. Because of this, a 160 GB Maxtor DiamondMax D540X can outperform a 40 GB version of the same drive.

Now, that said, the larger drive's advantage is mitigated by a few factors. First, locality's influence will restrict the majority of either drive's movements to a relatively tiny area. Second, keeping in mind this same locality in conjunction with buffer strategies, the impact of seek time on overall drive performance is not nearly as significant as we've stated in the past. Finally, though it may not be published in specs, a smaller drive's lighter actuator may be able to travel physical (ie, centimeters or inches... as opposed to tracks and cylinders) distances slightly faster.

In fact, the latter phenomenon sometimes results in manufacturers specifying a lower average seek time for a smaller capacity drive within a given family. Seagate's Barracuda ATA IV is the latest example. Two-platter versions of the drive feature a 9.5 millisecond seek time while the one-disk units receive a 9.0 ms claim. Remember, though, in an absolute sense that while the 40 gig drive may be able to move its head a distance of one centimeter slightly faster than the 80 gig drive can, the 80 gig drive will have to move such distances less often. In many cases, it all balances out.

At this time it is not feasible for SR to test every member of every family... we're not sure how many manufacturers would go for consistently shipping us every capacity point within every family. Further, we're having a hard enough time keeping up as is- something that'll be exacerbated by Testbed3's more comprehensive (and thus more lengthy) tests. We're going to stick to our policy of formally testing only the flagship (largest) drive within each family. Remember, however, that if an 80 GB drive of a given family outperforms an 80 GB competitor, the 40 gig version of the same drive will in all likelihood outperform the 40 GB version of the competition. So, within this context, drives within the same family truly do perform the same.

Case in point, the Hitachi 7K250 vs. the newer 7K400. They perform almost identically -- and notice the smaller drive has an 0.6ms faster access time. The 400GB version is 2.8% faster in the Office behcnmark. The 250GB version is 4.6% faster in the file server benchmark. Overall the differences are not the slightest bit significant. You would see similar differencs when looking at different revisions of the exact same drive.

So, while it is a myth that they will all perform the same, they perform quite similarly. Smaller drives are usually slightly faster in some benchmarks and slightly slower in some benchmarks, but they are very close, as the benchmarks show. You will probably not be able to find any examples where a larger drive manufactured at the same time from the same company with the same platter density will be significantly faster or slower than a smaller version. If you know of such benchmarks, please present them or give it a rest.

 

[Tostada]

desy:

Do you realize you just posted a quote which contradicts your theory that the bigger models are consistantly 5% faster?

 

[VirtualLarry]

Originally posted by: Chainzsaw
You and Tostada seem to share an unsettling resemblence, which is not a good thing. /stating the obvious

I thought you were a logical person, but after reading what you have said, well, it brings me to think about what I just said above.

Masquerading as multiple people is not a cool thing, and it's only usually done to boost ones ego.

Oh well, it's like trying to teach a wolf, of course he won't listen...

ROTFL. Is this "silly day" on AT forums or what??

 

[Chainzsaw]

desy, theres no point in going on, Tostada is forever trapped in his steel trap mind.

"ROTFL. Is this "silly day" on AT forums or what?? "

I guess I was right...

 

[VirtualLarry]

Originally posted by: Pariah
How many people have 250GB of frequently accessed data? Any home user? You're too close minded. You don't have to leave the drive empty to benefit from shortstroking. Years ago people came up with a way to "partition" drives. And by using this secret method you can force the drive to remain within a certain area of the drive for only frequently accessed data, while creating one or more additional partitions on the rest of the drive for infrequently used data. That way you can guarantee your data doesn't drift all over the platters and remains on the fastest part of the drive for important data, yet you can still use the whole drive. Pretty amazing, huh?

Well, for the record, I believe that this issue came up on SR some time ago (well, that, and whether or not to partition). The consensus seemed to be, that by partitioning, it would actually tend to *increase* the average access time, from any random point, to any other random point on the drive. Basically, while partitioning can give some of the same apparent theoretical performance advantages as short-stroking the drive, if one also accesses the data in the "slow" parts as well, then the overall effect is to increase average random access time, and that tends to negate any potential seek advantages from "short-stroke partitioning".

Basically, if you use the entire surface of the drive, then it turns out to be a wash overall.

This is also partially due to the biggest factors in track-to-track access time being rotational and mechanical seek latency, which are sector location-independent, and desktop access patterns are far more performance-dependent on those factors, than on pure STR, so the STR advantage from staying in the short-stroked partition of the drive is negligible.