Sign in to follow this  
Simen1

The new 200GB WD disks..

Recommended Posts

I was looking at the spesifications here:

http://www.wdc.com/products/Products.asp?DriveID=12

Notice that the 160 and 180GB models have the same number of platters (3), heads (6), seektimes and altso max transferrate (736Mbit/s) as the 200GB.

Isn't that a bit odd? I guess i can trust the number of heads and platters but:

If they have the same transferrate, it must mean they have the same areal density of the data, but that means that the drives use a smaller part of the available disc-surface. That again would make the seektimes lower. So why are they not?

Or told in another way: If the seektimes are true, that must mean the data is spread out over the same area of the disc-surfaces. Witch means that the areal density is lower on the smaller models. That again would mean that the transferrate would be lower.

Do WD lie? What are the TRUE spesifications? I hope storagereview will find that out in a coming review of all theese drives.

Share this post


Link to post
Share on other sites

No, it's not the same thing!

That discussion is about the areal density and number of platters. It is based on the press-realease witch said it had a 60GB/platter density and 200GB. (Nothing about number of platters/heads etc.)

The new thing is that now we know the 200GB has 3 platters and 6 heads, but the performance-spesifications of the different drives contradict them selves. The transferrarte and seektimes cannot all be true at all the drives! (160, 180 and 200GB)

Share this post


Link to post
Share on other sites

After looking at the specs, what I'd like to know is, how can they all have the same number of cylinders, head, platters, and sectors/track, but a different total number of sectors? And for the life of me, I can't multiply any of those numbers together to get something that resembles the total sectors either... :?

Share this post


Link to post
Share on other sites

Perhaps destroking when the drives are servowritten? The number of cylinders listed on a spec sheet is normally the logical number, not the physical number. The physical number of cylinders should be much larger. A quick and dirty estimate is to use the TPI to get the total cylinders, since data stroke usually hovers around 1 inch.

Seektimes should be related to the strength of magnet used and design of the ACA. Thus, if the 200 GB and 180 GB models use the same magnet-coil assembly (and I suspect they do), your seek time is the same.

Share this post


Link to post
Share on other sites
Seektimes should be related to the strength of magnet used and design of the ACA. Thus, if the 200 GB and 180 GB models use the same magnet-coil assembly (and I suspect they do), your seek time is the same.
Yes, i expect they use the same components. My point was that even with the same components it could have different seektime.

Example: A full stroke seek takes 21 ms. That is the time it takes the heads to go from the outermost cylinder to the innermost. (Lats say a 1 inch distance.) If the 160GB version uses only the outermost 80% of the disc surface, then the distance between the innermost and outermost cylinder is smaller then 1 inch. That means a a shorter distance to move the heads and thus a reduced full stroke seektime and average seektime.

Just a thougth: I wonder if it could be possible to buy the 160GB version, flash the firmvare to the 200GB-version, and low-format the drive to be 200GB? I'm shure WD have several obstacles to avoid that, but maby? Technically i belive its the exact same hardware, and no handpicking of drives capable of 200GB...

Share this post


Link to post
Share on other sites
Just a thougth: I wonder if it could be possible to buy the 160GB version, flash the firmvare to the 200GB-version, and low-format the drive to be 200GB? I'm shure WD have several obstacles to avoid that, but maby? Technically i belive its the exact same hardware, and no handpicking of drives capable of 200GB...

The formatted capacity of the drive is determined when the drive is servowritten. This is what lays out the total number of tracks for the drive. The firmware controls how the controller accesses the tracks, but a simple firmware update cannot magically "write" more cylinders/wedges than were there to begin with.

I don't think it's hand-picking of drives, per se, but using alternate formatting during servowrite to write more tracks, perhaps by slightly squeezing the size of each track down so more fit in the same stroke.

Share this post


Link to post
Share on other sites
I don't think it's hand-picking of drives, per se, but using alternate formatting during servowrite to write more tracks, perhaps by slightly squeezing the size of each track down so more fit in the same stroke.

Which means you would have slightly different amounts of data per platter. This goes back to the original point of the specs listed for the 160, 180 and 200GB drivers not adding up correctly. This theory explains it but means we don't have 60GB per platter on each model, rather 60GB on the 180GB and slightly more/less for the other two.

I guess when WD advertise the 60GB platter spec, they're talking about a 60GB "class", not 60GB exactly.

Share this post


Link to post
Share on other sites
I guess when WD advertise the 60GB platter spec, they're talking about a 60GB "class", not 60GB exactly.

True. Or maybe they're using 60 GB of engineering type (e.g. whatever 1024*1048576 comes out to) vs. 60 GB of marketing type (the whole 1 billion bytes thing). ;)

I really wish they weren't quite so vague with model numbers and definitions. It might simplify things for some folks, but for the discerning end-user it's rather a pain in the rear.

Share this post


Link to post
Share on other sites
The formatted capacity of the drive is determined when the drive is servowritten. This is what lays out the total number of tracks for the drive. The firmware controls how the controller accesses the tracks, but a simple firmware update cannot magically "write" more cylinders/wedges than were there to begin with.
How is a drive servowritten? Isn't that just a way of low-formatting the drive? Can't we just find some software to servowrite it again? (and flash the frimware of course so the bios and OS detects the right size)

Share this post


Link to post
Share on other sites

You need a servowriter to lay down the servo tracks, unless the drive does what's called "self-servowriting." I don't think the WD drives do; notice there's a small hole on the side? That's for the servo clock head. The banana-hole on the underside (or the top side, depending on model) is for the servowriter to lay tracks. My understanding is a drive that self-servowrites does not need these things.

A servowriter uses the clock head for keeping time and fancy positioning, usually with lasers and the like, to know "where" to lay the tracks with respect to certain datums. A low-level format, at least what is available at the end-user level, does not "erase" these wedges, nor do they have the ability to rewrite them.

I don't know enough about self-servowriting to say if it's feasible to "overclock" your own HDD, nor do I know which drives employ it.

Share this post


Link to post
Share on other sites
You need a servowriter to lay down the servo tracks, unless the drive does what's called "self-servowriting." I don't think the WD drives do; notice there's a small hole on the side? That's for the servo clock head. The banana-hole on the underside (or the top side, depending on model) is for the servowriter to lay tracks. My understanding is a drive that self-servowrites does not need these things.

A servowriter uses the clock head for keeping time and fancy positioning, usually with lasers and the like, to know "where" to lay the tracks with respect to certain datums. A low-level format, at least what is available at the end-user level, does not "erase" these wedges, nor do they have the ability to rewrite them.

I don't know enough about self-servowriting to say if it's feasible to "overclock" your own HDD, nor do I know which drives employ it.

As Mickey said, a servo writer is not a toy. Current writers run aroung $50k and need to be in a class 10 or better clean room (factory). The drive needs to know where it is physically and the servo wedges are what give that information. Mickey missed one point thogh, during servo track writing the banana shaped hole is used for an external actuator to push the head stack array 1/3 to 1/2 of track pitch for every servo track. Once track written, the servo area is completely protected. If anything ever overwrites a wedge all positioning information referenced by that wedge is gone, the drive is effectively blind there and no data can be stored.

An analogy I enjoy for how servo works is like this. You drive a vehicle at 100mph but are only allowed to open your eyes every 10 seconds and keep them open for 1/2 second. Every time you open your eyes you are allowed to turn the wheel to change your course. Between blinks you are reading or writing your novel...

Anyway, servo writing is set at the factory and cannot be changed later without a clean room, a servo writer, and a means to DC erase the media.

Free

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this