Brian

WD Black HDD Enhancements Announced Discussion

47 posts in this topic

So, as always no word on platter density and count. But they packed more processing power in there and hence increased performance drastically. Which, on the other hand, means their current Blacks for hundreds of $ are horribly limited by their controllers and could actually perform much faster. What a massive fail on their end!

May I suggest that most of the increased performance actually stems from finally going from 500 - 800 GB platters to 1 TB platters? The increased tracking accuracy they're talking about is probably just neccessary to enable this density at 7.2k rpm.

MrS

Share this post


Link to post
Share on other sites

Platter count question is in already, waiting for a response. Also review unit request is in, drives haven't shipped to us yet.

Share this post


Link to post
Share on other sites

Here's the platter breakdown.

1TB (one single 1TB platter)

2TB (three x 800GB platters)

3TB (four x 800GB platters)

4TB (five x 800GB platters)

Share this post


Link to post
Share on other sites

Here's the platter breakdown.

1TB (one single 1TB platter)

2TB (three x 800GB platters)

3TB (four x 800GB platters)

4TB (five x 800GB platters)

This is good and bad

Good because it should easy to bench and find out just how much a firmware/hardware improve the drive.

Bad because if they also increased the platter density they would have picked up even more throughput.

Share this post


Link to post
Share on other sites

Here's the platter breakdown.

1TB (one single 1TB platter)

2TB (three x 800GB platters)

3TB (four x 800GB platters)

4TB (five x 800GB platters)

Frankly, I would never want another non-1TB-per-platter drive, ever.

Share this post


Link to post
Share on other sites

I fail to see where the enhancements are.

If we were to take a Seagate 1TB-per-platter AF drive, its maximum sustained data rate at 7200RPM is 210MB/s. Now let's assume that we reduce the platter density to 500GB and allow all other things to remain equal. This means that the capacity would be halved, and that there would be a reduction in the number of bits per track by a factor of 1/sqrt(2), and a similar reduction in the number of tracks per inch.

Since the maximum transfer rate is directly related to the number of bits per track, then we would expect an equivalent Seagate 500GB-per-platter AF drive to have a transfer rate of ...

210 / sqrt(2) = 148.5 MB/s

Therefore ISTM that WD's current models perform only marginally better than last generation's Seagate drives.

Furthermore, if we were to reduce a 1TB-per-platter Seagate drive to 800GB-per-platter, then its transfer rate would be ...

210 / sqrt(1TB / 800GB) = 210 / sqrt(1.25) = 188MB/s.

This is still more than 10% faster than WD's drives.

Share this post


Link to post
Share on other sites

So the 1TB model must have two 500GB platters, not a single 1TB platter as claimed in "the platter breakdown"?

And how can WD claim that the new 4TB models have a 48% performance increase over the earlier versions? I see only 11%.

171 / 154 = 1.11

A figure of 11% is consistent with the bits-per-track increase for AF models over their non-AF counterparts.

Share this post


Link to post
Share on other sites

Its probably like the CPUs when they say 48% performance increase they are referring to only one component (the controller) not the overall performance of the drive.

Share this post


Link to post
Share on other sites

Here's the platter breakdown.

1TB (one single 1TB platter)

2TB (three x 800GB platters)

3TB (four x 800GB platters)

4TB (five x 800GB platters)

It doesn't make any sense to me why a manufacturer would use less than 1TB platters for any drive 1TB and over. I won't buy them either.

Share this post


Link to post
Share on other sites

What the heck is a "high resolution controller?" Is it like "blast processing" for hard drives?

I think tech writers need to be more critical of this kind of technobabble. Is WD no longer using Marvell controllers in favor of an in-house design? How does increased processor performance improve tracking precision, as opposed to improving tracking speed? Those would be interesting questions to answer, but as it is they might as well have said that the drives now have more "warp cores," and are like...faster, so buy them you silly sheep.

Share this post


Link to post
Share on other sites

It doesn't make any sense to me why a manufacturer would use less than 1TB platters for any drive 1TB and over. I won't buy them either.

It's not elegant, but there are possible reasons:

1) Short stroking for higher performance

2) More platters, faster sequential read/write

3.) Less aerial density--less demanding tracking/error correction, and possible speed increases.

4.) WD has a lot of 800GB platters they want to get rid of...

Share this post


Link to post
Share on other sites

good to know SR getting a quick sample on this imo WD HDDs desktop lineup is just way too expensive. anyway I'm still hoping to see some 4k sequential read/write tests at QD1 being done, not just some ATTO defaults at QD4 which is unrealistically faster. I use these big drives to transfer and storage files so I wanna know speed at high sequential, and small block sequential too.

onto waiting for the review.

Share this post


Link to post
Share on other sites

Since a long time manufacturers have preferred 5x800 GB over 4x1000 GB at 7.2k rpm. Even Seagate does 3x1000 at 7.2k rpm and 4x1000 only at 5.9k rpm. It could well be that the vibration generated by spinning multiple platters currently becomes too much at 4x1000 and high rpm. This would lead to pretty bad access times, which is something the Black and RE can't afford. Well, actually the 1st gen 4 TB Black had significantly worse access time than it's 2 TB predecessor.

2) More platters, faster sequential read/write

Uhm.. no! At any given point in time data is only read/written from/to a single platter in traditional HDDs. No interleaving or parallel access happens, since they've only got 1 motor for the actuators anyway. In the 90's Seagate tried several motors, but it was far too complex and expensive and quickly overtaken by simpler drives with denser platters.

What the heck is a "high resolution controller?"

It's a vague description, for sure, but I have no problem understanding it as "the higher processing power in the new controller enables more accurate tracking, which improves the drives performance". It will be interesting if they finally got back to random performance of the Blacks with 500 GB/platter :D

BTW: that 48% must be something else than STR, which obviously can't change much with just a switch to AF.

@Unityole: sequential 4k acess? Any software doing this (and failing to bundle those requests into larger blocks) should be probably be kicked right into its.. code. Random 4k at QD=1, on the other hand, happens sometimes in the real non-server world.

MrS

Share this post


Link to post
Share on other sites
@Unityole: sequential 4k acess? Any software doing this (and failing to bundle those requests into larger blocks) should be probably be kicked right into its.. code. Random 4k at QD=1, on the other hand, happens sometimes in the real non-server world.

MrS

@MrSpadge its good to have 4k random but since I'm using my HDD as storage, not OS drive I don't need high 4k random speed. I need high 4k blocksize sequential read and write because many of my files are small file size, and backing them up at 5-20 MB/s isnt fun. going back to my original post it is unfortunate reviews don't cover this area. sites using ATTO shows all block size sequential read/write but at queue depth of 4, which is unrealistically faster than queue depth of 1 (transfering files). HD Tune shows sequential read/write at QD 1 for different block size as well but most sites only do it at either block size 64/128 kb, or at 2 MB,

I wanted to see the speed at queue depth of 1, at block size of 4 kb or smaller to have a good idea which drive can perform well for for my needs, otherwise reading reviews for some benchmark numbers and none for real world usage is simply pointless.

Edited by unityole

Share this post


Link to post
Share on other sites

Uhm.. no! At any given point in time data is only read/written from/to a single platter in traditional HDDs. No interleaving or parallel access happens, since they've only got 1 motor for the actuators anyway. In the 90's Seagate tried several motors, but it was far too complex and expensive and quickly overtaken by simpler drives with denser platters.

MrS

Thanks! I knew there was only one motor, but I had assumed that the heads read in parallel. I believe the old MFM drives did read that way, but those were super-low density platters with armatures driven by stepping motors, so it was expected that the heads would all line up with the same cylinder.

Share this post


Link to post
Share on other sites

Unityole, that makes more sense to me now. Are you already doing incremental backups? Not needing to transfer files is of course the fastest way :)

It would also help a lot to pack them into simple zip files.. although in that case they have to be ready for archieving and not frequently being worked on any more.

MrS

Share this post


Link to post
Share on other sites

@MrSpadge its good to have 4k random but since I'm using my HDD as storage, not OS drive I don't need high 4k random speed. I need high 4k blocksize sequential read and write because many of my files are small file size, and backing them up at 5-20 MB/s isnt fun.

Have you tried robocopy with multithreading enabled? (i.e. /MT:n)

Share this post


Link to post
Share on other sites

Have you tried robocopy with multithreading enabled? (i.e. /MT:n)

what is robo copy?

Unityole, that makes more sense to me now. Are you already doing incremental backups? Not needing to transfer files is of course the fastest way :)

It would also help a lot to pack them into simple zip files.. although in that case they have to be ready for archieving and not frequently being worked on any more.

yes packing small files into zip files is generally the way I do things but only for long term backup that doesn't need constant changes. thing is I am always making file arrangement as well as copying stuff, benching and doing various different things to my hard drives.

It's kinda like a really good bonus to have when you know your smaller block size can transfer fast.. and I just found this out today that my HDD can transfer faster via esatap port to an esatap enclosure. my seagate HDD was able to do 80MB/s read/write at 4k block sequential where my WD blackscorpio in my computer sata port only able to do 35, was going at 55 in the esatap enclosure, pretty weird.

Edited by unityole

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