WD Green with faster interface - just BS marketing?

[qubit]

I've got the 2TB WD20EARS drive. This has a 3Gb/s interface. The new WD20EARX is exactly the same, but has a 6Gb/s interface.

The actual read/write performance is identical according to WD's own specs, so is this faster interface just marketing BS?

To see comparisons between these and all other Green click here: http://wdc.com/en/products/products.aspx?id=120

or load the PDF here: http://wdc.com/wdproducts/library/?id=271&type=8

 

[erocker]

No, it's true. It has a faster interface. It doesn't mean the drive itself is faster.

 

[qubit]

I know the drive isn't faster, as I've said, that's why I'm wondering if it's just marketing BS, or if there's some benefit somehow.

 

[95Viper]

No, it's true. It has a faster interface. It doesn't mean the drive itself is faster.

I understand what you are saying.

I know the drive isn't faster, as I've said, that's why I'm wondering if it's just marketing BS, or if there's some benefit somehow.

erocker is right... he is not being facetious. And, I am not great at explaining stuff.
Sorta right now, the Buffer to Host speed is BS, because the sustained Host to/from drive speed is going to max out at 123 MB/s on those drives.

Like, I said, I ain't great at explaining, so maybe this will help:

Quoted from External (Interface) Transfer Rate @ StorageReview.com (read this and maybe you will get what I fail at explaining)

External transfer rate is a perennial candidate for "most overrated hard disk specification". The reason is that external transfer rate specs are usually very high and impressive; manufacturers print them in big bold letters on their retail boxes, and system makers highlight them on their spec sheets. Unfortunately, they usually have very little to do with real-world performance, because the drive's internal characteristics limit transfer performance.

 

[qubit]

@95Viper

erocker didn't understand what I was saying and certainly wasn't being facetious - I would never accuse him of that.

Ok, let me phrase my OP slightly differently. The new drives perform exactly the same, yet have a SATA interface that now works at twice the speed. The drives can't max out the 3Gb/s interface, let alone the 6Gb/s one, therefore, where's the benefit?

Hence my question, did they improve the interface simply as marketing BS to sell a 'new' drive? It sure looks that way to me.

That Storage Review article is a bit of a heavy read just to get your point and I don't have the time to read it all, so I think if you did summarize it to explain your point, that would be a good idea.

 

[newtekie1]

so is this faster interface just marketing BS?

Everything beyond SATA 1.5Gb/s has been just marking BS with pretty much every mechanical drive...

 

[95Viper]

@95Viper

erocker didn't understand what I was saying and certainly wasn't being facetious - I would never accuse him of that.

All's good.

Ok, let me phrase my OP slightly differently. The new drives perform exactly the same, yet have a SATA interface that now works at twice the speed. The drives can't max out the 3Gb/s interface, let alone the 6Gb/s one, therefore, where's the benefit?

Yep, none really... not with mechanical drives.

Hence my question, did they improve the interface simply as marketing BS to sell a 'new' drive? It sure looks that way to me.

IMO, not really, future things, maybe.

That Storage Review article is a bit of a heavy read just to get your point and I don't have the time to read it all, so I think if you did summarize it to explain your point, that would be a good idea.

Sorry. However, it is a good read.

I do think the StorageReview.com quote summed it up though.

External transfer rate is a perennial candidate for "most overrated hard disk specification". The reason is that external transfer rate specs are usually very high and impressive; manufacturers print them in big bold letters on their retail boxes, and system makers highlight them on their spec sheets. Unfortunately, they usually have very little to do with real-world performance, because the drive's internal characteristics limit transfer performance.

Everything beyond SATA 1.5Gb/s has been just marketing BS with pretty much every mechanical drive...

True

EDIT: Sorry took so long to get back to you, had to take the dog out and fix lunch for tomorrow.

 

[MN12BIRD]

There might be other benefits to getting your drives on the latest interface. Think about it what happens if your mobo only has 2 SATA ports (simple example) and they're both SATA III 6.0GBps but they're port multiplied off a single controller (aren't they usually?) if you have a fast SATA III 6.0GBps SSD on port 0 and then you add a second Hard Drive for storage to port 1...

What happens if the 2nd drive on port 1 is only SATA II 3.0GBps??

Does it slow the SSD down?

Also aren't there other improvements to SATA III other than raw bandwidth? Like are there improvements to the NQL or other things?

Seriously this is a question. I just asked it in a really odd fashion!

 

[scaminatrix]

I'm thinking like MN12BIRD. What if you've got a mobo that works at either 6.0GBps or 3.0GBps and you've got a decent SSD? There's probably some entry level boards out there that can't use both at the same time, so this HDD wouldn't cause a meltdown slowdown.

 

[Mussels]

I'm thinking like MN12BIRD. What if you've got a mobo that works at either 6.0GBps or 3.0GBps and you've got a decent SSD? There's probably some entry level boards out there that can't use both at the same time, so this HDD wouldn't cause a meltdown slowdown.

speed negotiation is done per SATA port, so no, that doesnt happen.

 

[scaminatrix]

There's me googling for 20 minutes, the Doctor does it in 4!

 

[Mussels]

There's me googling for 20 minutes, the Doctor does it in 4!

But we cant do that, who'd clean up all the mess?


sorry, had to with your red dwarf avatar >.>

 

[remixedcat]

I know of a few webhosts that had problems with these drives (they used several hundred of those) so I'd be careful. They may be ok for non-critical stuff but anything serious and I wouldn't use em.

and BTW red dwarf FTW!!! loved that show!!!

 

[Steve-007UK]

I think the reason they have sata 6Gb/s interface is because thats where tech is heading and it will probably work out cheaper in the end, Bit like how my DDR2 ram is now twice the price of what i payed for it. There pushing DDR3 and prices are a lot cheaper now than DDR2 so yeah i think thats why your HDD has sata 6Gb/s interface.

 

[btarunr]

I know the drive isn't faster, as I've said, that's why I'm wondering if it's just marketing BS, or if there's some benefit somehow.

Standards compliance, so that they don't have to use two sets of controllers or components on their production line. You know that the Radeon HD 6450 doesn't need PCI-Express 2.0 x16, hell, it won't even need AGP-4X to perform to its peak. Yet its designers gave it PCI-E x16 Gen 2. Again, standards compliance.

Also, it's possible that the 64 MB cache is faster than on the SATA 3 Gb/s generation, and having SATA 6 Gb/s could yield high burst speeds that can make use of the additional bandwidth.

 

[repman244]

64 MB cache is faster than on the SATA 3 Gb/s generation, and having SATA 6 Gb/s could yield high burst speeds that can make use of the additional bandwidth.

This.

It's the only thing that benefits from SATA 6Gb/s (on a HDD).

I remember doing a small experiment by putting my SAS controller in a PCI-e x4 slot instead of x8, the HDD's obviously didn't perform slower but my cache did, it dropped from ~1100MB/s to something around 800MB/s.

Also SATA1 wouldn't NOT be enough for today's drives, it is 150MB/s but that is theoretical, some of that bandwidth is used for communication between the drive and the controller (For this I also did a test on my storage server, where all 5 drives are connected to one U320 SCSI port, maximum read was 270MB, which means the 50MB/s was "wasted", since the theoretical speed is 320MB/s).

 

[Izliecies]

on a related note - how do you explain burst speed? how does that help the performance of a drive?

 

[repman244]

on a related note - how do you explain burst speed? how does that help the performance of a drive?

http://www.intel.com/support/graphics/sb/CS-010481.htm

Sustained versus Burst transfer rates
Sustained transfers are continued transfers that do not occur from the drive cache. Burst rates are data transfers directly to/from the high speed cache. A true indicator of performance is sustained rate; but, most drives are advertised with their faster burst rate.

It does help, but my guess is only when cache is utilized.

 

[Mussels]

on a related note - how do you explain burst speed? how does that help the performance of a drive?

if the file was already in the cache.

 

[newtekie1]

EDIT: Sorry took so long to get back to you, had to take the dog out and fix lunch for tomorrow.

I read that as you had to take the dog out to get fixed at first...

if the file was already in the cache.

Yep, and most of the time it isn't. Even if it is small enough to fit in the cache, the drive has to guess which file you are going to want to access, and it isn't good at doing that.

 

[Widjaja]

General public will see the new interface and therefore immediately think it would be faster without any technical knowledge.

I remember posting my WD 1TB Black SATA II HDTune Becnhmark in the HDTune thread.
Shortly after someone posted their WD 1TB Black SATA III HDTune benchmark.

The specs of both drives were the same besides the interface.

There was an improvement but it was so insignificant.
The difference was most likely down to the other components in the machine rather than the new interface.

Also leads to remembering when I first went from a 80GB PATA to an early 160GB SATA II drive.
There was no difference in performance in HDTune benchmark.