Is wd updating the Raptor?

[keith97]

With perp drives, sata 300, and now hybrid disks coming out, does anyone know if WD is going to refresh the Raptors?

I am about to purchase a hd and noticed the markdowns/rebate at newegg for the raptor 150.

Jump now or wait?

 

[SuperFly03]

WD doesn't update the Raptor very often. The Raptor 150 came out just over a year ago, so don't expect another update for at least 8 months. It seems to be about a 18-20 month cycle between Raptor updates. The next Raptor should be 300GB, non hybrid disk most likely. Hybrid disks will make their first appearance in laptops and other energy conservative applications. It will take some time for the technology to trickle down into desktops because the need isn't nearly as great.

 

[Codesmith]

SATA 3.0 is in no way better than SATA 1.5 until hard drive approach 150 MBPs speeds.

Hybrid drives are at the moment pointless. Using flash memory to speed up the boot process and to cache small files where the latency cost of a hard drive will exceed the transfer rate cost of flash memory is an excellent idea.

But there is no need to build the memory into the hard drive. Flash memory attached to the system by any other means would work just as well.

 

[powerbaselx]

With perp drives, sata 300, and now hybrid disks coming out, does anyone know if WD is going to refresh the Raptors?

I am about to purchase a hd and noticed the markdowns/rebate at newegg for the raptor 150.

Jump now or wait?

I don't think also WD will change its offer in the next months.
Just see that Raptors still have SATA150 and no word from WD or rumour to update the interface.

So, if you want to buy, it's a good time IMO.

 

[SuperFly03]

The use of flash memory in hybrid hard drives isn't focused primarily on making hard drives faster, rather improving power usage. The idea is to read/write from the flash memory so that hard drive can stay spun down. Flash memory uses less power. the only time the hard would spin up then is when the flash memory gets near full, then it dumps to the hard drive to clear it out and then spins back down.

 

[darkguset]

SATA 3.0 is in no way better than SATA 1.5 until hard drive approach 150 MBPs speeds.

Hybrid drives are at the moment pointless. Using flash memory to speed up the boot process and to cache small files where the latency cost of a hard drive will exceed the transfer rate cost of flash memory is an excellent idea.

But there is no need to build the memory into the hard drive. Flash memory attached to the system by any other means would work just as well.

I do not agree with you. SATA spec 3.0 is not only about speed (which is still better - read RAID configurations) but about data security and reliability as well. So give me SATA 3.0 over the older specs any day! Why would you buy an older revision over 5$ difference anyway?

 

[SuperFly03]

How is SATA 300 better for data security and reliability? SATA 150 can achieve 150MB/s.... that is more than enough for RAID 0 arrays. The only time it would "limit" the array would be in burst reads, which can approach 200 MB/s.

 

[Slobogob]

If the "newer Version" is only different by offering higher bandwidth that can´t be reached with the current hardware, then i´d like to get me an icecream for that 5$ i safe by going old style. 8)

 

[darkguset]

Like i said earlier, there is a huge improvement in burst speed in RAID configurations. Now if you don't know what a RAID config is then i can't help you or even waste my time explaining why SATAII specs are better. I will just point you to the appropriate pages so you can enlight your brains:

Why is it better:
http://www.sata-io.org/namingguidelines.asp

Why is it faster:
http://www.hardcoreware.net/reviews/review-283-2.htm

My machine does 357MB/sec burst speed and yes they are both SATAII drives on RAID0. How much does your do on SATAI? And yes i am doing a lot of video editing so it beats yours anytime...

 

[turboflame]

That's great and all, but burst speed has no effect on performance. SATA II is just a higher bandwidth connection which doesn't do anything faster than SATA I.

 

[darkguset]

That's great and all, but burst speed has no effect on performance. SATA II is just a higher bandwidth connection which doesn't do anything faster than SATA I.

Ok, you people don't seem to understand some simple and basic things. So go and buy a SATAI disk. Oh! And before i forget... this year the hybrid disks are coming out... which they will eat SATAII's bandwidth for breakfast. So noooooooo, don't go buy SATAII nor Vista! Stay with your old crappy machines and be happy! And all that for 5$!!!

 

[Dante_Jose_Cuervo]

That's great and all, but burst speed has no effect on performance. SATA II is just a higher bandwidth connection which doesn't do anything faster than SATA I.

You do realize that SATA II isn't just for HDDs... although that's what it's used for. Remember the iRAM? Yeah... put that on a SATA II interface and you'll still run out of bandwidth. Now... that's not to say that Hybrid drives won't be able to take advantage of the bandwidth. Just look at the SSD drives that are crippled by their ATA interface. I'll be that if you switched them to SATA II they'd be quite a bit faster.

 

[darkguset]

That's great and all, but burst speed has no effect on performance. SATA II is just a higher bandwidth connection which doesn't do anything faster than SATA I.

You do realize that SATA II isn't just for HDDs... although that's what it's used for. Remember the iRAM? Yeah... put that on a SATA II interface and you'll still run out of bandwidth. Now... that's not to say that Hybrid drives won't be able to take advantage of the bandwidth. Just look at the SSD drives that are crippled by their ATA interface. I'll be that if you switched them to SATA II they'd be quite a bit faster.

My friend, they do not seem to understand some things... If the difference was over a big sum of money i might agreed with them. But for 5$ not to make your PC futureproof and faster... what can i say... I rest my case!

 

[Dante_Jose_Cuervo]

With perp drives, sata 300, and now hybrid disks coming out, does anyone know if WD is going to refresh the Raptors?

I am about to purchase a hd and noticed the markdowns/rebate at newegg for the raptor 150.

Jump now or wait?

Ya know... I was wondering the same thing, I would love to have a hybrid Raptor with SATA II... think of the added advantage... but only time will tell.

 

[Slobogob]

Ok, you people don't seem to understand some simple and basic things. So go and buy a SATAI disk. Oh! And before i forget... this year the hybrid disks are coming out... which they will eat SATAII's bandwidth for breakfast. So noooooooo, don't go buy SATAII nor Vista! Stay with your old crappy machines and be happy! And all that for 5$!!!

:lol:

I´d like to suggest to lay off the sugar, boy. It doesn´t do you any good. Really.

Now let´s be reasonable and look at what you are trying to say. You´re saying it´s all about the bandwidth wich would be a valid point if we were talking about NICs or RAID controllers. The slightly off topic discussion is about SATA though. As has been said before, the bandwidth of 1.5 GB is more than enough for current Harddrives. The links you showed proved that there is no difference in throughput between 3 GB and 1.5 GB Harddrives with the exception of burst.

 

[Dante_Jose_Cuervo]

No, you make a valid point. Now... the average HDD will probably NOT make use of the full bandwidth. The only other one I can see is Hitachi's 1TB hdd which claims to have a transfer rate of 1070Mb/s which comes out to about 134MB/s. Once you get close to saturating a device's bandwidth you'll start to see problems. Now... say the drive DOES get that bandwidth... then we're talking about plenty of headroom for bursting! Until then though... I can't really say much.

 

[SuperFly03]

Like i said earlier, there is a huge improvement in burst speed in RAID configurations. Now if you don't know what a RAID config is then i can't help you or even waste my time explaining why SATAII specs are better. I will just point you to the appropriate pages so you can enlight your brains:

Why is it better:
http://www.sata-io.org/namingguidelines.asp

Why is it faster:
http://www.hardcoreware.net/reviews/review-283-2.htm

My machine does 357MB/sec burst speed and yes they are both SATAII drives on RAID0. How much does your do on SATAI? And yes i am doing a lot of video editing so it beats yours anytime...

Burst transfer rate is the transfer rate from the cache on the hard drive to the system. So the maximum that any hard drive can transfer is 16MB, thus after the cache is empty the system falls back on the platters. So if you deal with a 500MB video... that leaves 468 MB to be read from the platters and that assumes that the cache is filled only with your video and you have two hard drives with 16MB caches. Burst transfer rate is probably worth 5% weight on a rating. Average transfer, seek, rotational latency (course that is a function of spindle speed), I/O's per second, those are relevant measures.

You still haven't answered my question: how does it deal with data reliability/security?

--Cuervo

I don't believe Hitachi's claims at all. 137MB/s is far faster than a Raptor. It just doesn't sound believable. Even if the areal density is that high, it can't explain the transfer rate alone.

Side Note:

Hybrid disks, how fast do you really think the flash memory will be? The design of hybrid disks is to reduce power consumption in laptops and to increase boot speed by caching certain files. In order to "chew" up the 1500Gbps the transfer rate would have to exceed 150MB/s, or at least approach 110-120 on each channel since each channel is a point to point connection and not a shared bus. Hmmm just my thoughts.

 

[Dante_Jose_Cuervo]

-Cuervo

I don't believe Hitachi's claims at all. 137MB/s is far faster than a Raptor. It just doesn't sound believable. Even if the areal density is that high, it can't explain the transfer rate alone.

Side Note:

Hybrid disks, how fast do you really think the flash memory will be? The design of hybrid disks is to reduce power consumption in laptops and to increase boot speed by caching certain files. In order to "chew" up the 1500Gbps the transfer rate would have to exceed 150MB/s, or at least approach 110-120 on each channel since each channel is a point to point connection and not a shared bus. Hmmm just my thoughts.

Like I said... Hitachi claims that the drive can reach those speeds. Can it? Probably not... but if it can come close... I'll be quite happy.

Also, on hybrid drives, while they WERE designed to lower power consumtion, one thing about the flash memory is that it's faster than a standard HDD. Fast enough to use SATA II bandwidth? Probably not, but that's what they're using port multipliers for. Why waste all that bandwidth when you can have it shared between two drives to better utilize the resources? Still, I think Hitachi might be onto something... but I'll reserve my judgments for now.

 

[SuperFly03]

I agree. I will withhold judgment on the drive, but on the sole claim of 137MB/s ... I find that hard to believe. If the evidence proves it true, then Hitachi has done something quite impressive and I will give props where props are due, but I am skeptical :tongue:

 

[bberson]

Like i said earlier, there is a huge improvement in burst speed in RAID configurations.

There's a huge improvement in burst speed at 3Gb/s in non-RAID configurations too. That'll make for huge real-life performance increases as long as all your I/O is limited to 16MB (the size of the biggest currently available HD buffer).

Or you could buy a RAID controller with its own RAM and then get burst speeds near PCIe speeds. Wouldn't that be impressive.

Now if you don't know what a RAID config is then i can't help you or even waste my time explaining why SATAII specs are better. I will just point you to the appropriate pages so you can enlight your brains. Why is it better:
http://www.sata-io.org/namingguidelines.asp

You can't help only because the clowns at SATA-IO want you to fork over your credit card to read the specifications document that explains the differences between SATA and SATA-II. And I don't blame ya' for not giving it up. But that's OK, Google is your friend. Try this: http://www.knowledgetek.com/datastorage/courses/SATA__2.5_Gold.pdf

Now I don't know about you but I think almost six hundred pages of propellor beanie stuff is a bit too much to swallow, so I resorted to Wikipedia for the Reader's Digest version...

What it says is pretty funny. "The new features are hotplugging, 300MB transfer (SATA300) and native command queuing".

But if you go to SATA-IO's web site it clearly says not to confuse 3Gb/s data transfer and SATA II.

And NCQ and hot-plug capability also exist prior to SATA II.

Ya' gotta wonder what the folks at SATA-IO are up to.

Why is it faster:
http://www.hardcoreware.net/reviews/review-283-2.htm

The problem with this review is that their testing not only toggled interface speed, it also toggled NCQ. This oversight skewed all the non sequential test results as well as the CPU utilization.

Remember, NCQ and SATA-II do not have to go hand in hand.

I'm sure there's an incremental real-world performance benefit to be had, but that review does not appear to be representative.

And yes i am doing a lot of video editing so it beats yours anytime...

Don't hurt yourself wagging your pee-pee around like that!

Look, it comes down to this... with the fastest SATA drives currently mechanically limited to about 65% of the capacity of a 1.5Gb/s interface, the only way to really exploit a 3Gb/s interface is to offer a huge cache and connect it to electronics that are up to the task. The 16MB on- board caches are a joke, relatively speaking. But putting the drives on a RAID controller with 256MB of RAM or even 1GB, that's where it'll start to pay off.

The rest of it is just pissing up a wall. Wanting SATA II because it's the latest and greatest spec is fine, but don't try so hard to justify an emotional desire on technical terms.

-Brad

 

[darkguset]

Like i said earlier, there is a huge improvement in burst speed in RAID configurations.

There's a huge improvement in burst speed at 3Gb/s in non-RAID configurations too. That'll make for huge real-life performance increases as long as all your I/O is limited to 16MB (the size of the biggest currently available HD buffer).

Or you could buy a RAID controller with its own RAM and then get burst speeds near PCIe speeds. Wouldn't that be impressive.

Now if you don't know what a RAID config is then i can't help you or even waste my time explaining why SATAII specs are better. I will just point you to the appropriate pages so you can enlight your brains. Why is it better:
http://www.sata-io.org/namingguidelines.asp

You can't help only because the clowns at SATA-IO want you to fork over your credit card to read the specifications document that explains the differences between SATA and SATA-II. And I don't blame ya' for not giving it up. But that's OK, Google is your friend. Try this: http://www.knowledgetek.com/datastorage/courses/SATA__2.5_Gold.pdf

Now I don't know about you but I think almost six hundred pages of propellor beanie stuff is a bit too much to swallow, so I resorted to Wikipedia for the Reader's Digest version...

What it says is pretty funny. "The new features are hotplugging, 300MB transfer (SATA300) and native command queuing".

But if you go to SATA-IO's web site it clearly says not to confuse 3Gb/s data transfer and SATA II.

And NCQ and hot-plug capability also exist prior to SATA II.

Ya' gotta wonder what the folks at SATA-IO are up to.

Why is it faster:
http://www.hardcoreware.net/reviews/review-283-2.htm

The problem with this review is that their testing not only toggled interface speed, it also toggled NCQ. This oversight skewed all the non sequential test results as well as the CPU utilization.

Remember, NCQ and SATA-II do not have to go hand in hand.

I'm sure there's an incremental real-world performance benefit to be had, but that review does not appear to be representative.

And yes i am doing a lot of video editing so it beats yours anytime...

Don't hurt yourself wagging your pee-pee around like that!

Look, it comes down to this... with the fastest SATA drives currently mechanically limited to about 65% of the capacity of a 1.5Gb/s interface, the only way to really exploit a 3Gb/s interface is to offer a huge cache and connect it to electronics that are up to the task. The 16MB on- board caches are a joke, relatively speaking. But putting the drives on a RAID controller with 256MB of RAM or even 1GB, that's where it'll start to pay off.

The rest of it is just pissing up a wall. Wanting SATA II because it's the latest and greatest spec is fine, but don't try so hard to justify an emotional desire on technical terms.

-Brad

An emotional desire? What are you, a psychiatrist or something? LOL

 

[SuperFly03]

Well your the one spouting benefits to a technology that has no purpose in the real world at this point. It is a paper gain, that's it. There is no evidence to support 3Gb/s is superior to 1.5Gb/s except on paper and burst cache transfer (ooooo 16MB, score).

 

[darkguset]

Well your the one spouting benefits to a technology that has no purpose in the real world at this point. It is a paper gain, that's it. There is no evidence to support 3Gb/s is superior to 1.5Gb/s except on paper and burst cache transfer (ooooo 16MB, score).

So its just me then eh?
Let's see, you seem to have SATA 2 Hard disks! Now that is an irony isn't it? And what about your 64bit CPU? There are no benefits, but you still opted for the futureproof hardware. Not to mention the tyres of 99% of the cars that run around the world. The maximum speed limit is 165km/h for a typical cheap set of tyres, yet nobody buys them, you all opt for something like 180km/h or even 200km/h. Now when are going to drive that fast? Probably never and nowhere, but you still go for that "theoretical" value. Not to mention another billion things in your everyday life (HD tvs, although only 2-3 stations broadcast true HD signals) etc etc etc...
So cut the crap and admit that YES a newer better technology is better wether you like it or not and yes you buy it despite the fact that you are trying to prove that it offers no better performance in the real world.

To conclude the differences between the theretical and practical values i will give you a very simple example: The Core 2 Duo's memory bandwidth is much smaller than the RAM's bandwidth at their maximum official speeds. Hence any kind of overclock on the CPU FSB would gain no benefits at all since the RAM bandwidth will never get saturated (not until huge speeds - impossible at the moment) are achieved. Well surprise surprise it does actually benefit on every single benchmark that is on the internet. Now go and convince 100000000 million people that that is not the case.
Good luck!

 

[MOREPOWERPLEASE123]

Man...if we READ what these people are saying, I don't see ANY of them saying NOT to buy future-proof hardware. The original question was about an update to the Raptors. They simply said that SATAII is not really a big improvement, it's not worth waiting for it on the Raptors. Does that mean they are saying "Don't buy SATAII drives"? No, they are saying it's not worth putting off buying a product HOPING that new technology is included in it that doesn't even raise real world performance to begin with. I agree. So what's the big deal that he owns SATAII drives? I will be buying SATAII really soon. Why? Because current mainstream hardware is based on it. That doesn't mean I think I need it, or that I honestly am under the belief that I'm really going to experience any speed increase from it. If I was saving a load of money to go with regular old SATA, I would.

 

[SuperFly03]

Let's see, you seem to have SATA 2 Hard disks! Now that is an irony isn't it?

They are SATA 1 ...

And what about your 64bit CPU?

Is this even in question? No. Does it provide me any benefit, no. Did I buy it for 64 bit, no. Am I even using 64 bit, no. Will it last until I use 64 bit, No. Why did I buy it.... because it spanked Intel's offering at the time.

There are no benefits, but you still opted for the futureproof hardware.

No such thing, nor do I ever plan to future proof my rig. I build it to last 2 years, that is it. Then a complete rebuild. Don't speak to motives you don't know or understand.

Not to mention the tyres of 99% of the cars that run around the world. The maximum speed limit is 165km/h for a typical cheap set of tyres, yet nobody buys them, you all opt for something like 180km/h or even 200km/h.Now when are going to drive that fast? Probably never and nowhere, but you still go for that "theoretical" value.

The faster tires have softer rubber and therefore more grip. I use that grip on a daily basis. I also have been ticketed for going 135mph, so yes I do use my $1000 tires. Nice try. I live in Texas, plenty of room to hit 120-130. Just because you are too scared to drive that fast doesn't mean others don't. There is nothing theoretical about the value of the Z rated tires. Let's see how great you "value" tires hold up in cornering, hard acceleration, and wet weather.

Not to mention another billion things in your everyday life (HD tvs, although only 2-3 stations broadcast true HD signals) etc etc etc...

Don't own any useless crap like that. I don't buy useless crap. Keep trying, you're bound to get one right.

So cut the crap and admit that YES a newer better technology is better wether you like it or not and yes you buy it despite the fact that you are trying to prove that it offers no better performance in the real world.

Spell check works wonders. It is newer, I agree, but to warrant the use of the term "better" it must show some sort of end user experience improvement and it doesn't. Therefore, it is purely theoretical. What can I do with a paper performance increase? Eat it for lunch and save $5?

To conclude the differences between the theretical and practical values i will give you a very simple example: The Core 2 Duo's memory bandwidth is much smaller than the RAM's bandwidth at their maximum official speeds. Hence any kind of overclock on the CPU FSB would gain no benefits at all since the RAM bandwidth will never get saturated (not until huge speeds - impossible at the moment) are achieved. Well surprise surprise it does actually benefit on every single benchmark that is on the internet. Now go and convince 100000000 million people that that is not the case.
Good luck!

Again spell check. You are talking purely bandwidth and leaving out frequency... The FSB is the communication between the memory controller linked to the north bridge and the CPU core. Therefore, if the speed of the communication path increases, the data travels faster. Now, whether that translates into a end user performance gains depends on the application. RAM intensive files, like compression obviously benefit, but ripping CD's doesn't gain anything. Bandwidth is the potential total memory throughput, not the actual throughput. Get your head out of the white paper and into reality.