• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

NCQ Raptor Rumor @ AT

J

jebo_4jc

[H]ard|DCer of the Month - April 2011
2FA
Joined
Apr 8, 2005
Messages
14,572
From Anandtech's Storage Systems Price Guide.

anandtech said:
We have heard some talk about a SATA IO "Raptor" with native NCQ instead of TCQ on the existing Raptors. Western Digital has not been a large proponent of NCQ, and since NCQ implementation seems very subjective per manufacturer, we are very anxious to see if the final product will live up to WD's reputation. Expect more details by the time we unleash our next price guide!
Click to expand...
Wow, I know we have all been begging for a drive like this, but could it be true?
 
The benefits of NCQ to the home user are questionable at best. Personally, I would rather see a 16mb cache or a native SATA interface instead of a converter.
 
Oh, yeah. I forgot that the raptor wasn't native SATA :(

16MB buffer would be nice too I guess :p
 
StorageJoe said:
16MB cache seems like it would make a bigger difference than NCQ in single user operation.
Click to expand...
16MB Cache would be nice. Would native SATA be beneficial?
 
SATA-IO compliance hints at a native SATA interface, although we must all remember that we are dealing with the maverick WD. Still nothing prominent on WD's site though.

Generally, larger/smarter buffer improves desktop performance, while better command queuing improves server performance. However, WD's first 16MB drive was something of a disappointment, and certainly fell short of the gains Maxtor realized when they went to the 16MB buffer.

Good job keeping your ear to the ground jebo, to the top this thread goes.
 
jebo_4jc said:
...Would native SATA be beneficial?
Click to expand...
It would probably reduce the number of Raptors WD has to replace because of bizarre interface conflicts and scrambled BIOS ID strings that make the drives unusable. (We've all heard the stories about Raptors showing up as PATA drives in SATA BIOSes)
 
DougLite said:
It would probably reduce the number of Raptors WD has to replace because of bizarre interface conflicts and scrambled BIOS ID strings that make the drives unusable. (We've all heard the stories about Raptors showing up as PATA drives in SATA BIOSes)
Click to expand...

You're right there. I've seen some of these Marvell SATA-PATA bridge chips doing some very strange (out of specification) things... I can imagine there are at least a few controllers out there that may have issues with them.
 
okay....this thread has attracted some high level attention in pretty short order.....

i'm WAY off of my home turf here, but i have to ask:

how is the raptor not native SATA? i got them because they are high performance, regardless of how you choose to cut it regaring making them better. did WD use most of the electronics from one of their IDE drives, and then stick the hardware to adapt SATA to IDE onto a free area of PCB? if there is a link or RTFM in my future, i really don't mind, as opposed to a lovingly (spitefully?) hand-typed message. i just want to know.

all in all, given the sheer number of users who are pleased with the performance of their 8 MB cache IDE drives, i would be kind of suprised to see this take off in a big way, at a reasonable price point for gamers/desktop users.

oh well, i hope that desktop HDD performance keeps progressing, regardless of my blisfull ignorance of how, exactly, it is done.
 
I put this thread at the top, as many users that browse this forum want to know when the Raptor will be improved or dethroned as the fastest ATA drive money can buy.

The Raptor, at present, is a PATA interface/controller on SCSI geometry platters and SCSI level seek performance with a SATA interconnect. It uses a Marvell bridge chip to allow the PATA controller board to interface with the SATA host. With the exception of WD2500KS, all of WD's SATA drives use this bridge chip and are PATA/IDE drives at heart. So do Hitachi 7K250/7K400, Maxtor DiamondMax Plus 9, and Samsung SP P80 for their SATA versions. Seagate has never used a bridge chip. The other makers have caught on to this, and realized that using native SATA without a bridge chip is a much better idea. Native SATA logic uses less space on the control board, requires fewer chips, increases compatiblity with hosts, and is cheaper to produce going forward after overcoming the initial investment of developing the native SATA design. Seagate, as the largest maker of HDDs, has the cash to do this at first. Other makers shipped SATA drives in very small volume, so couldn't justify the investment at first. Now that OEM builders are using SATA drives and PATA is fading into oblivion, all the makers are getting onboard. Samsung's new P120 and JJ SKUs are native SATA, as are Hitachi 7K80/T7K250/7K500, and Maxtor DM10/MLIII. As I mentioned earlier, WD is now getting on board as well. You are correct that there is no noticeable performance increase by eliminating the bridge chip, but there are other factors driving this transition that I describe above.
 
i see....i was just reading the buying guide, and i almost smacked myself in the forehead concerning my 36 GB raptors, until i realized that the 74 GB units had not yet come out when i purchased mine.

i am now eyeing a pair of adaptech 29160 cards that have happened to have fallen into my lap (many thanks, to DamienThorn) and thinking about SCSI for the first time in a long time. oh well, something to sit on and think about until i finish school.
 
A Overhaul of the Raptor is much needed and wanted. I know they are still sitting pretty with the fastest desktop drive so their is no reason for them to update the line, but they really should for our sake.

I for one would like to see larger raptors, keep them at 10K rpm, see a 16MB Buffer, Native SATA II and NCQ

It would be nice.
 
that would make them cost about 70 buks more. but the ppl. who actually buy raptors are enthusiasts of PCs so i guess thats not too bad. heck id buy another raptor and sell this one off.

USMC2Hard4U said:
A Overhaul of the Raptor is much needed and wanted. I know they are still sitting pretty with the fastest desktop drive so their is no reason for them to update the line, but they really should for our sake.

I for one would like to see larger raptors, keep them at 10K rpm, see a 16MB Buffer, Native SATA II and NCQ

It would be nice.
Click to expand...
 
chronic9 said:
that would make them cost about 70 buks more. but the ppl. who actually buy raptors are enthusiasts of PCs so i guess thats not too bad. heck id buy another raptor and sell this one off.
Click to expand...

Would buy a raptor here also instead of a Maxter.
 
Priorities in a Raptor redesign (most important first)

A) Eliminate the bridge chip by going with a native SATA interface.
B) Larger buffer. (If done right, this will give a good sized performance boost without raising cost much - the 73GB drive would probably stay under $200.)
C) Increase density. Higher density will allow it to make seeks even quicker, although it will come at a cost. Right now, low density is the principle limitation of performance on the Raptor, but it's also the most complicated facet to strengthen. Perhaps WD can get density on a level with Fujitsu MAT and Atlas 10K V would give a substantial boost to performance, but would also drive the price of the 73GB version over $200.
 
DougLite said:
Priorities in a Raptor redesign (most important first)

A) Eliminate the bridge chip by going with a native SATA interface.
B) Larger buffer. (If done right, this will give a good sized performance boost without raising cost much - the 73GB drive would probably stay under $200.)
C) Increase density. Higher density will allow it to make seeks even quicker, although it will come at a cost. Right now, low density is the principle limitation of performance on the Raptor, but it's also the most complicated facet to strengthen. Perhaps WD can get density on a level with Fujitsu MAT and Atlas 10K V would give a substantial boost to performance, but would also drive the price of the 73GB version over $200.
Click to expand...


Higher density will be a problem for WD. The Raptor is their first 10k drive since about 2000- 2001 when their enterprise division crashed (and had only been around since 1994). Getting a higher areal density at a 10k spidle speed will be tough for a company like WD. All the money is in enterprise as the desktop market is very very tight. Enterprise profits fuel R&D, which is where WD is lacking. What they really need to do is buy their way back into enterprise like every one else did. Seagate bought Connor, maxtor bought Quantum's SCSI business.....

Actually WD used to be a chip manufacturer and only got into disk drives by buying Tandon's disk drive business, so buying their way in isn't something new.
 
Are the 36 gig Raptors NCQ? Newegg says they are. Maybe WD added it.
 
If you are referring to the line in the product description "Ultra/150 Command Queuing technology -- increases data transfer efficiency for higher performance for enterprise applications." then, no. That's a reference to TCQ which is an older part of the (P)ATA specs, and has also been in SCSI drives for years. Remember that the controller board on the Raptor is PATA until further notice.
 
You must log in or register to reply here.
Back
Top