• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Best option? I'm looking for a highspeed windows drive

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
larva said:
In my experience a single top-drawer 7200rpm drive is good, two of them in RAID0 a touch better, and a single R74 a bit better yet. In the end a faster drive only helps overall satisfaction to a certain degree, but whatever that degree is, the R74 delivers it.

Seeing as how I have not personally had either the Raptor or Maxline III to test against each other I am not able to do more than research and have thought that the trade off between the Raptor's small size vs the Maxline III and the research that had the view of at least a close comparison led me to think that this thread's question asking whether a single Raptor vs Maxline III on Raid would bring about a better solution and thus I thought the reverse. That being: A pair of Maxline III's on Raid 0 would bring about a better performance than a single Raptor.

Your words don't really say that this is not the case Larva. They say that the Raptor is a better single drive solution than a Maxline III single drive solution so please allow me to pick your brain. Do you also think that the RAID 0 of Maxline III vs a single Raptor would still bring about an all around faster performance evaluation?

Thanks,

R
 
Yes, I fully believe that the R74's drastic seek performance superiority allows it to outperform any current 7200rpm IDE or SATA hard drives, even if you bring two. It's not like the benefits of RAID0 don't exist, as many would claim. But they are tiny when you are talking about the OS/application volume, and they in fact degrade seek performance slightly. Given that the R74's STR is so high, you can't hope to compete by taking drives with slower seek performance, and then trading some of that seek performance for STR. Only if the R74 was a weak sister from a STR standpoint would this have merit, and that it is not.

As you can see, the R74 generates in excess of 70MB/s at the outer edge, and greater than 50MB/s at the inner one. This is quite simply enough for 99% of the things an OS/application drive has to do. You can increase the STR to 100MB/s, or even 150, and the performance will still be limited by a 12ms seek performance. The R74s 8ms seek performance is for real, and its effect positively dwarfs interface factors and the impact of a RAID0 configuration for this application.

Make no mistake, something like a RAID0 of top 7200rpm drives like these Maxtors is a good thing. It is fast, slightly faster than one of the same drives. And it is cheap, and offers lots of storage. But a single R74 is indeed more capable from a performance standpoint unless you are capturing data streams that exceed the 50MB/s mark.
 

Attachments

  • R74.jpg
    R74.jpg
    88.8 KB · Views: 48
So, I would be better off relegating my OS to a Raptor and using two Maxline III's in a RAID 0 for off ramping video streams. This would increase both the OS response and the Media response.

If this is the case then I would like give it a try although justifying spending over $300.00 Canadian on a single 74Gig drive is a tough one. Can you give an approximate percentage differentiation between a Western Digital 8mb 200GB Primary and a 74GB Raptor when performing system drive operations?

Thanks,

R
 
No, I cannot, and neither can anyone else. Hard drive performance is the composite of the realized performance in an essentially infinite number of differerent circumstances. There is no one single numerical answer, the primary reason that people nearly completely and uniformly minunderstand drive performance. It is an extremely complex phenomenon, and society is in headlong flight from complexity at present. Unfortunately with this particular topic this trend seriously degrades understanding, as no adequate single numerical representation comes close to being revealing.

My opinion is that you unlikely to see the cost involved in buying a R74 for your boot volume as being well spent. Another factor responsible for the sucking void of understanding is that even if you make drastic changes from a baseline of adequate drive performance the impact is relatively slight. But that does not mean that the R74 does not have a drastic advantage, and anytime anyone asks the simple (and oft-repeated) question "which is faster, a single R74 or a RAID0 of 7200rpm drives", for OS/app use the answer is the Raptor.

The dissastifying aspect of continually answering these questions is that most times my effort is wasted. Even though people will ask precisely the question above, with no qualification, when they are given the answer they almost invariably launch into why they are going with the RAID for cost and storage space reasons. If you want cheap, or if you want space, there is no shortage of ways to get it. And that's a worthwhile goal. Just don't tell me the R74 isn't faster, because it is.

If you claim speed is priority 1, right now, the Raptor is without true competitors. Each user has to decide exactly what priority speed is, and how much it is worth to them. If you've not had a R74, you can easily do without it. After you've had one, it's hard to imagine settling for less.
 
larva said:
If you've not had a R74, you can easily do without it. After you've had one, it's hard to imagine settling for less.
That's clear enough for me even if tying you to a quantification is not possible :). I don't think that your effort is wasted as you have clarified this question for me with an understanding being the final result. I seldom need repetition after understanding is gained. :cool:

thanks,

R
 
ropey said:
That's clear enough for me even if tying you to a quantification is not possible :).
Well, as I explained, its not my preference, just the nature of this beast. To truly quantify the difference you must obtain examples of the drives in question and use them for the things of import to you. At every click the drive is responding to wildly different demands, and it is not feasible to simulate enough of them (or even the right ones for you) to paint a truly telling picture of the differences.
 
Back