• 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.

Raptor 74 vs 150GB and questions about RAID

CC_Pirate

Limp Gawd
Joined
Aug 1, 2005
Messages
463
With Vista Ultimate x64 doing what it can to feel at home, I am running out of space on my main system disk (a Raptor 74GB), so its time for an upgrade.

I am considering buying another 74GB Raptor and run RAID 0 with the two disks. I have built in RAID on my Asus Maximus mobo.

Second alternative is just to buy a 150GB Raptor instead. Its only like $30 more than the 74GB.

Questions then are:
Which option will be faster? The 150gig Raptor has 16MB cache vs 8MB on the other one. Remember that this is my system disk.
I read somewhere that Windows won't install to RAID. Is this true?
Is the built in RAID on my mobo high quality (suffice to use at least)?


Thanks!
 
I just upgraded from 2 36.7 8MB Cache Raptors to a single 80GB 16MB Cache raptor and the single drive is performing much better then the 2 in RAID.

You can install Windows on a RAID Array, you just have to install the RAID drivers at the beggining of the installation.

Most onboard RAID controllers do fairly well for themselves in terms of performance.


I would go with the 150GB 16MB drive over another 74GB 8MB drive.
 
Why not just go 150gb as your main drive; put the OS and maybe games and programs on there then keep your 74 gb as a secondary hard drive and put the swap file and documents and what ever else on that drive.
 
Thanks all. I think I'll go for the 150GB. Was just wondering if the RAID would give me a performance boost over the other option, but that does not seem to be the case.
Plus, its good to avoid the hassle with RAIDing.
 
RAID is the way to go. Its always going to be faster then a single drive (unless the drives in the array are complete trash)
In this case you will see a performance gain going raid with the 'lesser' drives then one single 'better' drive
 
RAID is the way to go. Its always going to be faster then a single drive (unless the drives in the array are complete trash)
In this case you will see a performance gain going raid with the 'lesser' drives then one single 'better' drive

How big will the performance increase be?
The recommendations here seem to differ. I always thought RAID was faster though.

Not sure what to do
 
The benefits are very good. How good is up to how much you want to spend.
If you dont care about data and your wallet can absorb 3 HD's go for raid 0
If you need redundancy and can only go 2 drives raid 1
if money is no matter then go multiple arrays 1 for boot and one for apps

Sticking to your original specs raid 1 with 2 drives will smoke your proposed 1 drive.

Write time performance will not increase and may decrease just slightly over the one drive but read performance will be much much better. For gaming/video editing/etc this is mostly read intensive so no negative factors in this scenario.
If your doing a lot or writing to the disk then go raid 0 or 5.
Either way performance will be much better then a single drive.
 
This has been discussed all over the place about RAID being faster or not, if it is two drives in RAID 0, they won't smoke a single drive. A little faster? Yes but smoking a single drive? Not a chance otherwise everyone would use RAID 0 since it could smoke just a single drive.

Get a 150 raptor for OS and games and use your current 74 raptor for programs and misc.
 
You know..... you could raid the 74gig with half of the 150gig and then use the other half of the 150gig for other stuff. So it would be like having two raided 74gigs and an extra 74gig for stuff you are scared of losing.
 
You know..... you could raid the 74gig with half of the 150gig and then use the other half of the 150gig for other stuff. So it would be like having two raided 74gigs and an extra 74gig for stuff you are scared of losing.


So would it be like running a RAID with three disks, and one having the backup-bits or whatever theyre called?
 
Sort of except using both the raid array and the spare space at the same time would cause a bit of a hit since the spare space is on the same physical disk. Woule be like having one big hard drive with 148gigs of really fast and 74 gigs of sorta fast on different partitions. But if you use both at the same time its gonna thrash and be slow instead of doing both things at full speed like it could with 3 seperate drives.
 
Sell the 8mb 74gb Raptor and get 2 740ADFD's and RAID0 'em. :)

This is EXACTLY what I did when the ADFD's came out. Huge difference.

Now I find I'm bored with that and am hoping WD will give us something even faster. 32mb cache?

Will SATA ever break the 10K rpm limit? Will SAS ever come in a "friendly" non-enterprise offering?

Hard drive technology seems to move so slowly at times.
 
Hmm. Its just that the 150GB is only $30 more than the 74GB. Its tempting to get a 150GB and have the 74GB as a secondary disc, no RAID. I wish I could see some comparison benchmarks. I know RAID is faster but Im unsure if its worth the hassle.

Thanks far all input
 
Yeah, I think what I said above is my decision. One 150GB.. If I need/want more space/performance in the future I'll get a second 150GB and RAID em.
Btw, does anyone know how much more noise the "windowed" Raptor makes than the regular one?

Thanks
 
This has been discussed all over the place about RAID being faster or not, if it is two drives in RAID 0, they won't smoke a single drive. A little faster? Yes but smoking a single drive? Not a chance otherwise everyone would use RAID 0 since it could smoke just a single drive.

Get a 150 raptor for OS and games and use your current 74 raptor for programs and misc.

Just for clarification RAID 0 needs a minimum 3 drives and it will smoke a single drive all day every day
 
Lol! I am going on the assumption that he put he 3 because he knows for a fact that RAID 0 smokes any hard drive, ide, sata, scsi 15k, all those are garbage when you got RAID 0! :p:p:p

FYI: Raid 0 is a minimum of 2 drives, RAID 5 is minimum of 3, RAID 10 or 01 is a minimum of 4. Also, google = friend.
 
Just for clarification RAID 0 needs a minimum 3 drives and it will smoke a single drive all day every day


Sigh.. another uneducated individual.. RAID 0 is not everything it's appeared to be.


Yes.. I absolutely agree that RAID 0 is 3 drives.. Just like you F10 in Steam games to pull up the latest stats.
 
Sigh.. another uneducated individual.. RAID 0 is not everything it's appeared to be.


Yes.. I absolutely agree that RAID 0 is 3 drives.. Just like you F10 in Steam games to pull up the latest stats.


Agreed and not agreed...Raid is not all its cracked to be and Raid0 Only requires a minumum of 2 drives. do your homework
 
Get the 150 and use the 74 for your games and such.
There is little to no benefit to having a RAID 0 on a desktop computer.

Or, if you have a ton of games, leave the OS on the 74 and use the 150 for the games and apps.
 
Sell the 74g raptor and buy single 150g + 320gb seagate/WD for storage. Running single 400gb + 2 750g (raid 1) seagate 7200.10 sataII drives now in main system. Hell alot less noise + heat and just as fast. In otherwords, raid0 isnt worth the $$ or hassle to ME.
 
I love these posts of it's better, no it's not, etc; honestly if you believe you will blow away the world's best hard drive in RAID 0, go for it. If you want to brag you are a risk taker and have all your data sitting in a RAID 0, go for it.
Honestly, I run 2 - first gen raptor 74s in RAID 0, is it faster? I don't know, from day one 2 1/2 years ago I ran RAID 0, I choose to because I wanted too and am now looking to break up the array and just go single drives. RAID 0 might be a smidge faster vs another drive on some thing and it might not, it is really up to what ever you want to run.
 
Of course RAID is faster. I do have to admit that the raptors are fast drives


It all depends on what you are doing... If you are doing a lot of I/O then yeah the RAID is going to be faster but if you are looking at access times (responsiveness) a single Raptor would be faster.

Go read the link I posted above it explains everything.

But I have a feeling I am typing this all in vain.
 
I have some data that I can give you that can help you out.

This is on a Vista Ultimate x64 (3 day old install).

E6600 2.4ghz
2Gb Ram
Asus P5N32-E SLI mobo

Running HD Tach 3040 under XP SP2 Comp mode.

A pair of WD 150gb Raptors in Raid 0 using the onboard Nvidia Sata Raid.
Random Access: 8.8ms
CPU Utilization: 4% (+/-2%)
Avg Read: 138.7MB/s
Burst Read: 218.1MB/s

A Pair of WD 500GB in Raid 0 on the same computer.
Random Access: 13.7ms
CPU Utilization: 6% (+/-2%)
Avg Read: 113.7MB/s
Burst Read: 209.1MB/s

To give you an idea of how much Raid is making a difference, here is a single identical WD 500GB HD not in Raid on the same computer.
Random Access: 13.5ms
CPU Util: 6% (+/-2%)
Avg Read: 70.5MB/s
Burst Read: 172.3MB/s



I bought those Raptors just a couple of days ago, and while they are fast, I'm not terribly impressed with them. They might get taken back and I might just get a pair of WD 500GB hd's for have the price.

I might buy two 1 TB hd's if I can find some that I like and raid those, but I don't know if I will be keeping these raptors. They are fresh drives, but they don't feel that any faster than my 500's.
 
I have some data that I can give you that can help you out.

This is on a Vista Ultimate x64 (3 day old install).

E6600 2.4ghz
2Gb Ram
Asus P5N32-E SLI mobo

Running HD Tach 3040 under XP SP2 Comp mode.

A pair of WD 150gb Raptors in Raid 0 using the onboard Nvidia Sata Raid.
Random Access: 8.8ms
CPU Utilization: 4% (+/-2%)
Avg Read: 138.7MB/s
Burst Read: 218.1MB/s

A Pair of WD 500GB in Raid 0 on the same computer.
Random Access: 13.7ms
CPU Utilization: 6% (+/-2%)
Avg Read: 113.7MB/s
Burst Read: 209.1MB/s

To give you an idea of how much Raid is making a difference, here is a single identical WD 500GB HD not in Raid on the same computer.
Random Access: 13.5ms
CPU Util: 6% (+/-2%)
Avg Read: 70.5MB/s
Burst Read: 172.3MB/s



I bought those Raptors just a couple of days ago, and while they are fast, I'm not terribly impressed with them. They might get taken back and I might just get a pair of WD 500GB hd's for have the price.

I might buy two 1 TB hd's if I can find some that I like and raid those, but I don't know if I will be keeping these raptors. They are fresh drives, but they don't feel that any faster than my 500's.

Very cool, thank you. I see that RAID gives a nice performance increase. However I will buy 1 150GB Raptor first and run it "alone" in my main system.

However, my HTPC has a 500GB WD drive now that is filled up. I am buying a second one of those and testing RAID 0 on that rig first.

Thanks for all input here. My decision has been made.
 
Agreed and not agreed...Raid is not all its cracked to be and Raid0 Only requires a minumum of 2 drives. do your homework

I have... Have you played Counter-Strike Source.. F10 does NOT give u the stats.. just boots you from the program..

Sigh... I ran 2 36GB raptors in RAID 0 for 2 years. It's not the "be all, end all" gaming / I/O solution. Personally, IMO, for home use, a single fast drive is more than adequate, as well a a single Drive for your OS.
 
I have some data that I can give you that can help you out.

Running HD Tach 3040 under XP SP2 Comp mode.

A pair of WD 150gb Raptors in Raid 0 using the onboard Nvidia Sata Raid.
Random Access: 8.8ms
<blah blah blah>

Nice synthetic benchmarks. Do you suggest I select which video card to buy based on theoretical pixel fillrate or would you go the [H] route and see how well it plays the game(s) that I care about? Yes, STR and average access times are nice metrics to have and may well define the I/O system's performance if we knew how to weight each of these correctly. However, I do not know how seek intensive or how STR intensive any game is, hence it is hard to make any decision based on the numbers.

Sure we can make a pretty simple statement: an I/O system with a higher STR and lower seek time is going to be faster -on average- than an equivalent system with less STR and a higher seek time, provided that the caching algorithm employed is the same.
 
I've never seen a hard drive review that had a benchmark for a game. I've seen some that had benchmarks for windows load times, but then again I don't look at to many hard drive reviews.

The OP was wanting to know about Raid performance of Raptor's, and I gave him/her benchmarks, plus benchmarks from other drives in my system (with and without Raid) to help him/her compare.

I also gave my subjective opinon on the product (that they don't really feel much faster, which is backed up by the data showing read speeds are close but seek time is much lower on the raptors), and then my personal opinon on what I'm going to do when my money hits the road.

I am personally unwilling to take the other drives in my system which are all nearly full and format them and copy my new windows install to them to preform identical benchmarks of windows load times and game load times. Now, if I made money doing it, or got the drives for free, sure, I would do that to the best of my ability, but not on my own free time on my own dime.\


In my opion, it seems that seek time doesn't matter as much as read rate when dealing with sequential data on a defragmented hard drive (or drives). Yes, a hard drive (or drives) with a faster seek time will get to the data quicker, but if the data is read at a slower rate it seems that if the data was of any segnificance size it would quickly be beaten by the faster read rate hard drive.

I would expect little to absolutely no difference in game on the differnt hard drives assuming sufficent ram. Once the game has the data in ram, it shouldn't have to read anything else of the hard drive. So, the only difference should be in map load times.


Running a few numbers here:
Assuming speeds give for my hard drives and compairing the Raptors in Raid to the WD 500gb's in raid to load 400mb worth of game data:

The raptors would take 2.891 seconds including seek time. (assuming the data is contiunous).
The 500's would take 3.526 seconds including seek time (assuming the data is continous).

The difference is .635 of a second. That's perceptible, but not a massive differnce.

Now, if were to include the new Seagate 1TB drives that have faster read rates than even the raptors, it wouldn't take very long for the seagates faster read rate to negate the raptors faster seek time.

The raptors have read only .6935 MB's of data before the 500's have completed their seek. A faster read rate would more than make up that difference in short notice. ~700kb won't make much of a difference in reading game data.

For the money, a pair of 500's 6.6x more space for roughly half the cost, and nearly the same speed. A pair of 1TB seagate would read faster than the raptors, and would provide more GB/$.

If one was really concerned with seek time, you could install a PCI based Ram drive for around the same cost of a pair of raptors, and have vertically no seek time. Space would be more of an issue, but it would be much faster than any hard drive base system.
 
I am sorry that my previous post was poorly worded. It wasn't meant as a personal insult and I would like to apologize. Since you are new, you are unaware of the flame-wars that have raged through this subforum about RAID-0.

True enough, there are almost no hardware review sites that use game loading times for HDD benchmarks. That doesn't mean that synthetic benchmarks are valuable to the gamer/ enthusiast. Measuring game loading times is much more time consuming than a quick Sandra/ HDTach/ Atto run...
 
Back
Top