RAID 0 - How much increase in speed

Use it and time it yourself, and Anandtechs review was later shown to be crap same as their whole site basically IMO.

Good thing I gave three links from independant sources then. Even if it anandtechs article ever was fabricated, which I very much doubt.

I have yet to see one reliable source showing raid0 giving any improvement where loading is concerned. Just people on forums saying "it loads faster", most probably due to the placebo effect or some anomaly in their previous setup.
 
Last edited:
500G seems like pretty big drives to put RAID0... close to a Terrabyte of data all on a volume with a halved MTBF. Would make me a bit nervous - hope you'll be backing up all the important stuff. :)

PS: I would tend to agree with the suggestions that RAID0 doesn't give all that much performance increase. I've oscillated between RAID0 and normal drives over a few windows installs and really haven't noticed all that much improvement in day to day operation.
 
I go by my own findings, I did know how long XP-SP2 took to load in normal and Raid 0 modes (aint used in Vista as cant fit HDD cage without Mod as of 8800 Ultra length).

Also daily unrared large files and found it great, game loads faster mostly but again its claimed thats down to better coding so not as much benefit but still does not claim no benefit.

I said this before, I was given a bit of advise by a peep in USA (older guy really clued up), he said instead of trying to get every MHZ out of a CPU and GPU, add a 2nd HDD in Raid0 cause its the bottleneck of any modern PC.
 
Last edited:
I've gone through numerous RAID setups with all kinds of hard drives and I've found desktop hard drives setup in RAID 0 to have next to no increase in performance for general tasks and gaming.

Forget synthetic benchmarks, they don't really mean anything for reasons already stated earlier in this thread. If you're looking to improve performance in games, get yourself a single Raptor.

Have a read around, there are plenty independent benchmarks around from well respected sites that will show RAID 0 does not live up to the hype that surrounds it in the enthusiast market.
 
There used to be quite a large difference with PATA drives but with the latest Sata 2 drives with 16mb+ cache the difference has decreased a lot. There is still a performance boost to be had(moving large files, application and game level loading) but the performance increase is a lot less now than it was a few years ago. Stripe size is dependent on what files you are working with. For me 16-32kb works best as I don't do a lot of video work.
 
this debate is quite interesting :p
It's always been a debate because there is little in it. Sure it's faster for applications which need to process vast amounts of throughput and an ideal choice for file servers but for everyday desktop PCs and gaming, there is little in it. It's much rather buy 2 drives and have the redundancy option.

I have a spare couple of drives floating around, if I get time then I'll run some comparisions using the same system..
 
I found a difference as it was 2 Raptor X's ;)

http://www.anandtech.com/storage/showdoc.aspx?i=2101&p=10

According to anandtech it doesn't matter in the slightest. I know its not easy to compare different setups once you're using the system but next time you need to format, try a stopwatch comparison on a clean setup.

AnandTech said:
If you haven't gotten the hint by now, we'll spell it out for you: there is no place, and no need for a RAID-0 array on a desktop computer. The real world performance increases are negligible at best and the reduction in reliability, thanks to a halving of the mean time between failure, makes RAID-0 far from worth it on the desktop.
There are some exceptions, especially if you are running a particular application that itself benefits considerably from a striped array, and obviously, our comments do not apply to server-class IO of any sort. But for the vast majority of desktop users and gamers alike, save your money and stay away from RAID-0.
 
Last edited:
You have just gave 3 reasons anyhow, boot time should be faster also.

1) Going from a single 7200rpm (IDE Mode - (ST3250410AS)) to a 2 HDD RAID0 setup of 7200rpm HDDs (ST3250410AS) I noticed no difference in boot time.

Going from a 2 x HDD RAID0 setup of 7200rpm HDDs (ST3250410AS) to a 2 x HDD RAID0 setup of 10000rpm HDDs (WD360ADFD) I noticed a considerable improvement in boot time, the (WD360ADFD) RAID0 setup being noticeably faster.

2) I noticed absolutely no difference in Access times from a single (WD360ADFD) to a 2 x HDD RAID0 setup of (WD360ADFD), both being around 8ms......:)
 
1) Going from a single 7200rpm (IDE Mode - (ST3250410AS)) to a 2 HDD RAID0 setup of 7200rpm HDDs (ST3250410AS) I noticed no difference in boot time.

Going from a 2 x HDD RAID0 setup of 7200rpm HDDs (ST3250410AS) to a 2 x HDD RAID0 setup of 10000rpm HDDs (WD360ADFD) I noticed a considerable improvement in boot time, the (WD360ADFD) RAID0 setup being noticeably faster.

2) I noticed absolutely no difference in Access times from a single (WD360ADFD) to a 2 x HDD RAID0 setup of (WD360ADFD), both being around 8ms......:)

This thread has certainly opened my eyes lol.

From what your saying you got the same performance when comparing single array versus RAID-0 and the only gains were made by buying a quicker drive
 
WD claim an MTBF of 600,000hrs for their SE16 range of drives

600,000hrs / 24hrs in a day = 25,000 days
25,000 days / 365.25 days in a year = 68.446 years.

if you raid 0 those drives, then you halve the mtbf you only get 34.2 years worth of use. i doubt many people on here use their drives for 5 years, let alone 34
 
MTBF = Mean Time Between Failures and not Mean Time Before Failure

Don't confuse MTBF with the life expectancy of the drive.

Read http://en.wikipedia.org/wiki/MTBF for more info.

Wikipedia said:
A common misconception about the MTBF is that it specifies the time (on average) when the probability of failure equals the probabiliity of not having a failure.

MTBF is not to be confused with life expectancy. MTBF is an indication of reliability. An 100,000 MTBF HD can have a life expectancy of 2 years while a 50,000 MTBF HD can have a life expectancy of 5 years yet the HD that's expected to break down after 2 years is still considered more reliable than the 5 years one.
 
Last edited:
http://www.anandtech.com/storage/showdoc.aspx?i=2101&p=10

According to anandtech it doesn't matter in the slightest. I know its not easy to compare different setups once you're using the system but next time you need to format, try a stopwatch comparison on a clean setup.

I hate that site, they tell lies all the time and aint visited in years. (already gaves reasons).

I dont need them to tell me if my PC is Faster or the same speed with 1 Raptor or 2 in Raido, I am the decider on that. ;)

Instead of reading a lot of BS all over the WEB, why not try it for yourself, use a Stripe of 16-32 for all round work.

Also you think Im that dumb I dont know a clean install is faster, I am compairing clean installs on same Raptors with Raid and non Raid, and I already said I cant run 2 HDD now as of HDD Cage hitting 8800Ultra (I would need do some mods to cage).

And as to Raid0 failing, it will fail no more than any other HDD set ups, 2 HDD, like 2 sticks of Ram obv means 2x as much chance as 1 failing, thats why everyone today should have a back up external HDD, if you use Raid or not it makes sence.
 
Last edited:
I hate that site, they tell lies all the time and aint visited in years. (already gaves reasons).
That's a strong allegation against a long-established and pretty well respected site. A lot of people on this forum and elsewhere would want some evidence of that before believing it. Besides, Anandtech isn't the issue here, there are plenty other respected review sites who say exactly the same. It really doesn't serve any purpose for them to lie on this issue. At the end of the day if people are buying more hard drives with the hope of improving performance in games, etc then its the manufacturers of said hard drives and retailers who stand to gain the most; so I don't see why well respected review sites would outright lie about it.

I dont need them to tell me if my PC is Faster or the same speed with 1 Raptor or 2 in Raido, I am the decider on that. ;)
I agree its always good to look at these things objectively and to form your own opinion, but without prior in-depth knowledge there's no harm in informing yourself via good old fashioned research.

Instead of reading a lot of BS all over the WEB, why not try it for yourself, use a Stripe of 16-32 for all round work.
I've tried it with all kinds of drives (7,200rpm desktop, raptors, 10krpm enterprise and my current 15krpm enterprise drives) except for solid state drives. I've also tried it with a range of RAID controllers, from onboard (various chipsets) to software-based RAID controllers and right up to my current £500+ high-end RAID controller. It does make a difference, just not as much as sythentic benchmarks would have you believe and not in the areas where people are expecting significant performance boost.

Also you think Im that dumb I dont know a clean install is faster,
No one has said that, we're just debating.

And as to Raid0 failing, it will fail no more than any other HDD set ups, 2 HDD, like 2 sticks of Ram obv means 2x as much chance as 1 failing, thats why everyone today should have a back up external HDD, if you use Raid or not it makes sence.
When a RAM stick dies from a pair, you take it out and carry on. When a hard disk from a pair in RAID 0 dies, it usually causes a lot of inconvenience and downtime and if you don't have backups (which lets face it, not a lot of people bother doing) then you've lost all your data on the array too.
 
I dont give a monkeys, they lied about Nvidia 6800Ultra and a so called fix of a few Dll's they had for PureVideo.

IMO that site are lamers.

And I aint getting at 1 stick of ram dying meaning the other still gets you running.

Im getting at 2 HDD's in a PC obv mean a HDD can fail 2x more than a PC with 1 HDD, same as a car with 4 tyres has got 2x more chance of a flat than a bike with 2 tyres.


I have read all the BS about Raid0 over the years and todays HDD's are so well made and External HDD are so cheap there is no real risk.
 
Last edited:
Imy I don't even think it's worth trying to debate this. Your not going to be able to get anywhere meaningful when all you get from someone is "lies lies lies, lamers" and won't even consider the obvious evidence from multiple sites.
 
Last edited:
I ignore that site totally and I do not need them to tell me if Raid0 actually does anything or not.

Every single Raid0 thread is the same with haters.

Anyone can go and Google a few sites, I can show dozens showing Raid0 as faster.

I wont talk anyone out of Raid0 here, I will tell them its faster overall and same as with single HDD usage, have a back up.

Not 1 of you haters has even answered the OP, Raid0 in real life will give about a 60% gain AFAIK.

There is some aspects that seek matter and its slower in Raid0 esp on normal 7200RPM HDD as they seek slower than says Raptor 16MB's but I would still say its worth it and more and more OEM's are coming with Raid0 on their gaming rigs.
 
Last edited:
Every single Raid0 thread is the same with haters.

Please don't be so childish. Why would anyone hate raid 0? The idea is ridiculous, it's called being unbiased. You look at the proven results and make an informed descision. I use raid 0 because it's brilliant for file transfers, so I don't know how I am a "hater".
Anyone can go and Google a few sites, I can show dozens showing Raid0 as faster.
Yet you haven't shown any site showing improvements in loading times so far, despite me saying that I'd like to see one.

I wont talk anyone out of Raid0 here, I will tell them its faster overall.
Great attitude there, misinform people and have them waste money on someone that doesn't do what they want.

Not 1 of you haters has even answered the OP, Raid0 in real life will give about a 60% gain AFAIK.
I answered it in my first post. Again you are coming up with figures out of nowhere.
 
Last edited:
Back
Top Bottom