Why have 2 raptors when you can have 4

yogibear

New member
Hi All Again

Am thinking about getting another two 36gig raptors and running all four on a striped raid arrey. Has anyone else done it and is it worth it. Obveiously its going to give me 140gig of very fast storage but is it a lot faster.

Yogi
 
I read somewhere that you won't get any performance increase from having more than three because with three you would have saturated the bandwidth available to them.
 
:confused:

Way out of my league, what I need is 'an morons guide to RAID' as I know nothing about it and its the one part of any rig I avoid due to my lack of knowledge and therefore fear.

My names is Maverik-sg1 and I have raid-o-phobia :eek:

Mav
 
yogibear said:
Hi All Again

Am thinking about getting another two 36gig raptors and running all four on a striped raid arrey. Has anyone else done it and is it worth it. Obveiously its going to give me 140gig of very fast storage but is it a lot faster.

Yogi

a while back when i was in the custom pc beat the office comp i have up to 5 34 gig raptors in stripped raid 0..............

i was top of the board for about 1 year the comp was a big con.........

but to be honest with u the best performance will be with 2x 34 gig not 74 gig raptors............

i got far better results in encoding with 2x 34 gig m8 hope this helps....
 
Dazboots said:
a while back when i was in the custom pc beat the office comp i have up to 5 34 gig raptors in stripped raid 0..............

i was top of the board for about 1 year the comp was a big con.........

but to be honest with u the best performance will be with 2x 34 gig not 74 gig raptors............

i got far better results in encoding with 2x 34 gig m8 hope this helps....

How big was the differance though, i would not have thought it could have been that much?

Also in what respect was the comp a big con?

Michael
 
In the early days of "Beat the Office," anyone who beat their benchmark computer, won it. They stopped upgrading the computer, and it was overtaken pretty quickly afterwards, then they moved the competition goalposts. The leaderboard is a bigger con, IMO. Some guys posted the same score 17 times, and every time he gets an 0.01 improvement, he posts that as well. There should only be one entry per computer, which can be updated as the computer is tweaked/ upgraded. I take an interest in the BTO comp, I think Im in the top ten for fastest single core air-cooled rigs :) Which as a category, doesnt exist, but I dont care :)

Kenny
 
K404 said:
In the early days of "Beat the Office," anyone who beat their benchmark computer, won it. They stopped upgrading the computer, and it was overtaken pretty quickly afterwards, then they moved the competition goalposts. The leaderboard is a bigger con, IMO. Some guys posted the same score 17 times, and every time he gets an 0.01 improvement, he posts that as well. There should only be one entry per computer, which can be updated as the computer is tweaked/ upgraded. I take an interest in the BTO comp, I think Im in the top ten for fastest single core air-cooled rigs :) Which as a category, doesnt exist, but I dont care :)

Kenny

couldnt have put it better myself nice 1...... ;)
 
Fair comment in all areas heer and you get know argument from me - especially as my Winchester score is the highest non FX A64 score there.

loooooooool

Mav
 
You rippin up with a Winchester too huh? What score u gettin? I was getting 1.66 on air. :) Still to try under phase. Dual-cores procs are making themselves felt on the leaderboard.....straight into no.1 spot

:)
 
K404 - yep thats me and too bad is what I say.

One think you have to remember about the BTO leaderboard.

Back in the day when it was a great benchmark (especially for P4 owners as it was the only one where we beat A64 lol) It was very tight getting into the top 5, which I did with me prescott with a score of 1.78, scored similar with Winchester several months later (sold ages ago now) so that 0.1 increase was literally the difference between getting published (which I did) or not (which you didn't).

The issue is not with people spamming - the issue is with the system not deleting the lower scores of the same system.

Believe me if I could delete all but the highest scores for each system which I used to submit marks then I would - but AFAIK I cant.

So get over it!!

loooooooooooooooooooooool
 
LOL... I dont mean it in a "boo-hoo, poor me not getting published" way...i`m unlikely to ever make top 5, I cant afford it. I agree its the way CPC have put the leaderboard together thats at fault, and may main issue is ppl posting the same damn score over and over. Hopefully their new leaderboard will sort it out. Theres still <10 ppl in the official top 10 tho... :)

Kenny
 
Looks like I am just out of the top 10 too, but still that was long time ago so I feel quite proud.

My A64 score of 1.73 is impressive also - althoug Dazboots would beat that if he benched using his Venice, making mine just the fastest winchester based system in the world!!!

No offence taken my friend, but be happier if you edited my name out of your post - ruin the persona of Maverik-SG1, Squadron Leader TEAM-EPUK.

LOL.

Mav
 
Out of respect...... Done.

:D

And....you`re 1.73 aint MILES ahead of my stock air Winchester core, which makes me happy.

Kenny
 
And this point you will understand how much that extra score of just 0.1 more than your last score is so important - Try and beat that score man, just a personal thing for ya.

Be warned, from experience it's binkin difficult.

Oh and to make matters worse the best Winchester score I managed was 1.77 No65 :p
 
1.77? On air, with a Winchester, that wont happen, but i`ll give it a damn good shot under phase :) If the leaderboard gets tidied up with the new benchmark reference, No.65 would suit me fine. Out of all the overclockers in the U.K....to be in the top 100 would delight me. I got small goals :)
 
Cant deny mate thats a great OC on air - but that test is cruel, so bloody cruel and your poor little machine has to work so hard for a measly tenth of a point - If I remember rightly, there was 3 or 4 of us all on a score of 1.74 then 1.76, then 1.77, then I managed 1.78, took about a week LOL.

A lot of A64 users backed away from it as the tests were so biased toward P4's, tehn P4 owners stopped as soon as a few dual xeons came online, now I see P4 dual cores are on there - it would be a right slap in their faces if a dual core A64 system took the crown that Intel suers have held so dear for nearly two years now.

Hehehe

Mav
 
To put those comments into perspective look at the guy in 113 - not a bad spec is it?

Then look at his scores too, thats the type of improvement you need to get an extra 0.1 points.

Decent watercooling would probably get you that.

Mav :)
 
Well...some of the guys (Shigs, Nick,....others) have X2`s...I`d love to see one apear at the old No.1 spot.

See....I think the CPC benches are a good way of REALLY testing an OC. SuperPi is <30secs, CPC is 15 minutes- thats much more like it.

My computers scores vary wildly depending on what kind of mood its in... up to 0.1 difference between runs...no joke.

Kenny

EDIT: LOL...I`m 116 at the mo. His Raptor setup prob helps with the image editing test... i`m using a crappy old Maxtor PATA with 2MB cache. To "only" get that score with an FX55 is selling it short, IMHO.

Still to try CPC benchmarks under phase... just reinstalled Windows, so might see whats happening with my rig at the mo.
 
Back
Top