Jump to content
Sign in to follow this  
Angry_Games

hard drives & raid - benchmark and compare!

Recommended Posts

Spread Spectrum

 

What is Spread Spectrum and what does it do?

 

Here's the lowdown.

 

If the drive and controller both support Spread Spectrum operation, it allows the system to modulate the frequencies that the devices communicate on.

 

This is done to allow proper operation in the event of a conflict with other devices operating on the same frequency.

 

That said, as best I can tell, it's rarely used but is part of the current PC specs to comply with FCC requirements.

 

In my testing, Spread Spectrum has a detrimental effect on overclocking and presents no obvious advantage during operation so therefore it should be avoided.

Share this post


Link to post
Share on other sites
Can anyone answer my question from about 1/2 page up?

 

What is better my increased read speeds on test ~1-5 with read caching on

 

or my burst rate of almost 400 with read caching off?

 

my pics are up there too :D

 

what does read caching do? ive looked it up a bunch and havent found tons of info on what it does.

 

Thanks again folks :D

 

 

I am assuming I am reading you correctly- the HD cash stores information read from the disk often in anticipation to furute needs. it speeds up the overall speed of the HD especaly in larger files. The benifits are not going to show up in these two benchmarking programs.

Share this post


Link to post
Share on other sites
Spread Spectrum

 

What is Spread Spectrum and what does it do?

 

Here's the lowdown.

 

If the drive and controller both support Spread Spectrum operation, it allows the system to modulate the frequencies that the devices communicate on.

 

This is done to allow proper operation in the event of a conflict with other devices operating on the same frequency.

 

That said, as best I can tell, it's rarely used but is part of the current PC specs to comply with FCC requirements.

 

In my testing, Spread Spectrum has a detrimental effect on overclocking and presents no obvious advantage during operation so therefore it should be avoided.

Recently I was thinking about this and was serioulsly having doubts about Spread Spectrum on my own... I think you totally confirmed this and of course its time to disable it and see what happens.

Share this post


Link to post
Share on other sites
Recently I was thinking about this and was serioulsly having doubts about Spread Spectrum on my own... I think you totally confirmed this and of course its time to disable it and see what happens.

I ran tests with SS on and off last year and didn't detect any difference.

Share this post


Link to post
Share on other sites

i just build my array with 16k stripe and 4 k cluster, to keep the 4:1 ratio of ExRoadie,

 

i got 2 seagate SATA II driv, seagte use a jumper to select 1.5 or 3.0gb/s mode, i remove the jumper to activate the 1.5gb/s mode like it was writen on the disk label,

 

after a nice and fresh install of windows with sp2, and driver package 6.70 (without IDE SW driver causing rig to crash)

 

with ncq enabled i end with something like this

 

http://www.dfi-street.com/forum/attachment...18&d=1153454116

 

and with ncq disabled i end with this

 

http://www.dfi-street.com/forum/attachment...19&d=1153454124

 

any ideas???

 

EDIT : i put the jumper back on the return to SATA 1.5gb/s mode, same result from HD TACH, i have also install IDE SW driver, after i had remove my CD-RW which doesn't like the ide SW driver

Share this post


Link to post
Share on other sites
I ran tests with SS on and off last year and didn't detect any difference.

Not even Burst Speed in HD Tach? I tested the whole array with and with out SS enabled ages ago and it showed a significant diference, however what is Burst Speed anyway :rolleyes:

Share this post


Link to post
Share on other sites
so noone still cares to answer my question?(roughly a week later) lol i see everyone else just skipping right over my comments.... Im not that mean :(

 

http://www.dfi-street.com/forum/showpost.p...&postcount=1284

 

I like mine with Command Queuing disabled and Read Cache enabled. That's what gives me the best real life performance, even if benchmarks with HD Tach show decrease in burst MB/s with Read Cache enabled.

 

I'm thinking the burst speed is effected "negativly" cause it has to go through the cache before being delivered to the program, but instead the speed from the cache to the program should be greater (in real life performance). But of course there is a overhead so the performance may differ depending on the task...

 

Don't know how HD Tach does it's calculation *shrug*

Share this post


Link to post
Share on other sites

HD tach 8mb, 32mb and ATTO

 

hdbenchescombinedlr9.th.jpg

 

Sorry for the thumby but I couldn't resize and maintain legibility

 

What's in my siggy X4 in RAID 0...Default Stripe/Cluster. These YD's are made for RAID...

Share this post


Link to post
Share on other sites

Oh you did buy the raid drives edition, with the Time-Limited Error Recovery (TLER). How do they work for you in the nF4 controller?

 

I was told these a specifically for a hardware controller that does it's own error checking so the drives own firmware doesn't interfere with the whole raid arrays rythm.

 

The speed looks good. With the default settings this array would be 64kb stripe and 4kb cluster, right?

 

Command Queing, Read Cache and Write Cache, disabled or enabled? all default?

Share this post


Link to post
Share on other sites
Oh you did buy the raid drives edition, with the Time-Limited Error Recovery (TLER). How do they work for you in the nF4 controller?

 

I was told these a specifically for a hardware controller that does it's own error checking so the drives own firmware doesn't interfere with the whole raid arrays rythm.

 

The speed looks good. With the default settings this array would be 64kb stripe and 4kb cluster, right?

 

Command Queing, Read Cache and Write Cache, disabled or enabled? all default?

 

All default. Haven't touched a thing...

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×