Does 8MB buffer make a difference?

[peter jensen]

Specifically I consider getting the Barracuda 80GB with either 2MB or
8MB - does it really make a difference? If so, for what type of
operations is it really noticable?

 

[John Smith]

Yes - a big difference. I just changed from a WD 120GB with 2MB Cache to a
Seagate 120GB with 8MB Cache and the Seagate is noticeably faster.

It is also much quieter and I think part of that is the fluid motor but also
that it can put more info into Cache. I used to be very aware of the WD
'crunching' when loading apps like Outlook or IE.

J.

 

[Rod Speed]

Specifically I consider getting the Barracuda 80GB with
either 2MB or 8MB - does it really make a difference?

I doubt you'd be able to pick between two otherwise
identical PCs, one with a 2MB cache drive and one
with an 8MB cache, with normal work and not being
allowed to use a benchmark or diagnostic.

If so, for what type of operations is it really noticable?

None in my opinion.

The sort of ops where it should make the biggest difference
is with random ops over the drive platter, rather than very
large contiguous blocks like with an image file etc done
using a single tasking app like ghost etc.

http://www.storagereview.com/articles/200210/20021018WD2000xB_1.html
has a comparison of the 2MB and 8MB cache WD2000 drives.

 

[Rod Speed]

Yes - a big difference. I just changed from a WD 120GB with 2MB Cache
to a Seagate 120GB with 8MB Cache and the Seagate is noticeably faster.

You cant compare them like that, you have to compare
two identical drives from the same manufacturer, one
with the 2MB cache and one with the 8MB cache.

http://www.storagereview.com/articles/200210/20021018WD2000xB_1.html
has a detailled comparison done the right way and there isnt
a lot in it with normal ops most do with desktop systems.

It is also much quieter and I think part of that is the fluid
motor but also that it can put more info into Cache.

Thats got absolutely NOTHING to do with the noise.

I used to be very aware of the WD 'crunching'
when loading apps like Outlook or IE.

Thats just head movements, and quite a bit of that is
between the area the app is being loaded from, the page
file, and the data files used, particularly with Outlook.
Presumably you also mean Outlook Express, not Outlook too.

 

[Arno Wagner]

Specifically I consider getting the Barracuda 80GB with either 2MB or
8MB - does it really make a difference? If so, for what type of
operations is it really noticable?

I see no noticable difference under Linux. However the

=120GB Maxtors only have 3 years warranty if they have 8MB

buffer.

Arno

 

[John Smith]

I wasn't making a direct comparison - I was passing on my personal
experiences. However, they are doubly relevant because the WDs, be they 2MB
or 8MB Cache, are supposed to be faster than the Seagates because the
Seagates concentrate on reducing noise.

As for the fluid drives having nothing to do with noise - what complete
bollox!

J.

 

[Rod Speed]

I wasn't making a direct comparison -
I was passing on my personal experiences.

Which isnt relevant to the question asked, how much difference
there is between IDENTICAL drives with just the cache size different.

However, they are doubly relevant
Nope.

because the WDs, be they 2MB or 8MB Cache, are supposed to be faster
than the Seagates because the Seagates concentrate on reducing noise.

Its MUCH more complicated than that too.

As for the fluid drives having nothing to do with noise

No one ever said anything even remotely resembling anything like that.

- what complete bollox!

In spades with your pathetic comprehension capabilitys.

 

[Arno Wagner]

I just put in a WD 120 with 8 meg. Problem is that I find when I run a
benchmark it shows relatively slow disk performance, with about the same
speed for data that is buffered and data that is not. I don't know what is
going on. This is under XP with a fast AMD processor.

Under Linux you can mount partitions with a 'sync' flag. It means
that the OS has to flush all data to that partition immediately.

Maybe somethng like that is in effect on your system?

Arno

 

[Richard]

I reformated and made sure it designated the drive as "C". Now the various
benchmarks show more normal benchbark values. Strange but true.

Richard.