Advice Please: The Importance of Hard Drive RPMs

Folkert Rienstra wrote:

"David Maynard" wrote in message

Folkert Rienstra wrote:

"David Maynard" wrote in message

CJT wrote:

David Maynard wrote:

CJT wrote:

kony wrote:

On Wed, 18 Aug 2004 21:29:03 GMT, CJT wrote:


Alright, I'll concede there, but a 7200 will still provide a very
noticable performance increase over a 5400. Not just in drive
benchmarks, but in day to day computer usage.

I think the devil will be in the details. If you mostly just browse the
Web, I doubt your disk will be exercised much. If you do a lot of video
editing, you probably want something pretty fast -- most likely RAID.
There's a whole range in between (and perhaps beyond).

Clueless.

Browser caches everything to disk, and reloads it all from this
cache until pages are refreshed unless brower is changed from defaults.


Och, the times that drives were slower than an internet connection, those were the times.

You've mixed replying to someone else inside of a reply you indicate at the
top, by the order of posters listed, is to me.


And that (a comment, not a reply)

And just what is the 'distinguishment' between a 'comment' and a 'reply',
and how is anyone to know?

confuses the hell out of you. Sounds like
purpose fulfilled, when the controversy in the statement fully eludes you.

BTW, what makes you think that comments made to your post is for your
consumption only?

If you want to 'comment' to someone then post to their message and not
inside one of mine, implying I said it.


The only time HDD speed doesn't matter much is when system has
A) Excess system memory to cache files B) Limited multitasking so
files are never flushed from this cache.

[snip]


And that's before we even get to doing a couple of things simultaneously
and/or burning CD/DVDs, playing videos/MP3s, etc.

I have yet to see a user who didn't notice the difference between a 15 gig
5400 RPM drive and a 120 gig 7200 RPM drive. The later is simply faster.

And I have yet to see a user who won't notice the difference between a 15 gig
7200 RPM drive and a 120 gig 5400 RPM drive. The latter is simply faster.

However, the 15GB/7200rpm drive may still load an internetpage (from cache),
existing of a thousand puny files, a bit faster than it's much newer 120GB/5400rpm
competitor.

Well, you've mixed apples and oranges in not only the messages you're replying to,


The apples and oranges are obviously all yours when you compare a 4-year old
dimpled 5400rpm apple with a current fresh 7200rpm orange.

The point of one being faster than the other apparently eluded you.


mixing someone else's in with mine, but in the supposed 'argument'.


Whatever floats your brain, err, boat.


When I said, "it's simply faster" I meant the 'faster' drive *is* faster by
every measurable standard..


And I said that your dimpled apple may still beat your fresh orange in some respect.

If that's what you were trying to say then you'd be wrong.


But I can't tell what you mean by it as you say 'faster' in one case and
then say precisely the opposite in the other, as if one is a 'fake'

I was only overly generalizing the statement in the same way that you overge-
neralized yours, there obviously was a reason for why I picked the same wording.
If you have a problem with that then you have a problem with your own statement.

No, because my statement was true while yours isn't.

"faster" with the other a real but 'irrelevant' "faster."


Yeah, never mind the example that I gave.

I was talking specifically about your 'example', not ignoring it.


Your 'faster' is purely based on sequential reading, i.e. STR.

No, it wasn't.

'Faster' is also having low access time, when reading random small files.

Which a newer 7200 RPM drive will have over an older 5400 RPM drive.


The difference in latency time for a 5400 and 7200rpm drive is 1.4 ms.
With 1000 files (of lets say 5kB each) the new 5400 rpm drive is 1.4 sec
'slower' than your old 7200rpm at same data rate and seektime. At about 0.1 s
to 0.2 s actual data transfer time in about 14 seconds (1000*14ms) of total
load time no speedup of STR will ever make good on that 1.4 second difference.

The 'old' 7200 RPM drive is YOUR 'example', not mine and no one in here
proposed comparing them. That is purely your invention.

You're arguing against yourself.

Let me be clear about it. My comment was that a (significantly) faster
drive, regardless of how it achieves the speed (RPM, linear density [which
I explained previously], etc), is noticed by the user in program load
times, at the least, whether the system 'needs' it to 'keep up' with some
notion of 'average load' or not.


I believe I just proved you wrong. When the excess latency time exceeds
the actual transfer time then there is no way of making good with STR.
There is no such thing as negative time.

I said 'faster'. You want to say I'm wrong when the drive isn't faster,
which is an oxymoron because my criteria is that the drive IS faster,
regardless of how that is achieved.

On Thu, 19 Aug 2004 17:23:07 +0200, "Folkert Rienstra"
put finger to keyboard and composed:


And I have yet to see a user who won't notice the difference between a 15 gig
7200 RPM drive and a 120 gig 5400 RPM drive. The latter is simply faster.

However, the 15GB/7200rpm drive may still load an internetpage (from cache)
existing of a thousand puny files a bit faster than it's much newer 120GB/5400rpm
competitor.

I doubt that there will be any perceptible difference, but show me
such a page and I'll test it out on my old PC.

Anyway, let's say you download a webpage that contains 1MB of content.
At dial-up speeds (5KB/s) this will take about 200 seconds. That's
quite a big page. Yet fetching this page from cache will require only
0.5 sec on my old Cyrix MII-333 PC with an old 1.2GB Maxtor HD. I
would hope that a modern PC would be faster. *Rendering* this page is
another story, however.

Here are the results of two tests in which I transfer 64 and 80 16KB
files (~1MB) from my Seagate 13GB HD to a ramdrive, and the result for
a similar test for my Maxtor 1.2GB HD. It appears that each 16KB read
requires 0.01 sec, so in one second you could perform 100 such
operations. Clearly your HD performance during Internet browsing is
not going to matter one iota in this case. Strangely, the older 4500
RPM Maxtor HD performs slightly better than the 5400 RPM Seagate.

================================================== ==============

C:\WIN98SE\TEMP\test16kread ------------ Seagate ST313021A

start read test
Current time is 9:17:22.97a
Enter new time:
Current time is 9:17:23.63a (elapsed time = 0.66s)
Enter new time: for 64 files
end read test

start read test
Current time is 9:22:00.90a
Enter new time:
Current time is 9:22:01.72a (elapsed time = 0.82s)
Enter new time: for 80 files
end read test


D:\WINDOWS\TEMP\test16kread ------------ Maxtor 71260AP

start read test
Current time is 10:27:03.85a
Enter new time:
Current time is 10:27:04.62a (elapsed time = 0.77s)
Enter new time: for 80 files
end read test

================================================== ==============

C:\WIN98SE\TEMP\testtype 16kread.bat

@echo off
echo.
md g:\test
echo start read test
echo.|time

for %%v in (0 1 2 3 4 5 6 7 8 9 a b c d e f) do copy 16k0%%v.bak
g:\test nul

for %%v in (0 1 2 3 4 5 6 7 8 9 a b c d e f) do copy 16k1%%v.bak
g:\test nul

for %%v in (0 1 2 3 4 5 6 7 8 9 a b c d e f) do copy 16k2%%v.bak
g:\test nul

for %%v in (0 1 2 3 4 5 6 7 8 9 a b c d e f) do copy 16k3%%v.bak
g:\test nul

for %%v in (0 1 2 3 4 5 6 7 8 9 a b c d e f) do copy 16k4%%v.bak
g:\test nul

echo.|time
echo end read test
del g:\test\*.bak
rd g:\test
echo.

================================================== ==============

Directory of C:\WIN98SE\TEMP\test

16K00.BAK 16K01.BAK 16K02.BAK 16K03.BAK 16K04.BAK

................ filenames snipped .....................

16K4B.BAK 16K4C.BAK 16K4D.BAK 16K4E.BAK 16K4F.BAK
80 file(s) 1,310,720 bytes

================================================== ==============


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

"JAD" wrote:

I think you're right. In my experience, upgrading an older machine
with a modern HD makes a much more noticeable improvement in system
response than a CPU upgrade does.


you are absolutely out on 'cloud 9'

Incorrect, oh clueless one.

I wasn't going to respond to this but I'm in a foul mood............if
you are saying that a 5400 rpm HD upgraded to a 7200rpm drive is more
of a UPGRADE than replacing the CPU with a newer faster CPU, your a
lot more clueless than I.


"chrisv" wrote in message
...
"JAD" wrote:

I think you're right. In my experience, upgrading an older machine
with a modern HD makes a much more noticeable improvement in
system
response than a CPU upgrade does.


you are absolutely out on 'cloud 9'

Incorrect, oh clueless one.

"JAD" wrote:

(top posting corrected)

"chrisv" wrote:

I think you're right. In my experience, upgrading an older machine
with a modern HD makes a much more noticeable improvement in
system response than a CPU upgrade does.

you are absolutely out on 'cloud 9'

Incorrect, oh clueless one.

I wasn't going to respond to this but I'm in a foul mood............if
you are saying that a 5400 rpm HD upgraded to a 7200rpm drive is more
of a UPGRADE than replacing the CPU with a newer faster CPU, your a
lot more clueless than I.

Having reading difficulties, JAD? I did not compare "5400 RPM HD vs.
7200RPM HD", it was "old HD vs. modern HD".

Also note that I said that a modern HD gives a more noticeable
improvement in "system response", compared to a CPU upgrade. Note
that "system response" does not mean more FPS in Quake, or whatever
you personally might consider an "upgrade". It means how quickly
common file-based operations (the things that usually make you wait)
occur. It's quite unusual to be waiting on your CPU to crunch-through
something, in normal computer usage...

(bottom posting corrected)

What are you comparing? Using arbitrary words like OLD NEW, doesn't
cut it. AND YOU SAID A HARDDRIVE upgrade will make more of a
performance kick than upgrading the CPU.
\
OK then lets talk what era's we are referring to. 1980's HD Vs 2002?
Why would we be comparing this?
OLD HDs are usually connected to an OLD chipsets, so no matter the HD
upgrade the performance will be gimped. Larger HD's (which are usually
faster RPM wise) after being filled up will bog the file system and
hence slow the system in some barely noticeable way.


"chrisv" wrote in message
...
"JAD" wrote:

(top posting corrected)

"chrisv" wrote:

I think you're right. In my experience, upgrading an older
machine
with a modern HD makes a much more noticeable improvement in
system response than a CPU upgrade does.

you are absolutely out on 'cloud 9'

Incorrect, oh clueless one.

I wasn't going to respond to this but I'm in a foul
mood............if
you are saying that a 5400 rpm HD upgraded to a 7200rpm drive is
more
of a UPGRADE than replacing the CPU with a newer faster CPU, your a
lot more clueless than I.

Having reading difficulties, JAD? I did not compare "5400 RPM HD
vs.
7200RPM HD", it was "old HD vs. modern HD".

Also note that I said that a modern HD gives a more noticeable
improvement in "system response", compared to a CPU upgrade. Note
that "system response" does not mean more FPS in Quake, or whatever
you personally might consider an "upgrade". It means how quickly
common file-based operations (the things that usually make you wait)
occur. It's quite unusual to be waiting on your CPU to
crunch-through
something, in normal computer usage...

On Mon, 23 Aug 2004 09:14:51 -0700, "JAD"
wrote:

I wasn't going to respond to this but I'm in a foul mood............if
you are saying that a 5400 rpm HD upgraded to a 7200rpm drive is more
of a UPGRADE than replacing the CPU with a newer faster CPU, your a
lot more clueless than I.

It depends on the task. For everyday, most common tasks, yes the
hard drive can be a better upgrade. 'Round here though, a lot of
power users will also benefit from the CPU or memory, video card,
etc, but then they may already have that faster hard drive too.

On Mon, 23 Aug 2004 12:20:33 -0700, "JAD"
wrote:

(bottom posting corrected)

What are you comparing? Using arbitrary words like OLD NEW, doesn't
cut it. AND YOU SAID A HARDDRIVE upgrade will make more of a
performance kick than upgrading the CPU.
\
OK then lets talk what era's we are referring to. 1980's HD Vs 2002?
Why would we be comparing this?
OLD HDs are usually connected to an OLD chipsets, so no matter the HD
upgrade the performance will be gimped. Larger HD's (which are usually
faster RPM wise) after being filled up will bog the file system and
hence slow the system in some barely noticeable way.


In a task where hard drive is the bottleneck, even a slight HDD
performance increase will be beneficial... it need not be a
comparison of a circa '80 to a circa '02 drive, but suppose a
more realistic comparision like a 7K2 '04 compared to a 5K4 '03,
which "can" make a large difference in job completion time.

If you're sitting in an office considering only someone who has 2
or 3 apps loaded and box sits like that all day, then in that
specific situation it may not matter as much (or it may, still
depending on the specific tasks).

"kony" wrote in message
On Mon, 23 Aug 2004 12:20:33 -0700, "JAD" wrote:

What are you comparing? Using arbitrary words like OLD NEW, doesn't
cut it. AND YOU SAID A HARDDRIVE upgrade will make more of a
performance kick than upgrading the CPU.
\
OK then lets talk what era's we are referring to. 1980's HD Vs 2002?
Why would we be comparing this?
OLD HDs are usually connected to an OLD chipsets, so no matter the HD
upgrade the performance will be gimped. Larger HD's (which are usually
faster RPM wise) after being filled up will bog the file system and
hence slow the system in some barely noticeable way.


In a task where hard drive is the bottleneck, even a slight HDD
performance increase will be beneficial... it need not be a
comparison of a circa '80 to a circa '02 drive, but suppose a
more realistic comparision like a 7K2 '04 compared to a 5K4 '03,

which "can" make a large difference in job completion time.

"Can", because of the access_time/latency difference, not
the '04 vs '03. STR has hardly changed the past year.


If you're sitting in an office considering only someone who has 2
or 3 apps loaded and box sits like that all day, then in that
specific situation it may not matter as much (or it may, still
depending on the specific tasks).

Right, like using an OS that does parallel IO vs an OS that serializes IO.

agreed,
in the big picture, what the OP was asking about, after educating
himself on the fundamentals of harddrive configuration, was out on the
fringe (which place on the HD will XP load faster)!!. I can understand
the performance gain of a buffered 7200/10000 rpm drive these days,
over a 5400 of latter year, scratch disks used by my software have
greatly improved. However, the statement was an OLD HD refitted with a
newER HD would be a BIGger performance gain than a cpu upgrade.
Granted he never said from what CPU to what CPU, but using the common
sense factor(why I bother with that, I don't know-uncommon sense
factor?), you wouldn't just update from a 2.2 to a 2.3, so I would
figure a significant upgrade. This would far better 'enhance the
system' than a HD replacement of just about any magnitude I could
think of. That is, stay on planet earth and compare by what can be
'noticed' without a BM running all day. I realize that there are far
'deeper' stats that apply, and that these are of importance, but I
prefer to stay in the everyday overall experience, and what I feel
while I'm using the appliance.


"kony" wrote in message
...
On Mon, 23 Aug 2004 09:14:51 -0700, "JAD"
wrote:

I wasn't going to respond to this but I'm in a foul
mood............if
you are saying that a 5400 rpm HD upgraded to a 7200rpm drive is
more
of a UPGRADE than replacing the CPU with a newer faster CPU, your a
lot more clueless than I.

It depends on the task. For everyday, most common tasks, yes the
hard drive can be a better upgrade. 'Round here though, a lot of
power users will also benefit from the CPU or memory, video card,
etc, but then they may already have that faster hard drive too.