Discussion Locked

Seeking hard drive speed opinions

0 helpful
18 replies
Closed to new repliesThis discussion is closed. It stays readable, but it takes no new replies.

What are your qualified opinions/observations in regards to the functional difference between these two "generations" of hard drives?

I realize that there are differences among various manufacturers, particular drive specifications, and the like. And I further realize that there is a "theoretical" difference of whatever percent (25%), but this is a "generally speaking" question.

The first "generation" would be those drives that are several years old, most commonly 5400 rpm, 10ms/11ms average seek time, 4 read/write heads, etc. A 6.5 GB Fijitsu Model MPC3064AT drive would fit into this category.

Here is a link to the specification page:

http://tinyurl.com/49ge

The second "generation" would be those most commonly sold the past one or two years, most commonly 7200 RMP, 8ms average seek time, 6 Read/Write heads, etc. The 20/40/60 GB IBM Deskstar drives would fit into this category.

A link to the specification page:

http://tinyurl.com/49gj

What I mean by "functional difference" is this: Would a normal user detect a difference in the speed of their computer running normal Windows 2000 functions, normal applications (MS Office), etc.?

18 replies

by dustyd

I believe a "normal user" today would definitely notice the difference in performance with only the change in drives you mentioned, especially if they use more apps than just email and IE. (Those apps that actually do a lot of read/writes to the disk.)
When you add in the performance gains from systems using ATA100/133 vs. ATA33/66, and those from faster memory (plus more of it), then the gains are going to be glaringly obvious.

by maxwell edison

Thanks everyone for taking the time to give your thoughts and opinions. Interesting results, don’t you think? If I counted the opinions correctly, out of six answers, four think that there would be no difference, while two believe that there could be a noticeable difference.

What led me to ask the question in the first place is this: While upgrading all my users from Windows 9x to Windows 2000 Pro, I would sometimes use a different drive in the same computer. I wanted to do a clean install, starting from scratch on all computers, and after the OS upgrade I would temporarily install the user’s old drive as a slave to easily copy any data, import old Outlook Express mail/address book, and the like. (Yes, I keep all company data on the file servers, but some users just don’t understand that I don’t backup any workstations – but that’s another whole issue - and Outlook Express can’t be saved on a server. But I digress.) Anyway, I replaced one of the IBM Deskstar drives with one of the Fijitsu, as described. Both the user and I noticed a significantly slower computer, especially accessing local files, opening folders, etc. I did his computer over again, but this time using his original drive, and the slow down disappeared. Everything was back to normal.

So I’ve concluded one of three things:
1. There is a noticeable difference.
2. Anna is correct in thinking that perhaps the drive is on its way to failure. (And it is about four years old.)
3. Something else that I missed entirely.

Thanks again for the input. Feel free to email me if you have any other opinions and/or are interested in discussing it any further.

Maxwell

by ann777

A normal user.... using the functionally same computers (by this I mean the same ram, mhz cpu, disk controllers, etc.... and same software).... I personally do not believe that they would be able to tell the difference of operation from just the rpms of hdd (...unless there is a problem with the drive, and the drive in reality is not operating to spec)

by ann777

Just a comment with Answerer #4's example.

With moving the swap file to a second drive... there'd be improvement in performance whether or not the drive was the same speed. The fact is that you are not using the "same computer" when you add a drive; two drives performing different tasks will increase performance simply because the first is not having to do the back and forth thing with doing all the tasks. How can you "prove" the performance increase was not due to separating functions instead of the drives rpms?

by maxwell edison

Thanks everyone for taking the time to give your thoughts and opinions. Interesting results, don’t you think? If I counted the opinions correctly, out of six answers, four think that there would be no difference, while two believe that there could be a noticeable difference.

What led me to ask the question in the first place is this: While upgrading all my users from Windows 9x to Windows 2000 Pro, I would sometimes use a different drive in the same computer. I wanted to do a clean install, starting from scratch on all computers, and after the OS upgrade I would temporarily install the user’s old drive as a slave to easily copy any data, import old Outlook Express mail/address book, and the like. (Yes, I keep all company data on the file servers, but some users just don’t understand that I don’t backup any workstations – but that’s another whole issue - and Outlook Express can’t be saved on a server. But I digress.) Anyway, I replaced one of the IBM Deskstar drives with one of the Fijitsu, as described. Both the user and I noticed a significantly slower computer, especially accessing local files, opening folders, etc. I did his computer over again, but this time using his original drive, and the slow down disappeared. Everything was back to normal.

So I’ve concluded one of three things:
1. There is a noticeable difference.
2. Anna is correct in thinking that perhaps the drive is on its way to failure. (And it is about four years old.)
3. Something else that I missed entirely.

Thanks again for the input. Feel free to email me if you have any other opinions and/or are interested in discussing it any further.

Maxwell

by dmiles

Besides the functional difference of the hard drives,you would also have to figure in the CPU and the Bus speed to transfer the data
So,a practical user may not notice the difference in data transfer depending on their system setup.
A PII with a 7200 rpm hard drive may not notice any difference in their speed compared to a user that may be using a PIII that is setup with a Controller card to connect the 7200 rpm hard drive
Then I would not think that they notice an difference in the drive speed

by maxwell edison

Thanks everyone for taking the time to give your thoughts and opinions. Interesting results, don’t you think? If I counted the opinions correctly, out of six answers, four think that there would be no difference, while two believe that there could be a noticeable difference.

What led me to ask the question in the first place is this: While upgrading all my users from Windows 9x to Windows 2000 Pro, I would sometimes use a different drive in the same computer. I wanted to do a clean install, starting from scratch on all computers, and after the OS upgrade I would temporarily install the user’s old drive as a slave to easily copy any data, import old Outlook Express mail/address book, and the like. (Yes, I keep all company data on the file servers, but some users just don’t understand that I don’t backup any workstations – but that’s another whole issue - and Outlook Express can’t be saved on a server. But I digress.) Anyway, I replaced one of the IBM Deskstar drives with one of the Fijitsu, as described. Both the user and I noticed a significantly slower computer, especially accessing local files, opening folders, etc. I did his computer over again, but this time using his original drive, and the slow down disappeared. Everything was back to normal.

So I’ve concluded one of three things:
1. There is a noticeable difference.
2. Anna is correct in thinking that perhaps the drive is on its way to failure. (And it is about four years old.)
3. Something else that I missed entirely.

Thanks again for the input. Feel free to email me if you have any other opinions and/or are interested in discussing it any further.

Maxwell

by joseph moore

My wife has been my test subject on several computing items over the years. When I need a "normal user" view on something -- a document I wrote, or a procedure on a workstation that I worked up -- I run it by her. She has no patience with me when I try and explain IT stuff (just like a normal user); she has no desire to get any understanding of how systems work other than you move the mouse and click things and they go (just like a normal user); she expects it to work easily and without problems or slowdowns all the time (just like a normal user).
So, in my opinion, I think a normal user WOULD notice a difference between the two drives you mentioned, and here is why:
Our old computer (that I just decommissioned late last year) was a 100 MHZ Pentium with 64 MG RAM, running Win95. Yep, a real barn-burner! This drive had a Seagate 5400 RPM IDE drive.
Last Christmas, I bought for myself as a Christmas present a Maxtor 7200RPM drive for extra storage. On Christmas day 2001, I opened thedrive, and installed in in the old computer. I ran it on its own IDE channel (slaving the CD-ROM off of it). I then brought up Win95 and set up the drive as E: drive.
AFter that, I moved the swapfile onto it, leaving no swapfile on drive C: (the old 5400RPM drive).
Reboot.

I remember later on (probably a week later) that my wife noticed a speed improvement with the new drive being installed. Apps launched a little quicker (I didn't move apps to this drive; it was only for drive space, the swap file, and the IE temp cache). She was happy with this experiment (she was initially dubious; "Don't break my computer like you do at work!" she jokingly told me a few times!)

by joseph moore

Now, I have never tried replacing a hard drive in an existing system with one that was faster on the RPMs without chaning out other hardware. This example with my old home system is the closest thing I have done. But based off of it, I say yes, your users will notice something being different. Not a lot, but noticeable.

And BTW, I now have a 566MHZ Celeron at home, running Win2KPro!

Good luck

by maxwell edison

Thanks everyone for taking the time to give your thoughts and opinions. Interesting results, don’t you think? If I counted the opinions correctly, out of six answers, four think that there would be no difference, while two believe that there could be a noticeable difference.

What led me to ask the question in the first place is this: While upgrading all my users from Windows 9x to Windows 2000 Pro, I would sometimes use a different drive in the same computer. I wanted to do a clean install, starting from scratch on all computers, and after the OS upgrade I would temporarily install the user’s old drive as a slave to easily copy any data, import old Outlook Express mail/address book, and the like. (Yes, I keep all company data on the file servers, but some users just don’t understand that I don’t backup any workstations – but that’s another whole issue - and Outlook Express can’t be saved on a server. But I digress.) Anyway, I replaced one of the IBM Deskstar drives with one of the Fijitsu, as described. Both the user and I noticed a significantly slower computer, especially accessing local files, opening folders, etc. I did his computer over again, but this time using his original drive, and the slow down disappeared. Everything was back to normal.

So I’ve concluded one of three things:
1. There is a noticeable difference.
2. Anna is correct in thinking that perhaps the drive is on its way to failure. (And it is about four years old.)
3. Something else that I missed entirely.

Thanks again for the input. Feel free to email me if you have any other opinions and/or are interested in discussing it any further.

Maxwell

by thechas

In my opinion, you are comparing apples and oranges.

The 'normal' user will see a much larger performance difference between these 2 drives than the difference in the read/write and rotational rates, assuming that the system supports ATA100.

The ATA 100 transfer rate coupled with the larger drive cache gives the newer drives roughly 3X the performance of the older drive.

I was ammazed at the performance improvement when I switched my ATA100 drives from an UDMA66 controller to an ATA100controller.
That is part of why I am so quick to recommend that users install new IDE controllers when their motherboards do not support their new large hard drive.

Now, if you install the new drive in a system that only supports UDMA33, the user will see little or no change in performance.
Without the faster data transfer rate, the drive will end up waiting for the buss to catch up.

If the systems support even UDMA66, the users will notice the difference in nearly all applications thataccess the drive.

Chas

by maxwell edison

Thanks everyone for taking the time to give your thoughts and opinions. Interesting results, don’t you think? If I counted the opinions correctly, out of six answers, four think that there would be no difference, while two believe that there could be a noticeable difference.

What led me to ask the question in the first place is this: While upgrading all my users from Windows 9x to Windows 2000 Pro, I would sometimes use a different drive in the same computer. I wanted to do a clean install, starting from scratch on all computers, and after the OS upgrade I would temporarily install the user’s old drive as a slave to easily copy any data, import old Outlook Express mail/address book, and the like. (Yes, I keep all company data on the file servers, but some users just don’t understand that I don’t backup any workstations – but that’s another whole issue - and Outlook Express can’t be saved on a server. But I digress.) Anyway, I replaced one of the IBM Deskstar drives with one of the Fijitsu, as described. Both the user and I noticed a significantly slower computer, especially accessing local files, opening folders, etc. I did his computer over again, but this time using his original drive, and the slow down disappeared. Everything was back to normal.

So I’ve concluded one of three things:
1. There is a noticeable difference.
2. Anna is correct in thinking that perhaps the drive is on its way to failure. (And it is about four years old.)
3. Something else that I missed entirely.

Thanks again for the input. Feel free to email me if you have any other opinions and/or are interested in discussing it any further.

Maxwell

by cactus pete

If all otehr things are equal, and the bottleneck of each system is actually the hard drive [given normal user tasks], then no, I don't believe that users can tell the difference. But I'm defining "normal user" as someone who uses pretty much on app at a time, with few files opening and closing at the same time [if any].

Pretty much - if there is ample processor and memory and the user is not running DISK INTENSIVE applications, there is no reason to bother with faster drives.

by cactus pete

Man... my typing was REALLY bad. My apologies for such poor spelling.

"if all other things..."
"pretty much one app"

by maxwell edison

Thanks everyone for taking the time to give your thoughts and opinions. Interesting results, don’t you think? If I counted the opinions correctly, out of six answers, four think that there would be no difference, while two believe that there could be a noticeable difference.

What led me to ask the question in the first place is this: While upgrading all my users from Windows 9x to Windows 2000 Pro, I would sometimes use a different drive in the same computer. I wanted to do a clean install, starting from scratch on all computers, and after the OS upgrade I would temporarily install the user’s old drive as a slave to easily copy any data, import old Outlook Express mail/address book, and the like. (Yes, I keep all company data on the file servers, but some users just don’t understand that I don’t backup any workstations – but that’s another whole issue - and Outlook Express can’t be saved on a server. But I digress.) Anyway, I replaced one of the IBM Deskstar drives with one of the Fijitsu, as described. Both the user and I noticed a significantly slower computer, especially accessing local files, opening folders, etc. I did his computer over again, but this time using his original drive, and the slow down disappeared. Everything was back to normal.

So I’ve concluded one of three things:
1. There is a noticeable difference.
2. Anna is correct in thinking that perhaps the drive is on its way to failure. (And it is about four years old.)
3. Something else that I missed entirely.

Thanks again for the input. Feel free to email me if you have any other opinions and/or are interested in discussing it any further.

Maxwell

by opatzg9

Unless your normal user includes gaming/graphic work/cad on computers, I'ld say no they will not
notice any difference. But there are other ways to get speed and maintain speed in a computer. And even the fastest hard drive when filled to near capacity is no faster than a slow hard drive with plenty of room. The faster speeds will be realized simply by having more space available but also if they have that extra space they really shouldn't be trying to run on a skimpy amount of ram. If they are gaming then speed and graphics, cdrom/dvd multpipliers or virtual cdrom software make a huge difference in performance. So it really depend on your definition of normal user. For things that acomplish work or homework for kids I'ld have to say no just enough space to keep your drive always at under 50% capacity works and about the same. Otherwise how do you compare an 80 gig meg buffered drive with 7200rpm to a 20gig 1meg buffered 5400. the 20 gig was fast when it had 5 gig used but started choking around 18 gig full, the 80 gig is still under 50% and running great. But I think that is due to available space not spindle speeds. I just saw no reason to buy another 20 gig when I could swap to an 80 for $20 more.

This discussion is closed to new replies.