Welcome to the MacNN Forums.

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

You are here: MacNN Forums > Hardware - Troubleshooting and Discussion > Mac Desktops > Better Choice? a 160GB Drive or 2-80GB drives?

Better Choice? a 160GB Drive or 2-80GB drives?
Thread Tools
JohnM15141
Mac Enthusiast
Join Date: Dec 1999
Status: Offline
Reply With Quote
Apr 20, 2003, 02:26 PM
 
I have a new PowerMac DP1.25 with the basic 80GB Hard Drive on the ATA 100 bus.

I use it to make home movies with iMovie 2 and a Sony TRV10 DV Camera. I save my finished movies on CD. And I render some into quicktime for e-mailing to family.

Usually a 5 minute movie takes up 30GB while I'm working with it. So I want to add more Hard Drive space. I have two configurations I am considering:

I can get a 160 GB hard drive with 8mb Cache and put it on the other ATA 100 channel. This is kind of expensive. Or...

I can get 2 relatively cheap 80GB hard drives and put them both on the ATA 66 channel and configure them as a RAID Array.

Which is the better choice for what I use my PowerMac for?
     
Jazzphone
Forum Regular
Join Date: Sep 2002
Status: Offline
Reply With Quote
Apr 20, 2003, 03:44 PM
 
2 80 Gigs are better because you can back up to 1 hard drive and run regular stuff on the other. If the 1 big hard drive goes down, then, well there goes all your mp3s .
     
CatOne
Mac Elite
Join Date: Nov 2001
Status: Offline
Reply With Quote
Apr 20, 2003, 08:29 PM
 
Originally posted by Jazzphone:
2 80 Gigs are better because you can back up to 1 hard drive and run regular stuff on the other. If the 1 big hard drive goes down, then, well there goes all your mp3s .
Well, if he configures the 2 80's as a RAID array then that's not gonna work, if they're striped Plus, if you put 2 drives as RAID 0 on the ATA66, the bus will surely limit your performance.

Disks are cheap. I'd go with the 160 MB 8 MB cache drive... it is LOTS faster than a 2 MB cache drive.
     
JohnM15141  (op)
Mac Enthusiast
Join Date: Dec 1999
Status: Offline
Reply With Quote
Apr 21, 2003, 11:37 AM
 
The idea to use one 80GB HD for backup hadn't occured to me since most of my critical data is already backed up in the form of DV tapes and I keep my MP 3's and such on CD-rw. I had other intentions when I asked the question.

Here are some more details that I haven't confirmed but I seem to remember reading somewhere.

The Throughput for a 160GB HD with 8MB Cache on the ATA bus is about 35 to 45MB/sec. on the ATA 100 bus(I can't remember the exact numbers.) While the 80GB drives on the ATA 66 bus is around 25 to 33MB/sec.

However, (here is the tricky part that I cannot confirm,) if I configure the two 80GB HD's as a RAID Array (0 striping) I can get double the normal throughput on an ATA 66 bus?

So I guess my real question is which is faster? A 160GB HD w/8MB Cache on the ATA 100 Channel or two 80GB HD in a RAID Array on the ATA 66 bus?
     
CatOne
Mac Elite
Join Date: Nov 2001
Status: Offline
Reply With Quote
Apr 21, 2003, 03:40 PM
 
Originally posted by JohnM15141:
The idea to use one 80GB HD for backup hadn't occured to me since most of my critical data is already backed up in the form of DV tapes and I keep my MP 3's and such on CD-rw. I had other intentions when I asked the question.

Here are some more details that I haven't confirmed but I seem to remember reading somewhere.

The Throughput for a 160GB HD with 8MB Cache on the ATA bus is about 35 to 45MB/sec. on the ATA 100 bus(I can't remember the exact numbers.) While the 80GB drives on the ATA 66 bus is around 25 to 33MB/sec.

However, (here is the tricky part that I cannot confirm,) if I configure the two 80GB HD's as a RAID Array (0 striping) I can get double the normal throughput on an ATA 66 bus?

So I guess my real question is which is faster? A 160GB HD w/8MB Cache on the ATA 100 Channel or two 80GB HD in a RAID Array on the ATA 66 bus?
I'm reasonably sure the 160 GB drive would be faster... Note the ATA66 bus is going to be the limiting factor in the Raid 0 case... they are on the same bus. Not to mention they drives themselves will be significantly slower if they have a 2 MB cache.
     
JohnM15141  (op)
Mac Enthusiast
Join Date: Dec 1999
Status: Offline
Reply With Quote
May 5, 2003, 07:40 PM
 
I just added two WD800J 80GB (W/8mb Cache)drives to my ATA 66 Bus. I set them as a RAID with Striping.

I've got the original 80 GB HD on the ATA 100 channel(Which incidently is a Seagate ST380024A Barracuda ATA V.)

Can anyone suggest a Hard Drive Benchmarking Utility that would allow me to test the performance of this setup?
     
Kenneth
Addicted to MacNN
Join Date: Mar 1999
Location: Bellevue, WA
Status: Offline
Reply With Quote
May 5, 2003, 08:35 PM
 
xBench would do the job.
     
JohnM15141  (op)
Mac Enthusiast
Join Date: Dec 1999
Status: Offline
Reply With Quote
May 5, 2003, 10:26 PM
 
Using XBench the Results are in:

2-80GB RAID on the ATA 66 is faster than 1-80GB HD on the ATA 100!!!

The numbers are:

ATA 100 Disk Test 91.49
ATA 66 Disk Test 112.73

ATA 100 Sequential 104.11
ATA 66 Sequential 103.57

ATA 100 Random 81.59
ATA 66 Random 123.68

Heres the actual results from XBench(Pay attention to the Random Read/Writes on the RAID, very high throughpus are seen!)

Drive Type ST380024A(This is on the ATA 100)
Disk Test 91.49
Sequential 104.11
Uncached Write 96.11 41.93 MB/sec [4K blocks]
Uncached Write 101.34 41.23 MB/sec [256K blocks]
Uncached Read 131.07 20.64 MB/sec [4K blocks]
Uncached Read 95.05 41.01 MB/sec [256K blocks]
Random 81.59
Uncached Write 56.40 0.85 MB/sec [4K blocks]
Uncached Write 88.50 20.30 MB/sec [256K blocks]
Uncached Read 96.10 0.62 MB/sec [4K blocks]
Uncached Read 104.29 20.44 MB/sec [256K blocks]

Drive Type RAID HD(These are on ATA 66)
Disk Test 112.73
Sequential 103.57
Uncached Write 136.20 59.42 MB/sec [4K blocks]
Uncached Write 131.28 53.41 MB/sec [256K blocks]
Uncached Read 94.36 14.86 MB/sec [4K blocks]
Uncached Read 76.54 33.02 MB/sec [256K blocks]
Random 123.68
Uncached Write 185.72 2.80 MB/sec [4K blocks]
Uncached Write 223.85 51.35 MB/sec [256K blocks]
Uncached Read 104.28 0.68 MB/sec [4K blocks]
Uncached Read 77.52 15.19 MB/sec [256K blocks]
----------------------------------------------------------
"He who is tired of Weird Al, is tired of life"
Homer J. Simpson, the 90's
----------------------------------------------------------
     
SSharon
Professional Poster
Join Date: Jan 2003
Location: Teaneck, NJ
Status: Offline
Reply With Quote
May 6, 2003, 09:27 AM
 
just some numbers to compare that to . . . both drives are on the ata-100 both are 2mb cache

Results 123.26
System Info
System Version 10.2.4
Physical RAM 768 MB
Processor PowerPC G4x2 @ 1.00 GHz
Drive Type Maxtor 6Y120L0
Disk Test 125.37
Sequential 129.94
Uncached Write 120.76 52.69 MB/sec [4K blocks]
Uncached Write 133.81 54.44 MB/sec [256K blocks]
Uncached Read 137.41 21.64 MB/sec [4K blocks]
Uncached Read 129.01 55.65 MB/sec [256K blocks]
Random 121.11
Uncached Write 152.55 2.30 MB/sec [4K blocks]
Uncached Write 133.44 30.61 MB/sec [256K blocks]
Uncached Read 95.76 0.62 MB/sec [4K blocks]
Uncached Read 117.16 22.96 MB/sec [256K blocks]

and here is the 80gb drive it came with . . .

Results 113.56
Drive Type IBM-IC35L080AVVA07-0
Disk Test 77.56
Sequential 71.28
Uncached Write 58.24 25.41 MB/sec [4K blocks]
Uncached Write 63.14 25.69 MB/sec [256K blocks]
Uncached Read 163.70 25.78 MB/sec [4K blocks]
Uncached Read 58.82 25.38 MB/sec [256K blocks]
Random 85.05
Uncached Write 91.17 1.38 MB/sec [4K blocks]
Uncached Write 75.41 17.30 MB/sec [256K blocks]
Uncached Read 95.35 0.62 MB/sec [4K blocks]
Uncached Read 81.21 15.92 MB/sec [256K blocks]
     
MDA
Forum Regular
Join Date: Jul 2000
Location: St. Louis Park, MN, USA
Status: Offline
Reply With Quote
May 6, 2003, 09:54 AM
 
Originally posted by SSharon:
just some numbers to compare that to . . . both drives are on the ata-100 both are 2mb cache

Results 123.26
System Info
System Version 10.2.4
Physical RAM 768 MB
Processor PowerPC G4x2 @ 1.00 GHz
Drive Type Maxtor 6Y120L0
Disk Test 125.37
Sequential 129.94
Uncached Write 120.76 52.69 MB/sec [4K blocks]
Uncached Write 133.81 54.44 MB/sec [256K blocks]
Uncached Read 137.41 21.64 MB/sec [4K blocks]
Uncached Read 129.01 55.65 MB/sec [256K blocks]
Random 121.11
Uncached Write 152.55 2.30 MB/sec [4K blocks]
Uncached Write 133.44 30.61 MB/sec [256K blocks]
Uncached Read 95.76 0.62 MB/sec [4K blocks]
Uncached Read 117.16 22.96 MB/sec [256K blocks]

and here is the 80gb drive it came with . . .

Results 113.56
Drive Type IBM-IC35L080AVVA07-0
Disk Test 77.56
Sequential 71.28
Uncached Write 58.24 25.41 MB/sec [4K blocks]
Uncached Write 63.14 25.69 MB/sec [256K blocks]
Uncached Read 163.70 25.78 MB/sec [4K blocks]
Uncached Read 58.82 25.38 MB/sec [256K blocks]
Random 85.05
Uncached Write 91.17 1.38 MB/sec [4K blocks]
Uncached Write 75.41 17.30 MB/sec [256K blocks]
Uncached Read 95.35 0.62 MB/sec [4K blocks]
Uncached Read 81.21 15.92 MB/sec [256K blocks]
Were both drives optimized? Those numbers are terrible for the IBM.

MDA
     
Cipher13
Registered User
Join Date: Apr 2000
Status: Offline
Reply With Quote
May 6, 2003, 10:32 AM
 
2 x 80.
     
Spliffdaddy
Posting Junkie
Join Date: Oct 2001
Location: South of the Mason-Dixon line
Status: Offline
Reply With Quote
May 6, 2003, 11:28 AM
 
2 x 80

gives you more configuration options.

Using a software-based RAID 0 striping scheme you'll get much improved read/write speeds at the cost of higher avg seek times.

How much improvement you'll see depends on the size of files you normally work with.

There is very little difference in speed between drives with a 2MB and a 8MB cache, so don't let that affect your decision. High platter density and rotational speed are your friends - they allow more bytes to be shown to the read heads in the least amount of time. A large cache isn't much benefit if it's waiting for read/write operations. It doesn't significantly improve disk performance, rather, it is intended to lessen the load on the host machine's CPU by providing a place for data to be written very quickly. The host machine has no idea that the data was merely cached, instead of being written to the disk immediately. Heck, you could put a 256MB cache on a hard disk and write operations would FLY...as far as the host machine was concerned. The fact that there is still 17 seconds of activity required to dump the cache onto the disk platter means the hard drive is still just as busy as it would have been without the cache. Read operations would be slowed and any subsequent write operations would be slowed - while the disk plays catch-up with the cache. There is no free ride.
     
dfiler
Grizzled Veteran
Join Date: Feb 2001
Location: Pittsburgh
Status: Offline
Reply With Quote
May 6, 2003, 01:35 PM
 
Originally posted by Spliffdaddy:
There is very little difference in speed between drives with a 2MB and a 8MB cache, so don't let that affect your decision.
Interestingly, users are reporting significant perceived performance improvements from 8MB drives. I'm not sure how the benchmarks stack up but in my experience, the larger cache makes the filesystem much more responsive on most systems. Clicking on a non-cached folder in the finder will produce visual feedback much more quickly when reading from a drive with 8mb of cache.

A search of this forum will turn up a number of threads on the topic of cache size.

Back on topic:
When using both 80gig drives for storage, instead of half for back-up, two drives aren't neccessarily better than one. Two drives decreases the chances of total information loss but increases the chances of partial info loss.
     
Spliffdaddy
Posting Junkie
Join Date: Oct 2001
Location: South of the Mason-Dixon line
Status: Offline
Reply With Quote
May 6, 2003, 01:44 PM
 
I perceive a difference, too, but the benchmark numbers don't seem to justify my perception. Perhaps it's the fact that nearly all hard drives with a big cache are fast drives to begin with. I can say without a doubt that Western Digital's 8MB SE 80GB drive feels pretty much the same as their similar 2MB offering. The benchmark numbers put the 8MB drive slightly ahead of the 2MB, but certainly not enough to notice in actual use.
     
dfiler
Grizzled Veteran
Join Date: Feb 2001
Location: Pittsburgh
Status: Offline
Reply With Quote
May 6, 2003, 02:34 PM
 
Human perception is an interesting beast. A 30 mili-second delay does not feel 3 times as long as a 10 mili-second delay. Some people can discern between 12 and 24 fps film animation while others are completely oblivious to the difference.

This is why I suggest that, if possible, personally test systems running on 2mb and 8mb drives. That is the only true way of telling if the marginal benchmarks disperity adds up to a perceived difference for you personally. For me, it must have reduced lag to just below my threshold. No longer is there an anoying delay between my mouse click and visual feedback in the finder.
     
SSharon
Professional Poster
Join Date: Jan 2003
Location: Teaneck, NJ
Status: Offline
Reply With Quote
May 6, 2003, 06:26 PM
 
In response to MDA, nothing is optimized, I just put in the second drive (the 120gb one) a day ago because I only had 49mb free on the IBM 80gb drive. That's probably another factor in those numbers since the new drive is maybe 15% full while the 80gb drive gave me the wierdest and most non-apple sounding error message ever. It went something like, "Your hard disk has no more space available for application memory. Quit some applications immediately, or your computer will have problems." then beneath that and not in bold it said, "Closing windows and removing files from your hard disk may also help."

So the point is, I need to optimize both drives, and not over stuff them with multi gigabyte eye-tv files, any suggestions?
     
scottiB
Professional Poster
Join Date: Jan 2000
Location: Near Antietam Creek
Status: Offline
Reply With Quote
May 6, 2003, 06:50 PM
 
For more comparison, here's results from 2x120GB Western Digital Jumbo drives striped with a SIIG RAID 100/133 card. Currently the volume is half full and probably fragmented to hell.

Results 131.98
System Info
Xbench Version 1.0
System Version 10.2.5
Physical RAM 1152 MB
Model PowerMac3,5
Processor PowerPC G4x2 @ 800 MHz
Version 7450 (V'ger) v2.1
L1 Cache 32K (instruction), 32K (data)
L2 Cache 256K @ 800 MHz
L3 Cache 2048K @ 200 MHz
Bus Frequency 134 MHz
Video Card GeForce3
Drive Type ATA ACARD Stripe HD0
Disk Test 131.98
Sequential 126.95
Uncached Write 103.56 45.18 MB/sec [4K blocks]
Uncached Write 207.19 84.29 MB/sec [256K blocks]
Uncached Read 99.64 15.69 MB/sec [4K blocks]
Uncached Read 143.09 61.73 MB/sec [256K blocks]
Random 137.43
Uncached Write 203.01 3.07 MB/sec [4K blocks]
Uncached Write 144.84 33.23 MB/sec [256K blocks]
Uncached Read 110.67 0.72 MB/sec [4K blocks]
Uncached Read 121.35 23.78 MB/sec [256K blocks]

-----

SSharon, add more memory. The message pertained to using your IBM drive for swap space (i.e., virtual memory). A 512MB stick would help a lot.
I am stupidest when I try to be funny.
     
SSharon
Professional Poster
Join Date: Jan 2003
Location: Teaneck, NJ
Status: Offline
Reply With Quote
May 6, 2003, 07:54 PM
 
I have 768 physical already in there, thanks for the suggestion though, I will add another 512 chip when I get the cash.
     
JohnM15141  (op)
Mac Enthusiast
Join Date: Dec 1999
Status: Offline
Reply With Quote
May 6, 2003, 08:18 PM
 
I posted my results to show the differences in performance of the drives. But now after I see the XBench scores others are getting I realize my drive performance is dismal!

Maybe I should start a new thread "how to improve drive performance in a powermac?"

I just installed the drives without any special settings or tweaks.
     
scottiB
Professional Poster
Join Date: Jan 2000
Location: Near Antietam Creek
Status: Offline
Reply With Quote
May 7, 2003, 12:59 PM
 
Originally posted by JohnM15141:
I posted my results to show the differences in performance of the drives. But now after I see the XBench scores others are getting I realize my drive performance is dismal!

Maybe I should start a new thread "how to improve drive performance in a powermac?"

I just installed the drives without any special settings or tweaks.
Don't sweat it. You've got good, cheap storage for what you want to do and have more than enough bandwidth. Plus, with your G4, you didn't need to purchase a RAID card to accomplish it (like I had to with my DP-800).

When your done with a project (and have backed-up) reinitialize your RAID from time-to-time to eliminate fragmentation.
I am stupidest when I try to be funny.
     
   
Thread Tools
 
Forum Links
Forum Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Top
Privacy Policy
All times are GMT -4. The time now is 02:58 AM.
All contents of these forums © 1995-2017 MacNN. All rights reserved.
Branding + Design: www.gesamtbild.com
vBulletin v.3.8.8 © 2000-2017, Jelsoft Enterprises Ltd.,