www.xlr8yourmac.com

Reviews and Daily News with a  Difference!
Select a Topic Area To See a List of Related Reviews & Articles
Systems  | CPU Upgrades  | SCSI | IDE | Firewire  | Video  | Audio  | Games  | Misc/OS  | Archives  | Search
Performance Tests: OWC Elite Firewire CaseReturn to News Page

Click for SMOKING FAST 6G SSDs!
Click for SMOKING FAST 6G SSDs!

OWC Elite case kit
Performance Tests: OWC's Elite Firewire Case
Real World Tests & Benchmark Comparisons
(w/G4 500 DP, PB G4/400, PB G3/500, 9600/350)

By Mike
Published: 4/27/2001

Note on DV Capture to Firewire Drives on early PowerBooks G4: If you're concerned about capturing Digital Video to a Firewire drive connected to a PowerBook (which as you can see from the tests below, Powerbooks have lower Firewire performance than desktop models), see my comments below. Tests with even the VST portable drive worked fine on my PB G4/400. (Digital Camcorder daisy-chained to the VST drive, using iMovie 2 and capturing DV to the VST drive, even while IE 5 was running in the background with an Airport internet connection.) Digital Origin (makers of EditDV, MotoDV, etc.) notes a drive with an avg. sustained rate of 5MB/sec or faster is required, so most Firewire drives should be fine. I later captured video from Final Cut Pro 1.2.5 with no problems also but I don't have FCP 2.0 to test. [Update - See Michael Kramer's comments on FCP 2.0/Quicktime 5.01 batch capture issues even with Desktop G4s in the 6/8/2001 news page. His experience indicated that QT 5.01 had issues he did not see in QT 4.x, even with desktop G4s.]
(The internal 2.5" IDE drives of PowerBooks also have lower rates than desktop G4 systems with 3.5" IDE hard drives, but still adequate for most anything you need to do. Benchmarks show a much higher delta of performance than real world tests in every system/drive. Pure benchmarks like ATTO's rates are much higher than the real world transfer rates as shown below.)

Cases/Drives Tested:

  • Other World Computing "Elite" (Oxford 911 IDE/Firewire bridge) Case kit (60GB IBM 75GXP 7200 RPM/2MB cache IDE drive installed)
  • Clubmac Firewire case (early 2000 version w/ATA3 bridge) Case kit (w/same IBM 60GB drive installed)
  • VST Portable 12GB drive (bus powered)
    (using latest VST 2.3 driver and v56 firmware)

Both the OWC Elite case kit and the Clubmac case kit tests used the same (brand new) IBM 75GXP 60GB drive (ATA/100, 7200 RPM with 2MB cache - just over 57GB formatted). I've taken photos of the OWC case kit for posting on a separate page on Monday. The Clubmac case (with LSI ATA/3 rated bridge board) details are shown in the first Firewire case kit article here from early in 2000. To see how both of these 3.5" drive case kits compared to a portable firewire drive, I included results of tests with my VST 12GB with each system. Remember each drive can have some variation, especially with drives like the VST where they may use different brands of drives inside the case. I saw this in past tests of two different 12GB VST drives for instance. Even drives of the same brand/model can have some performance variation. For tests that used a stopwatch, there's some margin for error in starting/stopping it, so consider times within 1/2 second or so identical.

3 drive photo

Intro: The purpose of this article is to show what actual performance advantage the latest fast IDE/Firewire bridge based Firewire cases provide over the early models and compared to a small, truly portable firewire drive such as the VST ultraslim I purchased last year. The AC powered firewire drives have the advantage of 2x to 3x the storage for the same price (typically); but as shown here, real-world performance differences are often far from what the pure benchmarks would lead you to believe. (Take a look at the 102MB file copy performance for instance, where effective rates are a fraction of ATTO's benchmark rates.) If you're running ATTO's benchmark on the OWC Elite case, on Powerbooks you'll see much lower rates than on a G4 desktop. The Oxford911 bridge cases are just the first ones able to show the higher benchmark rates; previous firewire cases used slower bridges that masked this. (See results below.)

Test Notes: Results are organized by Mac model, since performance varies with each. Powerbooks had lower performance than the G4 desktop, in pure benchmarks much lower in many cases. All PowerBook tests were run on AC power (not battery). Only the firewire drive being tested was connected to the firewire ports during the tests. All tests run after a clean reboot. All Firewire disks were erased before tests. (Since some may not read every system's test results, info on each test is noted in every Mac test system section before the results graph.)

Real world test results are shown first since they matter most. Pure benchmarks like ATTO (which I've used since 1997 in articles here) show much higher rates than typically occur in normal use. In any tests that use the internal system's hard drive (file copies to/from it), remember the internal drive can be a factor depending on its speed. However even when copying files to/from the internal drives, there was some performance difference in the 3 firewire drives. The main point was to show how each of the 3 drives performed when run from the exact same system, since copying files from/to the firewire drive is something everyone would be doing in actual use. As you will see from the results, often the differences in real world tests between the drives are much smaller than what you'd expect if you only looked at the pure benchmark results. (This fact is also shown in many past drive articles here. The same is true of any drive - IDE, Firewire or SCSI.)

For the 9600/350, I also tested MP3 playback from the Firewire drives to see if there was any stuttering or dropouts. DV streams were also smooth in iMovie2 played from the Firewire drives in that system. (I plan on doing capture tests from my DV camcorder within the next week.)

Boot Tests: Boot tests were run on all systems with each drive. The 9600/350 as expected would not boot from any of the firewire drives. On the PB G4, PB G3 firewire and Dual G4/500, all firewire drives were bootable by pressing the option key at startup. Selecting the drive in the Startup Disk control panel didn't work to boot from them, even during a warm restart.

System Details: Configurations of each system used is listed in that model's test results section.

FYI: The first thing I noticed when opening the OWC Elite case was that the (short) IDE cable is not an ATA/66 one. ATA/66 cables have additional shielding. I immediately ran tests on two systems with both the supplied cable and an ATA/66 cable and saw no differences in the performance with either. For those wondering if the lack of an ATA/66 cable (shielding) could affect data on the firewire drive, I ran tests using Tim Seufer's data corruption tester listed in the FAQ Apple G3 section here. Tests with a 2GB test file showed no errors.


Systems Tests Results
Dual G4/500 | PowerBook G4/400 | PowerBook G3/500 | 9600/350

Dual G4/500 Tests:
Gigabit Eithernet Dual G4/500, OS 9.1 with Firewire 2.7 extensions. (G4 Firmware update 4.1.8 had been applied.) Boot disk is an IBM 75GXP 45GB drive. VST's current v2.3 extension was used for that drive, other drives used only the Apple Firewire 2.7 extensions (which mounted the drives fine without any added software). As a check of the IBM 60GB drive's maximum potential, I included ATTO benchmark results with the drive connected to the onboard ATA/66 IDE bus. The G4 desktop had much better performance (especially in benchmarks like ATTO's) with the fast OWC Elite drive than did the Powerbooks or 9600/350.

Real World Tests:

Copy 102.3MB File to FW Drive:
Time to copy a single 102.3MB file to the Firewire drive.

Copy 102MB file -  G4/500 DP

Notice the effective MB/sec rate in real world tests like this are far lower than the pure benchmark results such as ATTO's tools. (Even the fastest drive had less than 10MB/sec effective rates.)
BTW - Copying this file from the onboard IDE (45GB 75GXP slave drive/boot disk) to the OEM Maxtor 20GB drive took 7.4 seconds [net rate of appx. 13.82 MB/sec]. (These disks were not blank of course, about 1/3 and 1/2 full respectively.)

Read 102.3MB File From FW Drive:
Time to copy a single 102.3MB file from the Firewire drive to the G4's internal IDE drive (IBM 75GXP, 7200RPM/2MB cache).

Read 102MB file -  G4/500 DP

Unstuff System Folder:
Time to unstuff a 102.3MB file (327.3MB/3,342 files unstuffed) from the Firewire drive. (Intensive Read/Write test with the drive almost constantly being read/written. Granted CPU speed helps on unstuffing files, but there was some difference in performance with the 3 drives.)

Unstuff SIT file -  G4/500 DP

BTW - The onboard IDE 75GXP 45GB drive took 6 Min, 30 sec to unstuff this file.


Benchmark Tests: For the G4/500 DP benchmarks, I've included tests of the 60GB IDE drive connected to the G4's onboard ATA/66 IDE bus for reference. (I then moved the drive to the OWC and Clubmac firewire cases for their tests.)

ATTO Tools 2.5 Benchmark:
Rates in MegaBytes per Second (longer bars are faster).

ATTO Benchmark - G4/500 DP

Note: The "60GB IDE" results in the graph above is the same drive used in the OWC and Clubmac case, but connected to the onboard G4's ATA/66 IDE bus. I did this to get an idea of how the drive performed using native IDE.

VST Benchmark:
The VST Format utility benchmark requires the VST extension to be enabled. This is the only test ran where the VST extension was used with the other drives. In the graph below, longer bars are faster for all tests *except* the Access results (in milliseconds), where lower scores are better. Other rates are in MegaBytes/second.

VST Benchmarks - G4/500 DP

Helio LanTest 2.5.2:
Although written to test server performance, I used Helios LanTest 2.5.2 to exercise each of the drives. It runs each test 10 times and reports the average for the 10 tests. For the Read/Write tests (in KB/sec), higher scores are better. For the other tests (opening/closing/deleting/etc.) the times are in seconds (or fractions), so lower scores are better.

LanTest Read/Write -  G4/500 DP

LanTest Other Tests -  G4/500 DP


PowerBook G4/400 Tests:
My PB G4/400 has 512MB of RAM, OS 9.1 (OS X on 2nd partition), IBM 32GB Travelstar (5400 rpm/2MB cache). PB G4 Firmware update 4.1.8 had been applied.

Real World Tests:

Copy 102.3MB File to FW Drive:
Time to copy a single 102.3MB file to the Firewire drive.

Copy 102MB file - PB G4/400

OS X Copy Test: I ran this copy test with the OWC Elite 60GB drive under OS X on the PB G4/400. The 102.3MB file copied in 10.5 seconds [effective net copy rate of appx. 9.74 MB/sec]. That was significantly better than the rates seen under OS 9.1 on that machine. Not an extensive test of course, but interesting.

Read 102.3MB File From FW Drive:
Time to copy a single 102.3MB file from the Firewire drive to the PB G4's internal drive (a 32GB/5400 RPM, 2MB cache IBM Travelstar).

Read 102MB file - PB G4/400

Unstuff System Folder:
Time to unstuff a 102.3MB file (327.3MB/3,342 files unstuffed) from the Firewire drive. (Intensive Read/Write test with the drive almost constantly being read/written. Granted CPU speed helps on unstuffing files, but there was some difference in performance with the 3 drives.)

Unstuff SIT file - PB G4/400


Benchmark Tests:

ATTO Tools 2.5 Benchmark:
Rates in MegaBytes per Second (longer bars are faster).

ATTO Benchmark - PB G4/400

VST Benchmark:
The VST Format utility benchmark requires the VST extension to be enabled. This is the only test ran where the VST extension was used with the other drives. In the graph below, longer bars are faster for all tests *except* the Access results (in milliseconds), where lower scores are better. Other rates are in MegaBytes/second.

VST Benchmarks - PB G4/400

Helio LanTest 2.5.2:
Although written to test server performance, I used Helios LanTest 2.5.2 to exercise each of the drives. It runs each test 10 times and reports the average for the 10 tests. For the Read/Write tests (in KB/sec), higher scores are better. For the other tests (opening/closing/deleting/etc.) the times are in seconds (or fractions), so lower scores are better.

LanTest Read/Write - PB G4/400

LanTest Other Tests - PB G4/400


PowerBook G3/500 Tests:
The test system was an early model PowerBook G3/500 firewire (summer 2000 models had a different firewire chip according to some sources) with 256MB of RAM. OS 9.04, but using OS 9.1's firewire 2.7 extensions. Internal drive is a 20GB IBM travelstar (4200 RPM/2MB cache). Later PB G3 Firewire models with the different Firewire controller may perform better at least on reads (OWC's tests with a PB G3/500 running OS 9.1 showed sustained reads of 29MB/sec in ATTO's benchmark, with sustained writes of 19MB/sec, higher than I saw on this early model PB G3/500.) However every Powerbook I've seen has lower rates than desktop G4s, but still adequate for DV. (See DV rate comments at the bottom of this article.)

Real World Tests:

Copy 102.3MB File to FW Drive:
Time to copy a single 102.3MB file to the Firewire drive.

Copy 102MB file - PB G3/500

Read 102.3MB File From FW Drive:
Time to copy a single 102.3MB file from the Firewire drive to the PB G3's internal drive (IBM 20GB).

Read 102MB file - PB G3/500

Unstuff System Folder:
Time to unstuff a 102.3MB file (327.3MB/3,342 files unstuffed) from the Firewire drive. (Intensive Read/Write test with the drive almost constantly being read/written. Granted CPU speed helps on unstuffing files, but there was some difference in performance with the 3 drives.)

Unstuff SIT file - PB G3/500


Benchmark Tests:

ATTO Tools 2.5 Benchmark:
Rates in MegaBytes per Second (longer bars are faster).

ATTO Benchmark - PB G3/500

VST Benchmark:
The VST Format utility benchmark requires the VST extension to be enabled. This is the only test ran where the VST extension was used with the other drives. In the graph below, longer bars are faster for all tests *except* the Access results (in milliseconds), where lower scores are better. Other rates are in MegaBytes/second.

VST Benchmarks - PB G3/500

Helio LanTest 2.5.2:
Although written to test server performance, I used Helios LanTest 2.5.2 to exercise each of the drives. It runs each test 10 times and reports the average for the 10 tests. For the Read/Write tests (in KB/sec), higher scores are better. For the other tests (opening/closing/deleting/etc.) the times are in seconds (or fractions), so lower scores are better.

LanTest Read/Write - PB G3/500

LanTest Other Tests - PB G3/500


9600/350 Tests:
My refurbished 9600/350 running OS 9.1, with an Adaptec 2940U2B (Ultra2 SCSI) controller card connected to a Quantum 18GB Atlas 10K (series 1), an ATI Radeon PCI (drivers v1.1.1) and a PCI Firewire card (from FWDepot.com, tested here last year). Apple's 2.7 Firewire extensions were used (included with OS 9.1).

Real World Tests:

Copy 102.3MB File to FW Drive:
Time to copy a single 102.3MB file to the Firewire drive.

Copy 102MB file - 9600/350

Read 102.3MB File From FW Drive:
Time to copy a single 102.3MB file from the Firewire drive to the 9600's Atlas 10K drive (connected to Adaptec 2940U2B card).

Read 102MB file - 9600/350

Unstuff System Folder:
Time to unstuff a 102.3MB file (327.3MB/3,342 files unstuffed) from the Firewire drive. (Intensive Read/Write test with the drive almost constantly being read/written. Granted CPU speed helps on unstuffing files, but there was some difference in performance with the 3 drives.)

Unstuff SIT file - 9600/350

MP3 Playback: For those older mac owners curious about audio stuttering issues (often seen with some older macs and PCI IDE cards/drives), I was able to play back MP3 files in Tunes from each of the 3 drives without any dropouts, etc. on this system.


Benchmark Tests:

ATTO Tools 2.5 Benchmark:
Rates in MegaBytes per Second (longer bars are faster).

ATTO Benchmark - 9600/350

VST Benchmark:
The VST Format utility benchmark requires the VST extension to be enabled. This is the only test ran where the VST extension was used with the other drives. In the graph below, longer bars are faster for all tests *except* the Access results (in milliseconds), where lower scores are better. Other rates are in MegaBytes/second.

VST Benchmarks - 9600/350

Helio LanTest 2.5.2:
Although written to test server performance, I used Helios LanTest 2.5.2 to exercise each of the drives. It runs each test 10 times and reports the average for the 10 tests. For the Read/Write tests (in KB/sec), higher scores are better. For the other tests (opening/closing/deleting/etc.) the times are in seconds (or fractions), so lower scores are better.

LanTest Read/Write - 9600/350

LanTest Other Tests - 9600/350


Summary: Although best performance requires a G4 Tower (or Cube I assume), I'm happy with the performance and value of the OWC Elite case. With 60GB of storage and a compact form factor, it will get a lot of use. I know for my older macs with PCI firewire cards, it will more than pay for itself in cost savings over upgrading their SCSI drives, and be usable with more than one Mac. The case was a lot smaller than I expected, which I consider a plus. The combined cost with the 60GB IBM drive makes it a good value for me and the ability to upgrade the drive inside as faster/larger IDE drives become available is a plus. The OWC Elite kit didn't come with any installation instructions, but the page I'm posting on Monday will have all the info you need to install the drive in the case. (It takes only a few minutes to install an IDE drive in the case.) [Update: I've posted a page with photos on installing a HD in the OWC Elite Case.]

For Powerbook owners, fast AC powered cases can't reach their full potential and don't offer a major performance difference over compact bus-powered drives. However they do offer far more storage for the money. It's your decision as to what is more important - portability or storage size/cost. Ideally, it's great to have one of each.

I'm very pleased with the OWC Elite case performance, but I also appreciate the advantages of portable drives that literally fit in a shirt pocket. Price per GB and performance potential (varies by Mac model) favors the AC powered drives. The OWC Elite is one of the most compact and best performing drive of this type I've seen to date. (Granite Digital also sells cases with the Oxford 911 bridge and will be sending one for testing soon. They are not however as compact as the OWC case, which is a plus for me personally since my desktop space is limited.)

The bottom line is, would I buy the OWC Elite case again, now that I have tested it with several Macs? The answer is yes. I know I'll get a lot of use from this case with many Macs here. I also know I can easily update the drive inside when larger ones become available (and reuse the 60GB drive in another Mac). When I bought my first firewire enabled Mac (a B&W G3 in January 1999), I didn't think firewire was that big a deal. In the last 2 years however it has proven one of the most useful and flexible interfaces I've seen for hard drives, cameras and CDROM recorders.
-Mike

Availability:
Prices change over time (going down as capacities go up) so check OWC's Firewire drive product page for the latest pricing/models. Also check the site specials page drive section.


Note on DV Capture to Firewire Drives on PowerBooks:
In reply to a post at Dealchat that warned due to the appx. 16MB/sec (ATTO Tools sustained rates) limit seen even with the fastest Firewire drives on Powerbooks that DV capture may show dropped frames I wanted to post a reality check. DV rates are far lower than that (appx. 3.8MB/sec as I remember), and even the fastest 2.5" IDE drives in Powerbooks can't sustain much over 16MB either (even my PB G4/400 with IBM 32GB drive does just over 15MB sustained reads, just a hair over 16MB/sec sustained write per ATTO's benchmark) - but as this article (above) illustrates, real world tests will show lower rates than ATTO's pure benchmark results. (The same is true of IDE or SCSI drives, ATTO's benchmark rates are much higher than you'll see in real world applications use.) I did some checking and found a page at Digital Origin (makers of EditDV, MotoDV, etc.) that notes a drive with an avg. sustained rate of 5MB/sec or faster is required. (This provides some margin over the DV rate.)

To prove that DV can be captured to a Firewire drive connected to a Powerbook without dropped frames, I decided to test that on the PB G4/400 here using a portable VST Firewire drive (slower than the Oxford 911 bridge based cases as noted in the tests above). With only one firewire port on the PB G4, having a camera and FW drive connected at once (and getting DV from the camera and to the FW drive) isn't as simple as the dual FW port PB G3 pismo models, it still worked.

While I had IE 5.0 running, I switched back to the finder, connected my Panasonic PV DV910 firewire camcorder to my VST 12GB portable drive (which was connected to the PB G4). I then ran iMovie 2, creating a new project on the VST FW drive. I captured live video (DV) to the VST drive in iMovie 2 fine - playback showed no dropped frames - it was smooth and showed no audio or video dropouts. If the VST portable drive can do this, the Oxford 911 bridge based cases could also (since they're faster).

With older macs, there have been many reports (depending on the mac model, firewire pci card, cpu speed, etc.) of dropped frames during DV capture or export to the camera, but in general PB G3 firewire and G4 PB's should be able to capture DV to a good firewire drive ok. (Wallstreets or Lombards with PCcard firewire cards may have different results depending on the card/cpu speed, drive, etc - but Digital Origin's motoDV kit was sold for this purpose, but suggested a G3/300 or faster Powerbook as I remember. It was sold before the PB firewire models were released.)

For PB G3 Wallstreets, lowering the disk cache size to minimum helped with capture rates in my experience with a PB G3/250 wallstreet (big increase in capture rates on my PB G3/250 to the original 4GB drive when the disk cache was lowered). That's not needed with more modern PB's however, based on comments from users and in my own tests.

Bottom line is, at least on this PB G4/400 here even using my (older) 12GB VST portable drive, capturing DV to a FW drive (even one slower than an Oxford 911 bridge based case) worked fine. (And this was while an Airport internet connection was active with IE 5 running in the background)-Mike
[return to top of page.]



Related Articles:

I've posted a page with photos on installing a HD in the OWC Elite Case. For previous Firewire reviews and install guides, see the Firewire articles page. (Includes reviews from last year on the first ADS Pyro, Macally and Clubmac cases, as well as portable Firewire drive tests, portable case kit build guides and Firewire PCI controller card tests.)


Click for SATA3 SSD!


OWC XYM SPECIALS!
(Ad/Sale items)

= UPGRADES by Mac =
Upgrades just for
YOUR Mac!

= Used Macs =
(Click for List)

SSDs from under $50!
Fast SSDs for Most Macs/PCs

= ThunderBolt =
Drives & More

= HARD DRIVES =
Up to 6 TB HDs!
Hitachi, WD, Seagate, Samsung HDs

= 2.5in HDs & SSDs =
Notebook Hard Drives and DIY drive/case kit bundles

= MEMORY =
Lifetime warranty RAM Upgrades!

= OPTICAL DRIVES =
Internal and External Superdrives/Blu-Ray drives

= VIDEO/LCDs/TV =
Graphics cards, TV tuners, DVR, adapters and more.

= SOFTWARE =
Apps, Utilities, OS, VM, Games and more

= WIRELESS =
WiFi and Bluetooth Devices/Adapters/More

= Repair Service =
for iPhone, iPad, Macs

= iPad/iPhone/iPod =
Accessories, Cases, Repairs & More

NuGuard iPhone Case *Extreme* Drop Tested!


XLR8YourMac T-Shirts


FasterMac.net
ISP/Hosting

Help Fight Hunger

 
= back to www.XLR8YOURMAC.com =


= Other Site Topic Areas =
Mac Mods/Upgrades | CPU Upgrades | Storage | Video | Audio/HT | Misc/Software | Search | Recent


Copyright © , 1997-2014. All Rights Reserved
All brand or product names mentioned are properties of their respective companies.

Legal: Site Privacy and terms/conditions of use.