While CPU, memory, OS and other variables don't often make a huge difference, there are some nonetheless. We have a computer that is used only for testing hardware. We do this so all tests can be compared reliably.
Test System:
CPU: | AMD Athlon 64 3200+ 2.0GHz | |
Motherboard: | ASUS K8V SE Deluxe with BIOS 1003 | |
Memory: | 2 x 256MB Kingston PC3200 | |
Video Card: | XFX NVIDIA GeForce FX5200 - ForceWare v56.72 | |
Hard Drive: | Hitachi Deskstar 7K80 80GB SATA 7200RPM | |
Operating System: | Windows XP SP2 - VIA Hyperion Pro v5.04 |
For DVD writers the performance tests are broken down into four sections: CD read, CD write, DVD read and DVD write tests. Each benchmark test has been run three times. The score given is an average of the three. DMA has been enabled in device manager for all IDE devices supporting it.
CD Read Tests: For the read performance section of the CD-ROM benchmarks, CD WinBench 99 v3.0 and CD Speed v4.51 are used to test read speeds, seek times and CPU usage. For DAE testing both CD Speed and CD DAE are used. CD Speed is used to give an overall DAE speed rating and CD DAE is used to give the track by track extraction speeds and to check the extracted tracks for errors the drive may have created.
CD Write Tests: To test the write speeds, Nero Burning Rom is used to write 650MB and 700MB to our test media. Times are recorded. To test rewrite scores, Nero is used again to time how long it takes to write 400MB of random files and directories. Then Sonic's DLA is used to test packet writing speeds. The same files are copied and pasted in Windows Explorer and timed.
DVD Read Tests: For the read performance section of the DVD benchmarks CD Speed v4.51 is used to test read speeds, seek times, and CPU usage.
DVD Write Tests: To test the DVD write speeds, Nero Burning Rom is used to write 4GB to our test media. Times are recorded. Then Sonic's DLA is used to test packet writing speeds. The same files are copied and pasted in Windows Explorer and timed.
CD Winbench 99 Scores:
BenQ DW1655 |
LG GSA-H10N |
Lite-On SHM-165H6S |
NEC ND-4550A |
|
CD Winbench 99 | 1400 KB/sec | 1190 KB/sec | 1817 KB/sec | 1530 KB/sec |
Transfer Rate: Inside | 2890 KB/sec | 3183 KB/sec | 2990 KB/sec | 3060 KB/sec |
Transfer Rate: Outside | 6550 KB/sec | 6350 KB/sec | 6578 KB/sec | 6400 KB/sec |
Random Access Time | 116ms | 106ms | 106ms | 101ms |
CPU Utilization | 0.31% | 0.58% | 0.34% | 1.25% |
The CD Winbench test is as close as we can get to testing every day usage. It fires off eight different applications using scripts. This tries to mimic the activities of a person loading these programs onto their own computer.
BenQ's new DVD writerdid pretty well in the CD Winbench portion of our tests. While the DW1655's overall score wasn't as high as some of the other drives, it had some fairly good transfer rates. The drive started reading at 19.3x (2890 / 150) and reached a maximum speed of about 43.7x (6550 / 150) on the outside.
CD Speed v4.51 - Pressed CD:
For this test I used a pressed CD containing one Mode 1 data track. The disc is 74:38 in size and is full of data and directories.
BenQ DW1655 |
LG GSA-H10N |
Lite-On SHM-165H6S |
NEC ND-4550A |
|
Transfer Speed Average: Start: End: |
36.54x 20.71x 48.30x |
35.57x 20.42x 46.78x |
36.68x 21.95x 48.50x |
35.60x 20.54x 47.09x |
Seek Times Random: 1/3: Full: |
107ms 125ms 176ms |
114ms 130ms 175ms |
103ms 116ms 168ms |
137ms 145ms 219ms |
CPU Usage 1x: 2x: 4x: 8x: |
1% 1% 3% 6% |
1% 1% 3% 21% |
2% 4% 7% 12% |
17% 1% 3% 6% |
The DW1655 performed very well when reading pressed CD's. It flew through CD Speed's transfer rate tests, reaching a maximum read speed of 48.30x. Looking at the other scores, you can see that the drive's seek times were also quite good.
CD Speed v4.51 - CD-R Media:
For this test I made a copy of our pressed test CD. I used 12x rated Memorex Gold 74 minute media for the tests.
BenQ DW1655 |
LG GSA-H10N |
Lite-On SHM-165H6S |
NEC ND-4550A |
|
Transfer Speed Average: Start: End: |
37.34x 20.99x 49.48x |
36.30x 20.76x 47.85x |
37.50x 21.74x 49.67x |
36.33x 20.83x 48.03x |
Seek Times Random: 1/3: Full: |
109ms 124ms 172ms |
113ms 132ms 205ms |
106ms 120ms 184ms |
152ms 159ms 250ms |
CPU Usage 1x: 2x: 4x: 8x: |
1% 1% 3% 6% |
1% 2% 3% 21% |
2% 5% 8% 13% |
17% 1% 3% 6% |
Like the other drives, the DW1655 was a little faster when reading CD-R media. Thanks to this small speed boost, the drive was able to reach a maximum transfer speed of more than 49x. On top of that, its seek times were again quite good.
BenQ's new DVD writer had no problems recognizing our 99 minute CompUSA media. The drive read our test disc from start to finish, reaching a maximum transfer speed of 53.19x! This is pretty good for a drive rated at only 48x.
CD Speed v4.51 - CD-RW Media:
For this test I made a copy of a pressed test CD. I used some PNY 80 minute CD-RW media for the tests.
BenQ DW1655 |
LG GSA-H10N |
Lite-On SHM-165H6S |
NEC ND-4550A |
|
Transfer Speed Average: Start: End: |
30.86x 17.59x 40.77x |
30.58x 17.66x 40.13x |
25.56x 15.01x 33.75x |
30.78x 17.94x 40.13x |
Seek Times Random: 1/3: Full: |
107ms 120ms 172ms |
112ms 129ms 182ms |
111ms 128ms 190ms |
153ms 161ms 252ms |
CPU Usage 1x: 2x: 4x: 8x: |
1% 1% 3% 6% |
1% 2% 3% 16% |
3% 6% 12% 21% |
12% 1% 4% 8% |
Like the GSA-H10N and ND-4550A, the DW1655's CD-RW read speeds are limited to only 40x. As you can see, the drive had no problems reaching this speed in our tests.
CD DAE and CD Speed v4.51 - Pressed CD:
For this test I used Pure Funk. The CD is almost exactly 74 minutes. This helps to squeeze the maximum performance out of the CD.
Exact Audio Copy can tell us a lot about a drive's capabilities. You can see from the screen shot that while the BenQ DW1655 supports accurate stream and has the ability to cache audio data, it does not have the ability to retrieve C2 error information.
CD Speed | BenQ DW1655 |
LG GSA-H10N |
Lite-On SHM-165H6S |
NEC ND-4550A |
Average: Start: End: DAE Quality: Accurate Stream: |
36.90x 21.00x 48.76x 10 Yes |
30.73x 17.95x 40.68x 10 Yes |
37.03x 21.26x 48.93x 10 Yes |
30.94x 17.99x 40.12x 10 Yes |
The DW1655 did very well in our DAE tests. While it wasn't able to top the drive from Lite-On, it gave it a run for its money thanks to a maximum DAE speed of 48.76x.
To get a better look at the quality of the extracted audio we use CD DAE. CD DAE is actually designed to be an audio ripping program. It converts the CD-DA on the CD to .wav files. We use it in our testing because it can also be used as a quick and easy way to test the quality. Where CD Speed tests the DAE as one large session, CD DAE extracts each CD track individually. It actually extracts each track twice and then compares them to check for any errors. Every error a drive creates could be a hiss or pop you would hear later in the audio tracks.
CD DAE | BenQ DW1655 |
LG GSA-H10N |
Lite-On SHM-165H6S |
NEC ND-4550A |
Average: Min: Max: |
34.4x 22.1x 47.3x |
29.0x 18.4x 39.8x |
34.5x 22.3x 47.9x |
29.3x 18.6x 39.5x |
Errors | 0 | 0 | 0 | 0 |
While the DW1655 started out a little faster when extracting our test CD with CD DAE, its average and maximum speeds were slower than what we saw with CD Speed.
CD DAE and CD Speed v4.51 - CD-R Media:
For this test I used a copy of the Pure Funk CD. It's burned onto the same Memorex Gold 74 minute media I used in the CD Speed tests.
CD Speed | BenQ DW1655 |
LG GSA-H10N |
Lite-On SHM-165H6S |
NEC ND-4550A |
Average: Start: End: DAE Quality: Accurate Stream: |
36.56x 20.97x 48.17x 10 Yes |
30.48x 17.94x 40.27x 10 Yes |
36.59x 21.69x 48.24x 10 Yes |
30.67x 17.92x 40.11x 10 Yes |
While the DW1655 was a little slower when extracting audio from CD-R media, this didn't keep it from reaching amaximum DAE speed of 48x in our tests. So how did it do in CD Speed's advanced DAE tests? Take a look below.
BenQ's new DVD writer completed CD Speed's advanced DAE tests with an average score of 34.78x. The drive had no problems passing all of the on-the-fly tests and did not create any errors. In the advanced tests, the DW1655 was able to read the CD-Text, subchannel data and lead in section of the CD, but for whatever reason was not able to read the lead out.
CD DAE | BenQ DW1655 |
LG GSA-H10N |
Lite-On SHM-165H6S |
NEC ND-4550A |
Average: Min: Max: |
34.2x 22.1x 46.7x |
28.8x 18.4x 39.3x |
34.2x 22.3x 47.4x |
29.1x 18.6x 39.5x |
Errors | 0 | 0 | 0 | 0 |
To see how well BenQ's DVD writer can read scratched and dirty discs, I used CD Speed's ScanDisc utility to see how many sectors were damaged or unreadable. This is a very rough, but good way to test the drive's error correcting abilities.
CD Speed - ScanDisc
While all of that green looks good, these ScanDisc results are a little misleading. Without the ability to retrieve error information, the DW1655 is unable to determine which sectors are damaged or unreadable.
You can also get an idea of how well the drive can read scratched and dirty discs by using CD DAE. CD DAE will extract the audio tracks twice and then compare them. From this information we can see what the drive's average speed was and how many errors it generated.
CD DAE | Avg. Speed | Errors | % of Disc |
BenQ DW1655 |
6.1x | 50646329 | 6.46% |
LG GSA-H10N |
28.7x | 37659512 | 4.81% |
Lite-On SHM-165H6S |
1.8x | 2069367 | 0.26% |
NEC ND-4550A |
8.2x | 167095050 | 21.32% |
The DW1655 didn't do as well as some of the other drives in this test. The scratches on the disc not only slowed it down considerably, the number of errors was a little higher than what we'd like to see.