Since I don't have any comparable drives to do a direct comparison with, I'll do what I can to paint as complete a picture as possible with different benchmarking utilities. This should give a reliable viewing of the capabilities of this burner while working in its main purpose plus that as a normal CD drive. Let's begin with the test system:
EPoX 8KHA+ Motherboard
512MB Crucial PC2100 DDR
Seagate Barracuda 40GB HDD
Pioneer 16x DVD
Software used will consist of SiSoft Sandra 2002, Ahead Nero v126.96.36.199, Nero CDSpeed, Nero CD DAE and CD Tach. These utilities should do a fair job of letting us know what the performance levels are concerning this drive and also validate any results between programs.
One of the first things we really want to know is whether we can expect a reasonable amount of normal speed from the TDK burner. With the performance levels on the rise, there is little need to even have a normal CD-ROM drive anymore. So how well can we expect this one to handle the daily grind?
We have discovered in the past that when a disk is capable of reading at a given rate, this reading is done towards the outer edge of the disk. Since there is a lot of data towards the inner portion of the disk as well, we don't get an average score that matches the given specifications from the manufacturer. As a general rule, if you can get more than half of the stated speed as an average speed, then you're normally doing pretty well for yourself.
That said, we have attained a result of 32.1x under CD Tach and 34.5x under Nero CD Speed. This equates to a very workable speed for your normal CD needs. Whether you're going to be installing programs, listening to music or watching a VCD disk, this drive will suffice easily in the task at hand. Also note that under CD Tach, this drive was able to test at 49x burst speed!
One thing that concerned me a little was the CPU utilization of this drive. While not the highest I've seen, it does tend to be a bit higher than some of the competitors out there. I don't think it will be a real issue with the strong processors available today, but it was an item that needed mentioning.
Now that we have determined that the veloCD is capable of being used as a regular CD drive, it is time to see how it does what it was really made for; namely copying disks. To do this, I simply took a 60-minute audio disk and burned a copy. Since I want to see how this drive performs, I will burn the disk as an ISO image to my hard drive and then burn that image to the blank media. This will put all the stress onto the drive itself and not have it slowed down by a slower optical device.
The numbers speak for themselves in this case. The disk was burned at 48x and the time to burn was a very fast 2:16 including Lead-Out. This is very sweet!
Digital Audio Extraction
With more and more folks recording music onto disk, there has been a lot of concern with the quality of the files being extracted. After all, what good is a fast copy of your favorite music if it sounds like garbage?
To check the ability of this drive to accurately extract this digital material, I used a program from Nero called CD DAE. What it does is take a music CD and extract the contents digitally and then copies this data to your hard drive. Not only that, but it checks the integrity of the data twice during the copy and then also converts it to a .wav format. It also gives you the number of errors that occurred during the copy process.
Again, the results speak for themselves. In a total of four minutes and three seconds, I had an hour of music on my hard drive in .wav format. Not only that, but there were absolutely no errors during the ripping process! This will ensure that you don't have those cracks and pops when listening to that custom music disk you want.
What drive review would be complete without the results of the SiSoft Sandra benchmarking utility? Certainly not this one, so here are the numbers:
Unfortunately there are no higher baseline standards within the Sandra interface than the ones listed above. Even with SP1 installed, the highest they compare to is 32x, but the scores still accurately reflect the top-notch performance gains to be expected by using a drive rated at 48x.
Of course, I still don't understand why Sandra identified this drive as a meager 20x when the results smoked everything they had on the chart.