i assume this topic has been discussed before but in light of some recent posts i’d like to offer up a thread devoted to it…
some individuals theorize that burning “real” data (e.g. DVD-Video) is more indicative of a particular drive’s write quality than simply running a CD/DVD Speed Create Data Disc (i.e. CDS_TEST_B2) due to the larger file sizes and the data not simply being the results of a CPU-generated function…
i’m in the “it doesn’t matter” camp…it’s all 1s and 0s and the drive can’t tell the difference between the two “sources”…
I agree; It does not matter at all. Data is data, and if you a gazillion small files vs. a few big files, it’s still data. The partitions that divide the data into files of different sizes is stored in a table at the start of the disc, so the difference between one huge file and a thousand small files will manifested only in that table (in fact, that’s how it works in other medium as well, such as HDDs–data is data, and it’s the [potentially complex] tables that handle the partition of raw data into files).
And since the generated data is gauranteed to take up the entire disc, it’s a better test if you are interested in just testing a disc. It also eliminates the possibility of hitting a buffer underrun and having the underrun affect the write results.
“Write quality” has nothing to do with what’s on the disc. It’s pits and lands and how they are defined in the dye. Many folks have the very mistaken idea that error scans have something to do with the data on the disc, but they don’t.
But then there are also folks who believe that the software used to burn somehow affects write quality, so there you go.
Another thing is that CD_TEST_B2 doesn’t use data from harddrive but something generated on-the-fly so if you get bad burn you can rule out the harddrive (and/or motherboard) as one the causes. Some may prefer this because it is testing of burner performance/disc quality and not harddrive or motherboard performance. But some may say that in real life you burn real data so you should test everything buy using real data from harddrive when using CD-DVD Speed.
Well, if you’re going to make scans to compare with other scans, then it’s best to control for as much as you can. Even though there is a lot that you can’t control for (like the random variance between each drive), it’s still best to control for what you can.
The only time when differences (if any) produced by these “real-world” results would ever matter is if you’re trying to determine for yourself if your particular system setup can handle things; they will be useless to anyone other than yourself, and such differences are thus not worth posting about.
I am not to sure about that. I believe the random data is generated at the beginning of create data disk and then written back to the burner. So CD_TEST_B2 also uses data from harddrive and thus supports what everyone says here ie CD_TEST_B2 is an indicative of real data burn.
Code65536 Wrote:they will be useless to anyone other than yourself, and such differences are thus not worth posting about.
Do have to insult the guy just because he has difference of opinion with you?, is this how the rule of moderatorshould be?. Then you better know I am one of those that disagree with your analysis and say no way the create data disc is good representation of drive performace as far as quality and burning speed and reality of real job concern.
Then you don’t understand what you are testing for, or how “quality” is being measured. As Code already said, if your concern is whether your hard drive can deliver “real” data fast enough, that’s a separate issue that has no relationship to burn quality or a burner’s speeds. Using datadisc tests eliminates that variable from the test, which is a good thing, not a bad thing. Burning test discs is not a test of your HD performance. Although, there are a number of related IDE and system performance problems that DO show up in datadisc tests, (more so than there are issues that do not show up).
To put it another way: There are no HD or system related issues, that do not show up in a datadisc test, that in any way relate to burner performance or burn quality. If you can name one, please do.
TCAS, I don’t think code65536 was pointing to anyone. Instead the word they in “they will be useless to anyone other than yourself” means that the data disk/burned disk/quality scans will not be comparable to other scans and it’s useful for yourself since they were produced in different settings, for example one burned Titanic movie with hardware spec and hdd speed xyz while others burned LOTR with different specs. This is one point that we are trying to explain, ie eliminating those differences in the tests as many as we can.
With all due respect please explain to me your qualification and involvement in software and hardware industry that you claim I do not understand what I am talking about and I don’t know the real meaning of what I am testing for?. Perhaps we have different opinion on this but isn’t that this forum is for every one to express his/or her opinion indecently without being blasted by other. I thought the purpose of this forums are to discuss, argue and finally to come to a consensus on he issues that we could learn from one another experiences.
When I do CD_TEST_B2, my harddrive LED hardly blinks. But when I burn ISO image using CD-DVD Speed, my harddrive LED has always blinked like mad.
Anyway, I’ve just tested it with NTFileMon and can now confirm that in CD_TEST_B2 test, CD-DVD Speed doesn’t read anything from any harddrives at all from the start of the burning until the end. Once finished burning, the first file its read/write is %TEMP% estresults.pn~ .
The main point here is that it makes no difference where the data comes from that gets burned. For testing purposes, what’s most important is that the disc is completely full. (The outer edge of a disc is where the most problems occur, and if it’s not burned you are missing important imformation.)
If you want to test your system performance or HD performance, there are better ways to do it than burning DVD’s.
I disagree. If your harddrive is slow, or it is somehow configured to run at slower speed than it should be, or it is somehow faulty, etc., you could get bad burn when you burn real data but get good burn when you do CD_TEST_B2. If that happens it can help isolating the problems. I’ve seen newbies who buy DVDRW for the first time and burn real data at 16x being unaware that their old harddrive/motherboard is not fast enough for 16x burning so they get bad burn and think that the burner/disc is to blame. If they do CD_TEST_B2 test and get good burn that would tell them that their burner/disc is good but the problem is with their harddrive/motherboard.
I think the main point here is that people should be aware of the difference in CD-DVD Speed between CD_TEST_B2 and burning ISO image from harddrive. Obviously some people think they are the same thing.
I agree. The disc should be fully filled. I think I read somewhere that CD-DVD Speed can fill the disc full when you burn ISO image smaller than the disc size. But I’ve just tried and that doesn’t seem to be the case.
True. I don’t think anyone here would do that. Testing system/harddrive performance by burning DVD that is.