|
Forum List
>
Café LA
>
Topic
Stump the Band..Data DVD Different Under Tiger/Panther?Posted by Greg Kozikowski
Everybody knows you can't make a Data DVD of 4.7 G. You're usually stuck with around 4G or so.
Is it possible you get a different number under Tiger than Panther? We tried to make a Data Dump of 4.32G or so and it will only burn under Tiger. Panther gives us a really nasty error. 10.4.2 10.3.4 ???? Koz
A computer uses 2's complement (1-2-4-8-16-32-64...) arithmetic to calculate bytes as something greater than media manufacture's calculate for their marketing. The reason that a 120 GB disk only comes out at 112 GB is due to the fact that a byte of data is 8 bits and a kilobyte is 1024 bits of data. When a drive manufacturer says they calculate a megabyte as 1,000,000 bytes, that flies directly in the face of how a computer calculates it as approximate 1,024,000 bytes of actual data.
So a 4.7 GB DVD can not hold an actual 4.7 GB of data bytes. It happens to hold approximately 4.37 GB of actual data and that doesn't change with different operating systems. It's the computers way of counting bytes.
<<<It happens to hold approximately 4.37 GB of actual data and that doesn't change with different operating systems. It's the computers way of counting bytes.>>.
Yes, but. We have two files in a folder that work out to 4.32GB. On a Panther machine, the "expanded number" is 4,640,281,708 bytes. If you try to burn that to a DVD, the system will tell you "The item dataTests cannot be copied because there is not enough free space." If you try to eject the disk, you get a burn dialog instead of eject and there is an eject option in that dialog. If you go ahead and burn, you will get a very damaged Data DVD. If you do all these tricks on a Tiger machine, the same folder now has an expanded size of: 4,640,275,560 bytes. There is no error message and the machine will cheerfully produce correct, stable, verified data DVDs one right after the other. They open up on Macs and PCs. Two different Tiger machines do this. 10.3.4 10.4.2 10.4.5 ????? Koz
You wouldn't by any chance be trying (or be able to try) to do this on one machine that has those 2 different OSs on it, would you? It's very peculiar that one OS would interpret the files as one size and the other a slightly different size - not accountable by the voodoo computer math syndrome. It would be nice to know that you were looking at exactly the same files, not just copies of them, when you compare the OS response. If not the same machine, can you put the source files on a FW drive to try on the different machines?
Do you suppose that the 6148 byte difference that is being reported represents the difference between "being small enough" and "being too big" for either OS to handle? There has to be a line somewhere, but that's getting pretty fine. You realize I'm just loading my gun with birdshot here and looking around for a lawyer.... Scott
Here's what I want to distinguish:
Is the difference in reported file size between the 2 different OSs due to a gain (or loss) of bytes in the copying process between 2 machines, or is it a difference in how the 2 OSs read and report file size? If you did copy the files between machines, how did you do it? Network? Burned disc? I'd like to know what kind of numbers are reported when they are the SAME FILES (not copied files) being read by the same OSs. That is, put the files on a FW drive (here we go copying again) and move the drive between machines and see what each reports about their sizes. This may sound like a bunch of trivial hair-splitting, but it could be that you've hit on why certain combinations of OS, QT and FCP work a whole lot better than others. There may be a fundamental file-management difference that occurred in Tiger and its "native" applications. It's a minuscule percentage (1.32e-6) of difference that wouldn't matter in a lot of places, but would still be critical in video processing. Lock and load. Scott
Sorry, only registered users may post in this forum.
|
|