sgc9812 Posted June 10, 2009 Posted June 10, 2009 How large a file is ImgBurn able to fit on one DVD+RDL? We want to archive a [software] development project to DVD [+RDL]. The archives are valuable and need to be encrypted, so our plan is to copy the source files set [1000's of files of various sizes - source files as small as 1KB, images of a few MBs, and clips up to 100s of MB] to a TrueCrypt container file, then burn the single container file to a DVD. As the project progresses, the source file set will eventually exceed the size of a single DVD and we'll need to burn multi-volume sets. ImgBurn reports the free space on the DVD+RDL media to be 8,547,991,552 bytes. TrueCrypt will allow us to precisely specify the size of the volume [container file]. We assume that a volume of 8152MB [exactly 8,547,991,552 bytes] will not fit due to some sort of directory overhead. But since we're only burning one huge container file to the DVD, how little will the overhead be for the one file? The bottom line question is, how large a container file will ImgBurn be able to successfully burn onto a single DVD+RDL? Thanks.
mmalves Posted June 10, 2009 Posted June 10, 2009 I've made a quick test with the largest file I have on my HDD (8,533,090,870 bytes), added only that file, used only UDF 1.02 filesystem and the overhead was just 546,816 bytes. My guess is that you can fit a 8,547,440,640 bytes file on a DVD+R DL disc and still have a little bit of room to spare, just in case If only TrueCrypt could mount "raw" optical media then you could use all the space available on a DVD+R DL disc
LIGHTNING UK! Posted June 10, 2009 Posted June 10, 2009 You'd have a bit more room if you just used ISO9660 (and Joliet), and a bit less if you up the UDF revision to 2.50+
sgc9812 Posted June 10, 2009 Author Posted June 10, 2009 One of our goals is to build highly-encrypted archives of the project - thus TrueCrypt [TC]. We're TC novices, so we assumed we'd have to utilize TC's "container files". We'd rather be managing fewer container files, so bigger files are better [fewer of them]. Won't ISO9660 impose a 4GB restriction on the container file size?
LIGHTNING UK! Posted June 10, 2009 Posted June 10, 2009 Oops, I forgot you were using a single file... just ignore me
sgc9812 Posted June 10, 2009 Author Posted June 10, 2009 I've made a quick test with the largest file I have on my HDD (8,533,090,870 bytes), added only that file, used only UDF 1.02 filesystem and the overhead was just 546,816 bytes. My guess is that you can fit a 8,547,440,640 bytes file on a DVD+R DL disc and still have a little bit of room to spare, just in case If only TrueCrypt could mount "raw" optical media then you could use all the space available on a DVD+R DL disc Thanks for the reply. I appreciate the proven number [8,533,090,870 bytes], and the guess [8,547,440,640 bytes]. I'm willing to create a test container file of the "guess" size... But, could you enlighten me as to the math behind the "guess" value? Why 538KB less than the media capacity? Regardless, I'll run the test and report the result.
mmalves Posted June 10, 2009 Posted June 10, 2009 But, could you enlighten me as to the math behind the "guess" value? Why 538KB less than the media capacity? It's 538KB less than the disc's total capacity to make sure there's enough room for the UDF filesystem
LIGHTNING UK! Posted June 11, 2009 Posted June 11, 2009 It should be easy to calculate. The file size (of a single file) has no bearing on the size of the file system portion of an image (taking into account that ImgBurn will round the size of the final image up to the nearest 16 or 32 sectors for the final UDF anchor). You can use fsutil to create a file of any given size quickly. Feed that file into ImgBurn's Build mode and hit the calculate button. Keep adding to the file size (2048 bytes at a time) and when it goes over DVD+R DL limit, you know you've gone too far
sgc9812 Posted June 11, 2009 Author Posted June 11, 2009 Here's the test result... The largest file [to 2048 byte resolution] you can burn to a DVD+RDL using UDF-1.02 is 8,547,452,928 bytes. The value was arrived at using mmavles's guess [8,547,440,640 bytes] as a starting point, then using LightningUK's fsutil/ImgBurn-calculate technique to create and check files of various sizes. You'll discover [quickly, if you use a binary search tree] that a file of 8,547,452,928 bytes will fit, but a file of 8,547,454,976 bytes [2048 bytes more] will not. mmalve, LigthningUK, thanks for all your help. - Scott [Perhaps an enhancement for a later version of ImgBurn would be to report the largest file that can be accomodated on the selected media, using the selected file system -- though we might be in a small minority of users who are in the flexible position of fitting our file to the media, rather than the media to our file.]
Recommended Posts