Topic: 24bit files reaching 2GB max file size---solutions????

[Evil Taper]

I'm discovering that editing 24 bit files can be a pain in the ass.  Neither Wavelab nor Soundforge will allow you to work with a file over 2GB in size (you've exceeded maximum file size of 2GB --- or something like that).  What I've had to do is split up a large file before adding the gain because it's not possible with the full file.  After mastering in partial files like this and reattatching them before tracking the track transitions are still flawless.  This doesn't seem to have introduced any trouble as far as I can tell, but I'm hoping there's a way to get around the max file size error. 

A related matter that's come to mind is the possibility of someone recording 24/96 on a laptop or similiar system to the point where the initial file exceeds the 2GB file size limit making it impossible to even save the data and lose the entire recording.

Is this possibly a problem unique to my PC configuration or is this a common glitch that needs to be worked out somehow?

[nickgregory]

the 2Gb is a limitation you cant get past.  I chop all the files up in cdwav first, join where the splits are then if I was to master, this would be when I would do it

[wbrisette]

As Nick points out this is a huge problem. Here are some possible solutions:

1) find an application that will allow you to use larger than 2 GB files (they do exist... at least on certain computer platforms). On the Mac this would include Spark and Digital Performer (as well as possibly other multi-track applications).

2) use Broadcast Wave files. The limitation seems to be ignored in Broadcast Wave files, however you might still run into software issues in opening these files.

3) Do as Nick suggests and break up the file.

All of these issues are going to start biting people as more and more people start using 24-bit files which take up a lot more room on the hard drive...

Wayne

[Ed.]

is there a way to convert a normal wav file to a bwf?  i'd imagine theres got to be something.

breaking the file up would probably still be easier, and if you needed to do mastering in post, there are apps that can do batch processing to all the files if you're too lazy to do it yourself.

[Brian Skalinder]

FWIW, Cool Edit Pro / Adobe Audition handles 2GB files - both primary and temp - just fine.  Its performs processing a little slower than Wavelab, but it doesn't choke outright on > 2GB files.

[wbrisette]

is there a way to convert a normal wav file to a bwf?  i'd imagine theres got to be something.

Because a BWF file IS a wav file simply with an extra header on it, there isn't any real advantage to "converting" so my guess is that you'll be hard pressed to find anything. On the other hand, I only mentioned this because the next rev of the Deva firmware is getting rid of the 4 GB partitions. Howy (the head sw developer for the Deva) is simply going to write the file out to the drive regardless of size. He claims that he hasn't had problems with extremely large BWF files. Again reading these into an application will be problematic, which is why he is also going to allow us to preset break points when we mirror the data off the DEVA and onto a drive or DVD-RAM disc.

Wayne

[Ed.]

will programs that choke on >2gb wav files choke on >2gb bwf files?  thats why i asked the question about converting.

seems that if the program works on bwf fine and not wav, then converting would be a solution...however, if the program just chokes on something larger than 2gb, then converting is pretty much pointless.

[nickgregory]

wavelab does batch processing so it really isnt much more than an extra step

[mmedley.]

The 2GB limit is a software based limitation when handling large wav files.

4GB is the FAT32 'hardware' limitation for ANY file type.

You can use SoundForge and save as a w64 file, I am not sure how WaveLab handles BWF files and different sizes >2GB. Again, limit is 4GB on a FAT32 partitioned drive. I think CDwave can handle 4GB files, although I have never tried, just reading what others have written. The reason everyone uses FAT32 is because all these recorders 'must' be compatible with all operating systems, etc.

Wayne,

Has Howy at Deva thought about using NTFS (or any other partition types) to avoid file size limitations completely?

[MattD]

At least with NTFS, isn't there a licensing issue there?

[Brian Skalinder]

Now that I think about it, I'm running NTFS, so I can't guarantee CEP will handle > 2GB file sizes under FAT32.  But if one of you with a FAT32 formatted drive would try it out and let us know, it'll help out a lot of people who are wondering.

[kindms]

Here is what I do using wavelab.

I run 24/48 and record with Wavelab. i have auto split set to 1500MB. Ususally I have 2 files that need to be put back together so to speak to have a continuous show. Wavelab allows you to combine files well over 2GB it is only when you go to save the file thqat it will puke on over 2GB. So I simply paste the files together and then find a nice place to make the breaks. I usually find a place I would normally put a track maker in anyway and split it there. That way I have files under the 2GB limit and work with them that way. You can also Get Peak when the 1 large file is way over 2GB if you want the peak for the whole show and not each set or whatever. It really only seems to have issues when Saving or processing files over 2GB but you can work with them.

[Rick]

Wavelab allows you to combine files well over 2GB it is only when you go to save the file thqat it will puke on over 2GB. So I simply paste the files together and then find a nice place to make the breaks.

I basically do the same thing...

Open > New Wav
Insert Audio Clips
Dither

[Scooter]

I split my files just like I was processing a 16b file.  Split the large file in pieces to fit on a CDR(s) timewise.  If you don't put fades in at the end of each file, then later if you want to recombine the files to put on DVD-V or DVD-A, no worries, just splice them back together.  The bonus is that all my 24b stuff is usually dithered/resampled for playback in the car anyway, so all I have to do is dither/resample my files that are already split to CD length.  No worries about the 2G barrier.  If you want to keep your file sizes smaller while recording, use the auto-split in Wavelab, or use a multitrack prog and record on two seperate mono tracks (L chan is track 1, mono, R chan is track 2, mono).  That way you can get around 4hrs of rec time, instead of 2hrs.

[dmonterisi]

the 2Gb is a limitation you cant get past.  I chop all the files up in cdwav first, join where the splits are then if I was to master, this would be when I would do it

i don't want to beat a dead horse on this batch processing point, but you might have missed this whole mess that i came up with on this topic:

http://taperssection.com/index.php?topic=49027.0

[nickgregory]

good god, dont know how I missed that...have to read through it when I am not bleary eyed...I udnerstand the problem I believe that SBEs are created on a batch resample which I saw on my own.  for the 16 bit output what i Have done is when encoding to flac, always check on align on sector boundaries, which I *assumed* fixed it....will have to read through the thread to confirm

[dmonterisi]

good god, dont know how I missed that...have to read through it when I am not bleary eyed...I udnerstand the problem I believe that SBEs are created on a batch resample which I saw on my own.  for the 16 bit output what i Have done is when encoding to flac, always check on align on sector boundaries, which I *assumed* fixed it....will have to read through the thread to confirm

it's not that fun of a read...basically, i don't think flac fixes the problems by fixing the sbe's because the resampling algorithm seems to act differently at the end of the file. 

[wbrisette]

Wayne,
Has Howy at Deva thought about using NTFS (or any other partition types) to avoid file size limitations completely?

Yes. Zaxcom developed a file format called MARF. This is what they started using on the Deva II 6 years ago and they are still using it today. 4 GB partitions were put in place because DVD-RAM discs which for a long time were the only media that the Deva could write to, could only handle slightly more than 4 GB. Plus most of the software out there couldn't handle these larger files, so it was driving post houses crazy.

Wayne

[mmedley.]

Wayne,
Has Howy at Deva thought about using NTFS (or any other partition types) to avoid file size limitations completely?

Yes. Zaxcom developed a file format called MARF. This is what they started using on the Deva II 6 years ago and they are still using it today. 4 GB partitions were put in place because DVD-RAM discs which for a long time were the only media that the Deva could write to, could only handle slightly more than 4 GB. Plus most of the software out there couldn't handle these larger files, so it was driving post houses crazy.

Wayne

Cool. That is good to know. I figured they would take that into consideration. The DVD-RAM thing is understandable, but is it still stuck to that standard or are other media formats being taken into consideration now like DVD-9 (DL), etc?

+T

[Evil Taper]

ok, so lets say we're recording something like:

mics>pre>adc>laptop @ 24/96 and the show is a continuous 3.5 hours without a set break.  what software would allow you to record something this huge save it as a single file?  Would a wavelab recording made with auto splitting create an audible glitch between the files or would they be seemless?

[wbrisette]

The DVD-RAM thing is understandable, but is it still stuck to that standard or are other media formats being taken into consideration now like DVD-9 (DL), etc?

Starting with the Deva IV (there never was a Deva III), a FireWire port was put on the box, this allowed hard drives to be added along with DVD-RAM drives, etc. One thing to remember is 98% of the Deva users don't ever touch their audio, post production facilities do. This means they are at the mercy of those facilities when it comes to how they can turn in their stuff. For most that still means using DVD-RAM because they have to use the Fostex DV40 to read back their data. It also means using UDF as the format of the disc, making sure the DV40 has the latest firmware, not putting just 3 channels (or tracks) on the disc (the DV40 has issues with 3 channel masters)... For as advanced as the video and film industry has gotten, there are areas where they simply don't move ahead and audio is one of them. Most post production houses have had 6 years to figure out non-linear recorders, yet film sound crews still struggle with post production houses using outdated firmware and software which means a lot of the advances made have them going backwards (in some cases the DVD-RAM discs are transposed back onto DAT and re-stripped with time code). It simply makes me very happy that I don't have to deal with these issues since I do all my own post production work.

Wayne

[kindms]

ok, so lets say we're recording something like:

mics>pre>adc>laptop @ 24/96 and the show is a continuous 3.5 hours without a set break.  what software would allow you to record something this huge save it as a single file?  Would a wavelab recording made with auto splitting create an audible glitch between the files or would they be seemless?

Wavelabs auto-split feature is seamless. I have been using it forever. It basically makes recording a set it and forget it as long as u have batteries.

[mmedley.]

Thanks for the explanation Wayne. Very cool and interesting!

+T