Print

Print


Thanks for the kind words, Steve.

I feel that we are both going down the proverbial rabbit hole. I have 
tried installing the BWF MetaEdit upgrade (wouldn't run, yes I got the 
X64 version for my 64bit W7 machine), reread the 2011 survey (not 
word-for-word), and tried a few other tools (one of which is being 
abandoned and was Mac centric though Windows versions were, at one time, 
available). Your simple question, coupled with my curiosity has cost me 
probably two hours this morning, but I was hoping that things would be 
better and that finally everything would fall into place for this workflow.

Suffice it to say, I continue to be horribly frustrated with metadata in 
general and embedded metadata specifically. This is more the case with 
audio than with still images, but there are issues there as well.

<non-audio aside>
For example, Nikon has not released a version of the top-of-the-line DX 
format camera (D7100 of which I have two and is now upgraded to the 
D7200) with internal GPS. So, when I care, Steve Puntolillo suggested 
just snapping an image with my smart phone from the same place. That 
worked fine and is probably more reliable than a cable-connected GPS. 
Other than that, the copyright and exposure metadata comes through 
pretty well (other than the encrypted as-shot white balance!!!!).
</non-audio aside>

I find the spreadsheet to be the easiest way to collect most metadata. 
If you need the MD5 checksum for a file, you might try FastSum which 
will produce one file, one file per folder, and one file per file with a 
list of filenames and MD5 checksums. You could then open the MD5 
checksum file in Notepad and copy and paste from there in one quick 
operation.

I have only done one project using BWF MetaEdit as the client and I both 
wanted to experiment with it. It was OK, but it has not been stable.

For the last half decade (in round numbers), I have been providing MD5 
checksums in a file-per-file format for almost all my projects.

I should get back to troubleshooting my 2nd dbx rack as I have a 
16-track dbx-encoded tape to digitize and the power supply rails are 
acting up. My guess is the bypass cap on one of the channel cards is 
shorting from the way it's behaving...grrrr.

Cheers,

Richard

On 2015-05-19 10:48 AM, Steven Smolian wrote:
> Hi Richard the Ever-Helpful,
>
> When opening the file in Wavelab 8.5 after running the MD% check box option, there are three distinct segments in the RH top panel that shows in bold Metadata, BWG and MD5, with the added or derived components listed under each.
>
> At this point, I copy and paste to a spreadsheet, a time-consuming, laborious process, especially as most of what I'm doing this is for large groups of under-15 minute files.   There's GOT to be a better way to find a dedicated utility for this without getting tangled up in larger, more expensive and more "learning required" programs.
>
> Steve
>
>
> -----Original Message-----
> From: Association for Recorded Sound Discussion List [mailto:[log in to unmask]] On Behalf Of Richard L. Hess
> Sent: Tuesday, May 19, 2015 10:16 AM
> To: [log in to unmask]
> Subject: Re: [ARSCLIST] another file format question
>
> Hi, Steve,
>
> This is much harder than track formats and balanced audio interconnects <sigh>.
>
> To the best of my knowledge, the only way to put an MD5 checksum (let's call it that or hash if you will) into a Broadcast WAV file is to use BWF MetaEdit which Dave Rice helped create.
>
> It is available here:
> http://sourceforge.net/projects/bwfmetaedit/
>
> I see there has been an update within the last month (I had gotten tired of checking for updates for a while) which is good news and I conditionally retract my comment about it being "buggy" until this version is proven--hopefully it won't be buggy.
>
> I use three applications which can create BWF metadata:
> Sound Devices recorder
> Samplitude ProX2 Suite
> Izotope RX4
>
> All three can make a "created date," but RX re-creates that if working on a file made on the Sound Devices recorder.
>
> The MD5 checksum is NOT part of the BWF unique identifier and recorder information. It is a separate chunk, as far as I can tell, specific to this function.
>
> It may be that Wavelab can write the MD5 checksum chunk into the file.
> (The biggest chunk in the WAV file is the actual audio data, what SMPTE calls the "essence" and then there are little chunks for various tags.
> Broadcast WAV "bext" chunk is the one with the coding "history" which I use advisedly.)
>
> Wavelab has been embraced by the archival field and may have integraged the MD5 checksum architecture embodied in BWF MetaEdit.
>
> If you created the file in Wavelab and then added the MD5 checksum in Wavelab (assuming it might do that) OR BWF MetaEdit and then processed the file in iZotope RX4 and exported it, retaining the original MD5 checksum would be illogical as the only reason to run RX4 would be to modify the contents of the audio chunk which would then invalidate the
> MD5 checksum.
>
> I have abandoned the use of internal MD5 checksums for files and still use FastSum to create external MD5 checksum files which then FastSum can automatically check.
>
> I do not know if the fixity checkers that are available work with MD5 files created by FastSum, but my intent is allow my clients to verify the files after download as I deliver much of my final work product electronically.
>
> Does this help at all to get you over the hump you are experiencing?
>
> Cheers,
>
> Richard
>
>
>
> On 2015-05-18 3:45 PM, Steven Smolian wrote:
>> Hi, Richard,
>>
>> This is precisely what I am doing- or think I'm doing, including unchecking the 32 bit box in RX4. .  I look at the metadata in Wavelapb 8.5 but see no MD5 number.  When I navigate to the check box for same in Waveab, the gear logo remains in the box at the upper left but doesn't turn or indicate it's alive in any way.
>>
>> I suppose I'm looking for a utility into which I can paste the filename that I want the  MD5 number for and it will run the program and fill in the numbers.  I would then like to export it to the hard drive on which I have loaded the identical files (I hope), have it calculate those as well and insert them in another column to the right of the first one, all as a plug-in or an integral part of one of these two programs.  Is there such an animal?
>>
>> Steve
>>
>>
>> -----Original Message-----
>> From: Association for Recorded Sound Discussion List
>> [mailto:[log in to unmask]] On Behalf Of Richard L. Hess
>> Sent: Monday, May 18, 2015 2:43 PM
>> To: [log in to unmask]
>> Subject: Re: [ARSCLIST] another file format question
>>
>> Hi, Steve,
>>
>> What I was thinking of in iZotope was the Export File dialogue which contains, across the top:
>>
>> WAV, AIFF, FLAC, OGG
>>
>> For WAV, the options are:
>>
>> Bit depth, Dither
>>
>> (check) BWF     (check) Preserve non-audio data
>>
>> (check) Reopen file in RX    OK     Cancel
>>
>> It is the two check boxes below Bit depth and dither:
>>
>> When exporting from RX, the WAV file is created anew.
>>
>> With BWF checked, I think that passes through the other BWF data.
>>
>> With Preserve non-audio data, I think that passes through other data (such as the BWF MetaEdit MD5 checksum and probably the tags that Tom is discussing).
>>
>> Thanks for asking, Steve. I just posted this to the RX forum:
>>
>>> I may be confused with this, but if I open a Broadcast Wave file
>>> (generated on a Sound Devices 722) in RX4 and then save it with the
>>> BWF option checked, it replaces the Sound Devices origination stamp
>>> with its own and the current date.
>>>
>>> I thought that the Origination stamp should not be over-written, but
>>> that RX should write into the coding history that it had processed
>>> the file.
>>>
>>> I don't know if this is a bug or not, but it seems not to work the
>>> way I thought it should.
>>
>> I know nothing about WaveLab since 6.1.
>>
>> Cheers,
>>
>> Richard
>>
>>
>> On 2015-05-18 2:07 PM, Steven Smolian wrote:
>>> Hi, Richard,
>>>
>>> I'm using Izotope- great program.
>>>
>>> However, I've not tried the MD5 option (nor even looked for it) as
>>> Izotope stores everything in 32 bit floating format and I usually
>>> convert to 24 bits so I know it plays on other machines.  I'm afraid
>>> to get it come up with a code with the 32 floating as part of the
>>> imbedded code which won't match the 24 bits in which I ultimately
>>> save the file.  Comments?
>>>
>>> So I export to Wavelab 8.5 where I add metadata.  It shows me  the
>>> MD5 logo, I hit the checkbox to invoke it, but nothing happens.  I'm
>>> awaiting an email back from Steinberg about this.
>>>
>>> Steve
>>
-- 
Richard L. Hess                   email: [log in to unmask]
Aurora, Ontario, Canada                             647 479 2800
http://www.richardhess.com/tape/contact.htm
Quality tape transfers -- even from hard-to-play tapes.