25/50 fps Over-crank failure

[Piotr Wozniacki]

I usually do not use any of those fancy features like timelapse, pre-buffer, or cranking - my regular stuff is shooting well prepared, live music performances. As some of you (and certainly Dan) might remember, I was very happy to report that my Transcend 400X cards can be recorded to at 280 Mbps I-Frame (even though they've been approved by CD for 220 Mbps max)....

However, today I just needed a couple of slowmo shots. I set my EX1 to 720/50p, and my nanoFlash accordingly (to 25fps at 50) - and to my very nasty surprise, it didn't work at all!

The nanoFlash kept reporting "card in slot x too slow", dropping the bitrate to as low as 40, then trying to revert back to the (expected) 110 Mbps (I set it for 220 I-Frame). I also tried with L-GOP at the moderate 100 Mbps - same story... Just what the heck, I thought?

Well, when I looked at what has been left over on my CF card (mind you - perfectly capable of 280 Mbps in 1080/25p mode), I noticed that all files' size is just some 109 MB. Why is that? My file size is set to 100% - so, when recording without overcranking, they're some 3.5 GB each.

I'm not 100% certain but I think that if as small a file size as just 100 MB is imposed, the nanoFlash and/or the CF card has a lot to do just closing and opening new files every couple of seconds, and this could be the reason for the lost footage and "card is too slow..." messages.

BTW, sound is recorded in the overcrank mode - this, and the filesize limit of 109MB - is something I don't understand at all. I presume this is due to some error of mine - as I said, I hardly ever use overcranking - but anyway, I'd be grateful for some advise.

TIA,

Piotr

[Julio Veas P.]

Piotr:

I had exactly the same problem you are describing, I then discovered that it occurred only when recording over/undercranking with MXF format, If you record .MOV it works fine.

I know it is not the way it should work, but it solves the problem.

greetings

Julio

[Piotr Wozniacki]

Thanks Julio - just for the sake of trouble-shooting, I'll check it with the .MOV format.

This of course is not a long-term solution for me though, as I need full functionality in the mxf format.

Dan, please?

[Dan Keaton]

Dear Piotr,

I am reporting this to our lab.

They will attempt to duplciate the problem and report back.

Dear Piotr,

Could you please create a "Settings.Txt" file, on a newly formatted card then email me the file.

Andy in our lab will want to check your settings.

I have reported this thread to Andy.

[Piotr Wozniacki]

Dear Dan,

Thanks a lot - please keep me updated. As I said, I do NOT exclude an user error, as I don't use those features too often :)

TIA

Piotr

[Andy Mangrum]

Hello Everyone,

When it comes to Crank, the bit rate reflects the recorded bit rate, which can verify depending on which direction that you go with the Crank Function, you also must keep in mind that when a Card is approved for 220 Mb that is recording with 8 Channels audio, for it to pass, we have seen various cards that will be on the edge, but typically for each channel of audio it adds 1-2 Mb. and must be able to do so for the entirety of the Compact Flash Card.

As for the 100MB files, at 20% file size the files should be 750MB each, so the card was unable to keep up, and each time it steps the bit rate down it creates a new file, you can monitor this with our Fifo Buffer, when you are ready to record Hold down the Left Arrow and press record it will show how well the card can keep up with the bit rate selected.

Likewise MXF requires a little be more overhead, so that a card that may be on the edge may perform better with QT, I would suggest to do the same test again and set the bit rate one or two levels lower (180 or 140), and it should work just fine, or You could use our free file Converter to convert your QT files to MXF (PC and Mac)

Best Regards

[Piotr Wozniacki]

Dear Andy,

1) Re: file size: Why 20% ? Why not 100%?

2) Also, why is audio being recorded?

3) Finally, the nanoFlash specs are the same for MXF and MOV.

All in all - no, with all due respect, I do NOT accept your explanation. Something is definitely WRONG with overcranking.

I just reverted to normal 1080/25p, and can record to the the very same card at as high as 280 (not to mention 220, or 100) Mbps without any single error. So, my CF cards are not at fault, either.



Piotr

[Andy Mangrum]

Dear Piotr,

1)The Example of 20% file size was to demon straight the smallest files that we typically record in MXF or MOV without the use of time-lapse (Which can go down to 2% File Size). If the nanoFlash is recording smaller files it is either due to short clips(User made), or the card cannot keep up, thus it closes the file and creates a new one at a lower bit rate and tried again, till it sustain that bit rate.

2)The mention of audio is it is a factor in what the card can record the more audio channels the more data present, Unless you set the audio channels to zero, So if a Compact Flash card can sustain a higher bit rate but if it is on the edge, the audio channels can become a factor, Regardless of if you use the audio in post or not.

"3) Finally, the nanoFlash specs are the same for MXF and MOV." Technically speaking there are very close, or the same as far as an editor is concerned, but the information I provided was from our lead engineer. and I was speaking at the Software level there is slight differences which may be contributing to this problem.

But I was able to replicate this Issue to some degree in our lab today, and we found that MXF does have an Issue in regards dropping the bit rate, if the card cannot keep up,and that it will drop the bit rate down to 50Mb, (also note that depending on Crank, if you are doing slow motion with a bit rate set for 280 we display 140, ) so the actual bit rate was 100.

This issue does not seem to be present in Mov, it will step the bit rate down level by level till it can sustain the bit rate, as we would expect when not using Crank mode,

As for the 400x Cards being able to record at 280, I have not tried to replicate this yet,

But we will be working on a fix for this Crank Issue, Thank you for your Posts.

[Adam Stanislav]

Originally Posted by Andy Mangrum

1)The Example of 20% file size was to demon straight

Andy, I have to say, that demon straight has to be most fascinating typo I have ever seen. :)

[Piotr Wozniacki]

Thanks Andy/

I understood all your points (including that on the bitrate displayed during overcranking being half of that set in the nanoFlash) - my point though is again: if this particular card is good for up and including 280 Mbps in "straight" recording, why should it pose speed deficiencies in overcrank?

Also, with some older firmware (don't remember now which), I tested overcranking and it was working just fine (with the nano set to 220, it recorded at 110 without a single issue).

I'm glad you will be seeing to this. Thanks again,

Piotr

[Piotr Wozniacki]

Dear Dan and Andy,

Any news on the subject yet?

Thanks

Piotr

[Andy Mangrum]

Hello Everyone,

Sorry about the typo, which was pretty entertaining,

But I do not have a release date at this time, As you all know it takes a great deal of time to make changes and add features as well as to test these features.

As soon as we are near beta status we will post here on DVinfo

Best Regards

[Piotr Wozniacki]

Thanks Andy!

Of course I didn't mean a ready solution - I realize perfectly a new firmware build will not be posted just for fixing this minor issue :)

Good to hear that you have nailed and fixed the issue though.

Piotr

[Colm Whelan]

so iframe is not suitable for crank but there is a firmware fix on the way?

is that correct?

[Dan Keaton]

Dear Colm,

For Slo-Mo, please try shooting in ".MOV" mode.

Then, on a Mac or PC, one can use our File Converter to obtain ".MXF" files.

This is just a temporary workaround until we can post a fix.