locked
Error: Not Enough Memory to Complete Requested Operation ????? RRS feed

  • Question

  • Have 3.23GB RAM, every other program is closed, and I get the "not enough memory" message halfway through the 50-minute file that I want encoded. It's a Gotowebinar WMV file and I want to encode it to regular Windows WMV. I have successfully encoded much shorter files of the same format, and don't see what the problem is.
    Thursday, January 13, 2011 6:09 PM

All replies

  • OS? (also if it's 32-bit or 64-bit)

    Source information? Frame size?

    Output settings? Frame size and bandwidth? Etc

    Hardware?

    Do you have any 3rd party codecs enabled? If so, have you tried to disabled all but the Microsoft ones (via the Options->Compatibility dialog)?

    We need more information to be able to help you.

    Thursday, January 13, 2011 7:39 PM
    Moderator
  • While not identical I can confirm similar behaviour.

    I have a number of .m2t files that I want to stitch together to one mp4 output file. If I do 2-3 files this is no problem. If I do 8-10 files Encoder starts doing the encoding, and at some point gives up with "Not enough memory to complete requested operation.". (I have not verified the exact number of files causing it to happen.) In the last attempt I did the files were about 1GB in total size.

    This is on Windows Server 2008 R2 64-bit with 24GB RAM. (Not that the total amount of RAM helps much when Encoder is 32-bit of course.) Source files are 1440x1080i, and I'm using the H.264 HD 1080p VBR output preset. (I am running a lot of VMs as well, but there should still be about 6 GB RAM left for the host.)

    Monday, January 17, 2011 8:33 PM
  • Hi Andreas,

    Your issue, which might or might not be related to the issue of the OP, is a limitation of our stitching algorithm and the error is actually true in your case: we are running out of memory. We unfortunately don't support more than a few source files on the same timeline. It's not about the file size or the number of clips, but about the number of files involved on one timeline. For HD content, it's about 7-8 in my experience.

    To circumvent this limitation, you could break your project in several parts with about 8 source files each, encode the different projects, reimport all the outputs and stitch them using smart recompression (aka Video=Source). When done properly, the final output will have the same quality and the smart recompression operation will be very quick. Another way would be to stitch all your sources together using a free 3rd party tool and import the one file which you could then split, trim, etc.

    But unfortunately, Expression Encoder was not really built to replace the functionality of a NLE. Feel free to vote for such a functionality to be added here: http://expressionencoder.uservoice.com/forums/58981-general.

     

    Monday, January 17, 2011 8:54 PM
    Moderator
  • To circumvent this limitation, you could break your project in several parts with about 8 source files each, encode the different projects, reimport all the outputs and stitch them using smart recompression (aka Video=Source). When done properly, the final output will have the same quality and the smart recompression operation will be very quick. Another way would be to stitch all your sources together using a free 3rd party tool and import the one file which you could then split, trim, etc.

    Yup, this is what I do. And luckily it doesn't add that much encoding time when using the apply source settings. Bit of annoying manual labor of course though :) I have considered automated the process by using the SDK, but I haven't gotten that far yet. (AKA: I've been lazy.)

    Still; it is better to know that it's a limitation of Encoder than having to spend time troubleshooting the issue further.

    Monday, January 17, 2011 9:35 PM