Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Well, they did their demo a little differently. Example files seem to be built into an .EXE file itself or one of DLLs and are accessible via special "example files" menu. You can't open an arbitrary file at all.Have you tried renaming your clip to whatever the demo file is called? Chris
I tried opening several different MTS files with H264visa; it crashed on GH1(3) ones and couldn't even open the Sony NEX5 ones. It seems very buggy.h264vista parse files by that specification ... (for GH1 firstly it extract stream from mpeg transport stream, if not done - crash caused /0001 is header as for 264 block as for mpeg block but they very different ... so when it crossed in not right sequence - will exception /)
Starting a wishlist for the outstanding GH13StreamParser by one cbrandin.
I wrote the utility originally to analyze the rogue frame issue, which it solved within hours proving that rogue frames don't exist.
The design goal is to analyze how the GH1 CODEC actually works, not how it should work according to the standard - thus I'm doing my own packet decoding. The program never will be a EyeStream replacement; that would require months of work. However, most of the suggestions offered so far will not be too hard to do. Sometimes a feature that sounds complicated turns out to be easy, and visa versa. So I don't mind suggestions - keep them coming. Just don't be disappointed if some of them aren't done. It only took me a few days to write the program because I'm not trying to write a generic H.264 analyzer; rather it's a simple dedicated GH1 AVCHD analyzer.
Good stuff, Chris. I really couldn't find anyI added a lot of logic today so don't be too surprised if a few bugs show up.
Here is the next version of GH13 Stream Parser.
Chris, I think it looks about the same graph and data-wise, but I did get an error screen during opening the file this time, which I believe I haven't seen before. It's kind of weird that by looking for giant rogue frames we have found tiny little ones to be the real problem ...I considerably improved the robustness of the parsing algorithm; you may want to re-run your tests with this version as it is more accurate (svecher, you may want to re-test that file from richtrav - let me know if anything changes).