I just read in the faq.
DC. This is the lowest effective QP value used for DC coefficients in the frame calculated by combining QP with the Quantization Scaling Matrix. A DC value below 4 is wasted in 8-bit codecs and just results in extra processing and no gain in quality. If values below 4 appear it is probably appropriate to raise the Q parameter value (or lower the AQ value) until no value under 4 appears.
I am stuck on a value of 2, best that i can get is 3.
i use a gop 1 iframe based 720 pal setting.
AQ is set to 0 All to motion
IQ = set to 28
And Q for 720 = set to 20
If i set Q for 720 for example to 22, 24, 26 or 28, the DC go's to 3, but the image is terrible.
And there will be a wright error, and the accu must be removed.
If i set Q to 18, 20 or 22 the value is 2, but the image is oké.
And no writing errors.
But there is indeed macro blocking in very fine "green" details.
i think the log file's looks oké...
Big question is, any good tips how i can improve this>
Hellow everybody, After view the fantastic page http://www.personal-view.com/faqs/streamparser and don't find an aswer to my question, here it goes:
Anybody knows what "trick mode" means and his convenience or not in a video stream? I can't find info about that. Just some comments in FlowMotion topic, but not enougt for me, sorry.
Thanks in advance.
@cbrandin. Man, you are a great person. Thanks for all your work.
@pc_bel This is what Chris states;-
"The "trick mode" occurs when the timer stamps in the transport stream don't agree with the timer stamps in the elementary stream. StreamParser has to guess at it because there are no flags to indicate that it is happening. "
Why do Spanned files show the second and onwards sequence as 'Trick Mode'?
"Probably what is happening is that with spanned files one set of timers is reset (i.e. the transport stream timers) between files and the other set (i.e. the elementary stream timers) is not. If that's true - then StreamParser is inappropriately reporting trick mode for all but the first file in a sequence... This has never been tested with spanned files and I will take a look into it..."
@driftwood Thank you very much for your help. Nice to get an answer from you. Same as cbranding: Thanks for your work with GH2.
hello every body, I'm mac user,how can I use this sofeware?
Hi guys, is there a similar software for linux? I can't install it using wine! Does anyone had any success instaling it on linux?
Install virtual win os.
StreamParser can also analyse AVCHD files recorded on the Sony NEX-7!
Only one thing to mind: I, P and B frames are displayed without the colors. But it seems to me that the StreamParser recognizes the frames.
If anybody interested, more about that in the GH2 vs NEX-7 comparison topic:
http://www.personal-view.com/talks/discussion/comment/103750#Comment_103750
@cbrandon, When I use Streamparser for analyzing video taken using Moon 24H, the graph is grayed out and "Big info" area shown max/avg/min as all zeros. Moon setting by driftwood is the only one that has that problem with 24H.
This happens with some settings because StreamParser can't automatically detect the GH2 camera. Go to "Configuration>Camera", deselect "Auto Detect Camera", and select "Panasonic GH2". That should fix it!
Thank you Chris, problem fixed.
I want to analize edited files that are not recognized by stream pharser. Is there another program, or coud stream pharser do it in the future?
The importation of .mov files will somehow have to be integrated into @cbrandin 's excellent Streamparser proggie.
BTW Elecard Streameye also cannot correctly import Intra GH3 files - only the 50Mbps IPBs (and often only around 10 second duration files) can be read.
Unfortunately there isn't any way to integrate MOV analysis into StreamParser. Streamparser is a Transport Stream (TS) analyzer. The MOV container isn't a transport stream. The only way to analyze such files is with a full blown H.264 analyzer - like StreamEye.
I don't know what the deal is with the all intra files and StreamEye - I can't get to work either, in fact I can't even demux those files properly.
The only programs that can correctly analyse .mov streams (including the Intra stuff) is VEGA Analyzer, MTS4EA and CodecVisa 4.14.
The long winded way is to use something like -vcodec copy -acodec copy switches to preserve the essences, and then just have ffmpeg write them out to a transport stream. But this is tedious. :-)
ffmpeg -i input.mov -vcodec copy output.h264 to get the Annex B stream (with 00 00 00 01) correctly as a single file. The option;-
-vbsf h264_mp4toannexb
may also be required.
This (see graphic) would look good in Clipwrap (Im trying to convince Divergent Media to incorporate this at the moment) However, from a biz prospective it has to be a value added addition. Therefore, I would be grateful if people would pm me their support for this feature to be incorporated into a future Clipwrap - the more the better. Anyone going to NAB should visit their stand and mention this. :-)
The problem with that is that the Transport Stream would not be the same as Panasonic and probably doesn't differentiate between frame types (that's all non-standardized). I'll have a look, though. Also, I haven't supported 8x8 macroblocks (in the JM-SP part).
It's been a while since somebody has been on this forum, but as a cheap indie filmmaker, I got myself a GH2, and it seems like your software cannot be downloaded anymore (I could be wrong). I'm just wondering if you can send it to me. Thanks! @cbrandin
It looks like you're new here. If you want to get involved, click one of these buttons!