Questions? Suggestions? Need help? Talk about anything related to Yawcam...
Posts: 2
Joined: Tue Sep 20, 2016 11:07 pm


Postby anwaypasible » Wed Sep 21, 2016 12:17 am

using ffmpeg with .avi as the container, i can also record 320kbps aac audio with open broadcaster studio @ 2,500 bitrate
recording the screen, the results are identical without any loss of sharpness & the file sizes are largely different compared to .mkv
36 seconds of .mkv = 17.9MB
27 seconds of .avi = 3.85MB

i know it might be senseless to bring up such a thing with|about yaw, since .avi isn't streamable they say (though no reason why it can't be other than flags set & buffering routines for the audio & video sync).

using a core2quad q6600 without a graphics card to accelerate the encoding, it's the lightweight high quality method.
mpeg4 (default encoder) is used for the encoder.

i just go on to think about how easy things moved along back in windows xp days & how file sizes are getting larger & processing load is getting heavier, yet the quality isn't scaling up along with it.
then there's the data caps adding to emotions about conscious choices to get a good high quality final result the most efficient & sensible way forward.
i've always found it odd how simply cloning an output with copy & paste turns out to be a big cpu heavy load, that is until i realize subtracting from the original quality adds to the processing load - only for them to see an opportunity of changing the way the data is received, from a pure onslaught to a more drawn realization of what is there as if staring at it to bring it into view.
but even the rendering processing doesn't totally add up, because doing it one layer at a time to add onto each sample after the first ; makes me wonder why it doesn't go straight to accurate & exact the first movement - as doing calculations are nothing but a text file list, say a spreadsheet & the hardware has been said to be able to do calculations quick of the one number to another.
the frame of what it is, the frame of what is being added or subtracted, the final of those two frames as there isn't a need to drop in another frame to add a localized addition or subtraction that doesn't cover the entire frame of the picture|video.

i guess rather than totally waste my time, i'll say yaw looks good & strong of visual quality, rather than washed out or compressed the extent of lessening the high quality that comes from what the camera can output.
using nv12 & 709 yuv color space & full yuv color range with the other software & it shouldn't be anything more than a choice of what color compression to use, what codec to put in the header of the pixel data, and then any bitrate tweaks to the encoder to add additional compression on top of it.
because some day, we are gonna be able to choose to roll back the bit depth and|or frames per second to lower the file size for times when the final copy won't be seen on a screen that can do all the colors & resolution anyways - with different file sizes coming from different codecs apparently a choice, there draws room to do the opposite of compression expanding the bit depths & frame rates to get the most from the final result when the hardware allows it.

one of the very first differences i seen was how adobe's version of pdf reader was taking way too long compared to foxit's pdf reader.
straight to the point efficient relying on the system as a whole, or adding your own security features as if the system as a whole doesn't exist.
in part why i question facebook's messenger app needing to be 60MB from google play when it's supposed to be a hook into the browser with the extension of using an overlay on top of other screens.
my phone refused to charge using the last method that was working (nothing more than chargers not adding up to hold the connection as the port wasn't the problem other than a little bit of loose from the usb connector not being engineered to fit exactly) .. that was a nightmare seeing my cpu usage stuck at 100% and cpu speed stuck at 100% & changing the homescreen picture twice & setting the phone down without wifi or bluetooth or gps on expecting it to last as usual, only to see the battery drain down to 0% & shut off by the time i looked at the phone again (twice).
remarkable is quite the topic of the subject & that is what brings people to use a camera to begin with.

in some ways that is a poor man's donation, though knowing people might not stumble onto the information unless they follow the forum more closely.
i started this post forgetting the save to file stream is just a picture & not video and|or audio.
but hows that for the way youtube & other rtmp servers work compared to using ftp with the server name, username & password, then directory being the verification key to their server - all because people aren't allowed to choose the port the receiving system uses.
a bit much for a bit of nothing a translator couldn't filter through.
but that goes to show how they've attempted to maximize the efficiency of their hardware to get as much done per clock cycle as possible.

i come from a computer that uses 3% - 5% of the cpu for audio processing with vst plugins , yet manage to see the cpu usage average up to around 30% sometimes & it's good knowing there's software out there that can get the job done with high quality & not needing a bunch of overwhelming additions of calculations attempting to bring the computer alive rather than treat it as the machine it is.
because as a whole, there's a limit to how much can be allocated to the hardware as far as broadscope handles|threads|processing requirements , to that of the baseline necessary to make it work of yesterday compared to 'new requirements' of today.

adding additional stitching to weak fabric, it isn't an apples to apples comparison until you run out of room to add new stitches & the material stiffens up from thickness , much easier & less worrisome of the future doing the operating system balled up & locked rather than a labyrinth of code that seems to be a tax deduction all the way around that doesn't even prove to be 100% accurate the same way.

i suppose it could go the way of lowering bitdepth in some portions of a picture or video to draw emphasis of emotions shaped in the way the creator intended, compared to all raw & on it's way to tricking the mind about realism.
what people seen as possible to be fake on the internet hiding behind a screen name, people are going to be able to do the same thing again with how much reality they choose to share.
if quality is scarce, then the objective is about the content existing rather than the quality of what the content is.
people choosing to stay there to claim an emptiness as an excuse to not put out quality of what the content is.
perhaps that needed to be said to get some people to upgrade their cam's quality & continuing to thank yawcam for existing.
being able to setup your own webpage with a video & audio feed, & then move on from there to add chat - it's creating their own skype where presets aren't guaranteed to rule the quality beyond the grip of the users with choices about resolution & chroma & color depths & relying on internet structure aside from corporate groups that might be experiencing high workloads.
because from there it is the people who are able to be on cam for that much time & people dropping in with enough energy leftover to acknowledge the people watching - i'd say that user base is quite a catch in the fishing world of software programming.


Return to “General”

Who is online

Users browsing this forum: Baidu [Spider] and 7 guests