You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Question / Help Specific bitrate numbers, why?
- Thread starter Razswe
- Start date
to simply answer your question. no there is no reason it's not practically gonna gain you quality or make the encoder run more efficiently. what it will do is make the data/encoder buffer run through a tcp connection at a more synchronized rate but this won't help breaking any limit's.
Some people believe there is a absolute setting for bitrate/buffer, there isn't. Your bitrate and buffer depend heavily upon what you're encoding. Whatever formula he's using to get that number is pointless.
Higher action scenes require more bandwidth in order to achieve a high quality experience. All of this is subjective, because quality is quite subjective as is bandwidth to achieve that level of quality. For instance you could get away with streaming LoL at 720p@30 on 1500 or 2000 bit rate. Where as if you play CoD or another FPS you'll probably need around 2500 or 3000 to get similar quality using the same exact resolution and FPS. The amount of action on the screen is entirely different.
That is just an example though. Someone who plays LoL may spend a lot of time looking all around the map so action on the screen may be higher and may result in reduced quality because of this. So they would need a higher bitrate in order to compensate for this. Conversely they can lower their quality setting to compensate for this, but that reduces quality if more bandwidth is needed.
People generally recommend the same buffer as bitrate because there is generally very little reason to set it differently. I go with a 1.5x buffer size, but that's just me. In practice I've seen very little reason to change this option outside of a 1:1 ratio.
Higher action scenes require more bandwidth in order to achieve a high quality experience. All of this is subjective, because quality is quite subjective as is bandwidth to achieve that level of quality. For instance you could get away with streaming LoL at 720p@30 on 1500 or 2000 bit rate. Where as if you play CoD or another FPS you'll probably need around 2500 or 3000 to get similar quality using the same exact resolution and FPS. The amount of action on the screen is entirely different.
That is just an example though. Someone who plays LoL may spend a lot of time looking all around the map so action on the screen may be higher and may result in reduced quality because of this. So they would need a higher bitrate in order to compensate for this. Conversely they can lower their quality setting to compensate for this, but that reduces quality if more bandwidth is needed.
People generally recommend the same buffer as bitrate because there is generally very little reason to set it differently. I go with a 1.5x buffer size, but that's just me. In practice I've seen very little reason to change this option outside of a 1:1 ratio.
you would have to understand why this is for tcp and what is vbv-bufsize and how does the decoder engages the coming buffer to understand the point behind those values. people always tend to think that a big bufsize would cost you more bandwidth and that is totally wrong, it potentially causes a bitrate spike for a short period of time but that shouldn't cause the playback to go wrong what's more important is the "average" number of bits . a simple switch to trigger the decoder to start decoding the buffer immediately can help making up for any flaws during the transmission and can simply make the player compensate for it if the rest of buffer comes in the same send wave.
the general usage/design for vbv-bufsize is to use it in calculation with vbv-maxrate to match playback buffer length in "seconds" of the player/decoder side. the worse the crf value is, the sooner you want the decoder to initialize the next playback and not wait for the whole receiving buffer get loaded up first. same goes for how big bufsize compared to the maxrate.
ratio bs values are made up and every one would give you a different opinion when it comes to this depending on their personal experience.
the general usage/design for vbv-bufsize is to use it in calculation with vbv-maxrate to match playback buffer length in "seconds" of the player/decoder side. the worse the crf value is, the sooner you want the decoder to initialize the next playback and not wait for the whole receiving buffer get loaded up first. same goes for how big bufsize compared to the maxrate.
ratio bs values are made up and every one would give you a different opinion when it comes to this depending on their personal experience.