Full video uploaded to Youtube = My video loss a lot of quality in certain parts.

Short clips of that "certain parts" uploaded to Youtube = Quality remains almost "intact".

 

 

Can you link those videos & include the time codes in the full version where YT reduces the quality. That would be interesting to see.

"Certain parts" means areas of high motion complexity, fades, transitions, pans, etc.

YouTube is very concerned with capping bitrates to conserve bandwidth. That, of course, is a recipe for mediocrity.

 

Video completo de Youtube = Mi video pierde mucha calidad en ciertas partes.

Clips cortos de esas "ciertas partes" Subidas a Youtube = La calidad permanece casi "intacta".

 

 

¿Puede vincular esos videos e incluir los códigos de tiempo en la versión completa donde YT reducir la calidad? Sería interesante verlo.

 

Damn, I just deleted the videos. I'll try to upload them again and then give you the links and info.

It's not something you have control over, like the tail trying to wag the dog.

Some were experimenting with higher upload resolutions to force higher bitrates, but it didn't seem to hold interest here for very long.

You'll get better results on Vimeo.

I did the test uploading the same video with different bit rates (but same resolution), the difference is huge (at least in what I can perceive with my eyes).

Sony AVC/MVS / 25,999,360 bps:

Link:

MAGIX AVC/AAC MP4 / 50,000,000 bps:

Link:

MAGIX AVC/AAC MP4 / 135,000,000 bps:

Link:

As you can see, the higher bitrate, the higher quality the video looks. I think this information is very interesting for those who upload videos to youtube.

All the screenshots were taken in the second 2:46. That part of the video is the one that loses the highest quality with the Youtube recompression.

I will now try to render 1080p videos in a higher resolution template to see if there is any change in the results.

"I applaud your curiosity." said the crestfallen old cynic, beaten-down and forced to compromise quality for bandwidth with every technology since he got his first 300 baud modem.

Here is a link that explains the trade-offs that streaming companies and individuals have to make. It will give you some insight into youtube's goals and limitations. You should memorize the bits/pixel concept and learn to do the math as part of your personal growth program.

Here is a discussion from the forum that deals with the issue that you are facing. You may be surprised to learn that people have dealt with these issues before you and I appeared on the scene.

Quote of the day.

"Laurence wrote on 7/26/2014, 8:04 AM

It's not just the artifacts, it's the artifacts on top of artifacts."

I miss Laurence on the forum.

You uploaded "MAGIX AVC/AAC MP4 / 135,000,000 bps:"

My cynical self surfaces once again when I predict that size of upload will get old quickly for you or youtube or both.

"I applaud your curiosity." said the crestfallen old cynic, beaten-down and forced to compromise quality for bandwidth with every technology since he got his first 300 baud modem.

Here is a link that explains the trade-offs that streaming companies and individuals have to make. It will give you some insight into youtube's goals and limitations. You should memorize the bits/pixel concept and learn to do the math as part of your personal growth program.

Here is a discussion from the forum that deals with the issue that you are facing. You may be surprised to learn that people have dealt with these issues before you and I appeared on the scene.

Quote of the day.

"Laurence wrote on 7/26/2014, 8:04 AM

It's not just the artifacts, it's the artifacts on top of artifacts."

I miss Laurence on the forum.

You uploaded "MAGIX AVC/AAC MP4 / 135,000,000 bps:"

My cynical self surfaces once again when I predict that size of upload will get old quickly for you or youtube or both.

I'm sorry, but my level of english is very basic, so I dont understand a lot of things you wrote. Could you be more.. mmm, clearer with the message? more specific? I notice a somewhat poetic tone in your way of speaking, so its difficult to me to understand. Maybe Im wrong.

What do you mean by "My cynical self surfaces once again when I predict that size of upload will get old quickly for you or youtube or both"?

Youtube will likely start rejecting 135 Mbps files, particularly if they are long videos. Also, it will take a long time to upload as most Internet Service Providers offer lower upload speeds than download speeds.

You're not wrong about my poetic style. It is a rejection of having worked and written with precision in a technical environment for forty years.

Youtube probablemente comenzará a rechazar archivos de 135 Mbps, especialmente si son videos largos. Además, llevará mucho tiempo subir, ya que la mayoría de los proveedores de servicios de Internet ofrecen velocidades de carga menores que las velocidades de descarga.

No te equivocas acerca de mi estilo poético. Es un rechazo a haber trabajado y escrito con precisión en un entorno técnico durante cuarenta años.

Now I get you. 40 years is a lot 😅 I think i will take the option of upload them at 135 Mbps while youtube allows it. The truth is that I can see a lot of difference between that and 29Mbps. The 135 Mbps video took like 15 minutes to upload; I can deal with that.

Do you know if the size of the video affects the user in some way when watching it on YouTube? if I'm not mistaken, I belive due the Youtube compression everything is in order to watch it by anybody once its uploaded.

What quantitative metric are you using to reference and compare your three YouTube downloads to the source?

I would ask this even if I had good vision. "

"As you can see, the higher bitrate, the higher quality the video looks" is called a subjective qualifier, not a quantifier (that involves data, my friend!), and no, I cannot clearly see what you claim to report.

Your 30Mbps version is 4288 Kbps on Youtube.

Your 135Mbps version is 4233 Kbps on Youtube.

Your observation is not supported by low-level bandwidth metrics. I stand by my first advice.

Internet 1080p template in Vegas is just the same and should work just peachy for you.

Required Reading:

https://fs.blog/2018/05/pygmalion-effect/

 

"As you can see, the higher bitrate, the higher quality the video looks" is called a subjective qualifier, not a quantifier (that involves data, my friend!), and no, I cannot clearly see what you claim to report.

This shows the difference between the 50mbit & 135mbit upload using magix codec, at least for the 2 screencaps he uploaded. Hopefully it's not a comparision of his screencap .png quality Occurs mainly in highly detailed/contrast outline areas of the flowers & leaves.

With youtube supporting 4K HDR, I don't see them rejecting sub 200mbps clips any time soon, especially since they accept extremely high bitrate formats such as prores. Also, many internet connections are jumping in speed dramatically. In the last year my internet speed jumped from 100mbps up 20 down to 1000mbps up and down.

What quantitative metric are you using to reference and compare your three YouTube downloads to the source?

You're right, I'm not using any quantitative measure and it's not my intention either.

I am (in my personal case) satisfied with the fact that my eyes can tell the difference.

I don't need to know the chemical composition of my food to realize if it tastes good.

Did you really see the frames I pointed out? it's very easy for me to tell the difference, even with the PNG screenshots of the video.

If you can not see it, I'm sorry, but there's nothing I can do about it.

I invite you to make a quantitative measure about it and share your "hard data".

MAGIX AVC/AAC / 135 Mbps

SONY AVC /25 Mbps:

Your 30Mbps version is 4288 Kbps on Youtube.

Your 135Mbps version is 4233 Kbps on Youtube.

Where or how did you get those numbers? As far as I know it's impossible to get an exact figure of the bitrate shown in a Youtube video. It doesn't matter if you're using an app (or a web site) to download the video from Youtube with "no convertion" and then you analize it with Mediainfo. The exact bitrate shown on youtube is something that only Google knows, I repeat, as far as I know.

I would like to add that if you download the videos with an app like aTube Catcher ("without convert") and then you put them in the Vegas timeline and compare them frame by frame, the difference is quite obvious. The one that was uploaded to YouTube with the highest bitrate has the highest quality in textures and details (yes, this is an subjective affirmation), even if the two videos have the same bitrate once they were downloaded.

Seeing a difference and judging a difference are two different things. Just trying to save you some time here, as if you would be producing this stuff.

Good Luck!

Oh, the only valid comparison, since your uploads used different encoders, is to compare each encode with the source, preferably using SSIM and PSNR.

And yes, bitstream capture from YouTube is quite easy with the right tool.

Here's a bit of background for you, Oscar.

We will assume that within YouTube's bandwidth cap for your video (48xx Kbps), they want to process it in the least amount of time to put it live. Well, the third factor is quality, which is a concern to many.

"Speed, Size, Quality. Pick Two" is how the saying goes. That refers to a triaxial relationship, where it is assumed that there is no way to force YouTube to take longer to process your video, in order to preserve more grain and motion, for instance. If there were such a backend control, practically no one would use it and YouTube would shut it off immediately. Why, because it would take longer.

so we make videos with 4000kbps- 8000kbps

With youtube supporting 4K HDR, I don't see them rejecting sub 200mbps clips any time soon, especially since they accept extremely high bitrate formats such as prores.

[Anecdotal]

Last week, I rendered a 15:30 (MM:SS) video at ~90 Mbps. That wasn't my intention. I normally upload UHD for upload at ~ 35 Mbps, but I decided to leave the house and let the upload to youtube proceed in my absence. They rejected it without stating a reason. Since this was an aberration for me, I just rendered the video at my normal bit rate and the upload went to completion.

[/Anecdotal]

In spite of my one-off experience, I agree with your logic.