Another scene from my short animation. I used Terragen, Lightwave and Hitfilm.
very nice work
I thought to open a new thread about a small problem with Youtube but decided to post here.
One of my friend showed me that the shadow is jerky-jumping kinda in the Youtube video. I thought it was because of the time change that blended the frames a bit and-or the compression Youtube uses. The original i uploaded was 52 MB, the compressed Youtube file is 10 MB. But when i played the compressed Youtube file i downloaded from Youtube on my computer there was not such a big jerkiness of the small moons moving shadow. looks like it is related to the way it plays from Youtube only. You can see the difference by downloading the file or watching it on Vimeo here:
Vimeo is better from a quality aspect mostly anyway as it looks.
I saw stuttering with Youtube in the past but that was related to other problems mostly on the hardware-software side of the users.
But this is directly related to Youtube as it looks.
Not sure if everyone does have the same problem with Youtube but just be aware it might not be related to Hitfilm for example if you see such a problem.
It does look significantly better on Vimeo. Nice work!
Kadri, you've hit the nail on the head. Vimeo's compression is absolutely better looking than YouTube's, and, yes, we do get quite a few threads from users saying, "my YouTube upload looks terrible," to which the response is, "That's YouTube's fault."
Very nice animation, btw. Great sense of scale!
Thanks all. Compression is certainly bad in Youtube. But this kind of micro stuttering was new for me. I didn't saw it first. Only after my friend showed it me in Youtube. After a while i can't see such things on my own videos. Too much familiarity or whatever
Just curious if this is the same for everybody ?
@Kadri I've had the same problem turn up with a very simple Quad Warp example from @Palacono 's thread. There's no jerkiness to the sign in my project or render it's just in the YouTube video
Could be because YouTube is making everything 30 FPS. so shouldn't exporting in 30 FPS eliminate that? Still, 24 FPS support on YouTube is just bad. Watching a blu-rays 24 FPS is night and day in terms of how smooth it is.
@Aladdin4d i see. Yes interesting that it is only on Youtube.
I thought if it was because of some hardware-software use of the browser, HTML5 etc. thing . But i opened Youtube and the Vimeo link in Explorer too (i use Firefox) and it was the same.
Opss! @CitrusNinjaKick i didn't saw your post before. Are you sure everything is playing at 30 FPS on Youtube? If so it could be the reason.
By the way can't we have a basic quoting of other posts here? Would be much easier.
No Im not sure. Your video is playing back at 25 FPS though. If you gave it a 24 FPS video, could this be why?
@CitrusNinjaKick it was 25 FPS and the downloaded file from Youtube is the same. The problem i somewhere else i think. I searched a little but couldn't find exactly this kind of problem (but there were many others). So many variables... The stuttering is small not all saw it. Help from others is limited too because of this. Anyway i can accept this . At least i know that the problem is most probably not on my side.
Well I don't understand much about the technical behind the scenes stuff. I do know that YouTube's bitrate for 1080p 24 FPS is like 5x lower on average than Vimeo though.
@CitrusNinjaKick "YouTube's bitrate for 1080p 24 FPS is like 5x lower on average than Vimeo though."
Can you elaborate on that?
1080p on YouTube is around 4 Mbit people are saying, unless I'm understanding this wrong, Vimeo is about 5x higher as both of these are peaks, at least from my understanding without tricking YouTube into HFR for more bitrate.
They might be talking about just uploading, but it's very high compared to YouTube's page:
So if in reality after uploading 8 Mbit to YouTube, then it's playing back at 4 Mbit (peak), so Vimeo should be half as well. So it would still be 5x, or am I wrong here?
Check bandwidth usage after watching the same video on Vimeo then YouTube I suppose is a good way to find out, but since YouTube is variable (Vimeo might be as well) then YouTube should fall behind pretty far. Vimeo looks way better than YouTube, even without a Pro account.
@CitrusNinjaKick I am going to disagree with your assertion(s).
It's been years since I did misc tests but I just did a quick one to verify my memory. I would not trust any visual bitrate readouts in a video player, or other, to figure bandwidth or video codec settings. I will trust a MediaInfo readout a file. Even better yet a bitrate analyzer but that is excessive.
Downloading from Vimeo is easy (1080 or 720). On Youtube it used to be easy but YT has been playing with their detection of these things to stop high res DLs and I can only get a 720 download.
I uploaded a test clip that is not very compressible. This best shows available bitrate given potential encoder settings tricks. More on that at post bottom. The clip is a GoPro mounted to my bike and I am moving down a trail. Nothing on the ground compresses for spit due to the effective "zoom", but this clip has a bit of sky and that compresses well.
Youtube resulted in 720p30 (29,97 actual) at 3.15Mbps. A 24p upload of the same video gets the same bitrate (anal results => 3.149). Vegas+MainConcept did the encodes and 24p resample. Setting was 12Mbps average single pass VBR. As stated I cannot get a 1080 DL from YT. Directly or via a couple of web based downloaders I tried.
This is the bitrate I expected. To be anal I expected more like 3Mbps tops. I also expected that 24p and 30p would get the same/similar bitrate and they did. Ideally 1080 would be 2x that of 720 but most encoding expects/hopes for better compression of higher res sources.
Vimeo Basic/free gave me a bitrate of 2.5Mbps for 720p30 and 4.5Mbps for 1080p30. I did not expect this. They are lower than I expected.
Here is the unlisted YT Video. The video is soft because it does not compress very well. By that I mean it does not get the type of compression that does not lose video quality.
If people see a different average bitrate than my result and their source material is the more common people talking type stuff, then this can be an indication of an encode with a constant quality type of encode settings. In x264 speak, CRF with suitable VBV params setup for streaming restrictions. I have no interest in creating a test case to find out. The talking people type of stuff compresses like crazy with the good kind(s) of compression (technically redundancy).
I'm kind of late to the party, but really liked that moon video on Vimeo.
It made my heart hurt in the same way that the destruction of the Hosnian system did in Star Wars. Nicely done!
@NormanPCN i got curious and looked at the difference of my video on Vimeo and Youtube with MediaInfo.
Youtube is 3 947 kb/s and Vimeo is 4 458 kb/s (overall bit rate for the 1960x960 HD file). That is roughly a 12 % difference. Not small and from my subjective feeling about the videos it looks right. The difference will be important for guys like we here around but probably hardly detectable for most people.
This is only for this file of course and i don't know if this is true in general for all the videos there.
Just wanted to add. Bitrate is of course not the only important part ... Youtube uses "isom (isom/iso2/avc1/mp41)" codec and Vimeo "mp42 (mp42/mp41/isom/avc1)" . I haven't gone more deeper in the differences of those two codecs and don't know if the difference in bitrate is kind of balanced or not because of the different AVC codecs. My subjective guess is still no.
isom and mp42 are not codecs and are just markers for the MP4 file format version. ISOM being mp4 4.1 and mp42 is obviously 4.2.
MP4 is just a generic file format. MP4 files most commonly have AVC/H.264 video within. The file format version would have (edit) NO effect on the AVC, or other, video contained within. Codec data streams are independent of file format structure definitions.
@NormanPCN i was waiting for such an answer It is just hard to write down and that was the easy way for me. I mentioned AVC especially to state that they are only different versions of the same. I am just curious how much difference there is?
I edited my post. I noticed I missed the word NO. To repeat that sentence here. The file format version would have NO effect on the AVC, or other, video contained within.
AVC is AVC wherever it is used. It is its own standard and no file container, or streaming structure, can change that. AVC is a very large spec and of course encoders make a ton of difference. AFAIK the online services all use x264. I've not read of any other encoder, at any price, that matches it in quality. Especially at low bitrates and online streaming is very low bitrate. Of course, even with the same encoder, the exact encoding params can make a difference.
I don't really have an opinion about the online services other than I see them as pretty much similar in quality. More similar than different. I include Youtube, Vimeo and Smugmug in this list.
@NormanPCN i see. You don't commented on the bitrate difference. For me there is a difference from a quantitative and of of course subjective level. But from a aspect to show the work to people that difference is kinda mood probably.
Strange when i try to edit my post the first time i click "save comment" it does nothing. Then when i click the second time it post the same as a new post. Then i wanted to change the double post but it didn't changed at all when i clicked .
I use Firefox anyone does have the same problem here?
Now it said " body required" or something similar in red and didn't posted.
@Kadri I don't think 12% is that big of a difference in bitrate. At the low bitrates of online streaming it could have a bigger impact. Encoding to low bitrates so often varies by the source material being encoded. Not something I've tested. I have generally considered bitrates in 50% jumps for a more apparent difference.
How did you DL your file from Youtube. The specs are above 720 and the utils I've used in the past can only get 720 from them these days.
I've occasionally got the "body required" message when I click save comment. I just refresh the page and my edits are back and then I click save comment again and all goes well. I use Chrome.
I use Freemake downloader and have no problem at downloading HD files. A little caution is needed at the install but it works since years. Sometimes Youtube screws it up but a new update makes it work ones again.
Since uncompressed 1280x720 @24fps is over 500mbps, by the time you've crammed it to 3 or 4 I think that 12% difference is probably significant.
How did you calculate that @Triem23 ?
What about 4k in say 24 FPS DNxHD, would that be about 4000 mbps uncompressed and how much compressed?
We're talking about disk speed in order to playback smoothly right?