Video Quality: Vimeo vs YouTube

As I’ve written in the past, since we bought our Roku XD|S, we watch lots of online video. Of course, my favorites are mostly on Vimeo, but not because YouTube doesn’t carry the same videos (it does), but simply because it’s easier to find them on Vimeo — since quality content is encouraged, and promoted there (rather than these stupid vlogs that are getting promoted by YouTube).

So while we were watching such content, on our 50″ Pioneer plasma TV, from 2.5-3 meters away, my husband and I clearly noticed that YouTube HD videos were not as detailed as Vimeo’s. Ignoring the PNSR result graphs below, even with the naked eye the difference is noticeable. There was lack of detail and sharpness. And I’m not talking about artificial sharpness that anyone can add during post-processing, but lack of true sharpness and detail in the video compared to the original. I could see this problem on my PC too for months now, but I thought that on a TV this would not be as evident. But it was.

So I went ahead, ran some tests, and here’s what I found. Utilities used was pDiff, a pixel differentiation meter over the original, and a bandwidth calculator. The originally uploaded videos can be downloaded from Vimeo, if you’re logged-in. I used video samples from PLUS users because these are the only ones that the original file is still in tact after a week (non-PLUS users can only offer for download the re-encoded version after a week). You can double-check and verify the results after downloading some of the frames used for the comparison (12 MB).

1080p sample
YouTube: 4 mbps (includes 128 kbps audio). Pixel difference: 644980 px.
Vimeo: 4.62 mbps, 2-pass (4.5 mbps for video, 128 kbps for audio). Pixel difference: 513204 px. Please note that 1080p is only activated on Vimeo IF a Vimeo PLUS user specifically checks the 1080p box in the video’s preferences.

720p sample
YouTube: 2 mbps (includes 128 kbps audio). Pixel difference: 258864 px.
Vimeo: 2 mbps, 2-pass (includes 128 kbps audio). Pixel difference: 213288 px.

The difference looks small, but it’s actually very visible while watching something. YouTube’s version has too much softness over both Vimeo’s and the original’s. I personally think that both Vimeo and Youtube should upgrade to 2.5 mbps for 720p (with 160 kbps VBR audio), quality could be better in that resolution — there were times that I wasn’t satisfied with either of the two!

But for 1080p, Vimeo seems to have found a sweet spot at 4.6 mbps: quality is indeed stellar, and better than our Comcast cable box! This was one of the multitude of reasons we actually got rid of our cable TV just last week. On a second thought, I’d love to see Vimeo supporting 5.1 audio in 1080p, bringing the bitrate at 5 mbps overall.

9 Comments »

Henry wrote on December 22nd, 2010 at 10:16 AM PST:

Hey Eugenia…I’ve always noticed the contrast between the two. YouTube,on it’s best day,always seemed a tad too fuzzy for my taste,especially after I stumbled upon Vimeo…even the better quality professionally produced ones.But in my personal opinion,Vimeo seems to be geared more towards those with a certain aesthetic for the cinematic look.And they promote that as well.Not the viral of the week.That’s kinda where YouTube has it’s strong point…that and corporate ads.


This is the admin speaking...
Eugenia wrote on December 22nd, 2010 at 10:44 AM PST:

Yes, I’m glad that Vimeo does not do virals. Virals are cheap skits, fast laughs that everyone forgets soon after. A truly artistic video can stay with someone for much longer.


Karel Bata wrote on December 25th, 2010 at 9:01 AM PST:

Thanks for that.

What you say is absolutely true for the average user, but I I’ve always wondered if commercial YouTube users get superior encoding. And if they choose to do the trans-coding to flv themselves (with software tweaked specifically for YT) they get it looking exactly how they want it. Vimeo doesn’t allow for uploading flvs directly, so YT would have the edge there.

And you can’t beat YT for exposure/publicity, and it has some awesome tools for analyzing who’s watching and how long you retain their attention. Take a look at this
YT Audience Attention curve for the video Sid Vicious Auditions for X-Factor I’ve learnt a lot from these, like that it’s not how many hits you get that matters.

BTW Don’t forget ExposureRoom, who IMHO have the best quality. I’d be interested to see you run your tests there. 😉


Richard Allen Crook wrote on December 25th, 2010 at 2:29 PM PST:

Great info Eugenia. I all but forget to update our YouTube site, because Vimeo is just so much better looking and now I know why. Thanks!


This is the admin speaking...
Eugenia wrote on December 25th, 2010 at 4:14 PM PST:

Karel, you have misconceptions about a few things. The On2 FLV format is way worse than MP4 h.264, so not having support for it on Vimeo, is a GOOD thing. Vimeo decided to not support FLV because FLVs are not normal video exports usually, but rips from Youtube — which could put Vimeo in a bad legal situation. Uploading FLVs is the wrong way to do it anyway, because of their bad quality at the same bitrate compared to h.264. I don’t think that Youtube offers better quality to their commercial partners btw.

As for analyzing traffic, Vimeo PLUS has exactly the same tools. They added them about a year ago.

Regarding exposure, this is both true and false. YT has MORE viewership, but the majority of these viewers don’t appreciate what they’re viewing. YouTube only has more viewers in comedy skits and virals. When it comes to ART, Vimeo is on par, AND the kind of viewers that watch the videos are the kind that MATTER. They all leave nice comments, and good feedback, in contrast to Youtube’s trolling.

Also, Vimeo is way more presentable than YouTube, and this is why some of us are using Vimeo as our portfolio. We would never use the butt-ugly YT for that, it’s just not well-designed.

As for ExposureRoom, I don’t think they have the best quality. I just watched a 720p video there, I wasn’t impressed. I didn’t benchmarked it though so I don’t have hard numbers.


Karel Bata wrote on December 25th, 2010 at 5:48 PM PST:

Hmm…

Fact is that the Vimeo videos you’re using for your comparison are from Vimeo ‘Plus’ users, who benefit from 2-pass encoding, and pay for the privilege. A fairer comparison would have been using regular Vimeo users who get the poorer quality.

I bet a comparison between ‘free’ users, where the YouTube video had been encoded using good 2-pass software before uploading, and the Vimeo video had to accept the standard one-pass service, would find YT to be the clear winner.

BTW I see no mention on Vimeo’s stats marketing page of any Audience Attention curves like YT have.

But I would certainly agree about YouTube’s poor interface.


Karel Bata wrote on December 25th, 2010 at 6:12 PM PST:

Oh, and Philip Bloom uses Exposure Room for his blog. He clearly rates them. You might want to check out the quality of his uploads for your test.


This is the admin speaking...
Eugenia wrote on December 25th, 2010 at 6:31 PM PST:

I used PLUS videos because there is no 1080p support for non-PLUS users, so the comparison would have been impossible.

Also, you’re wrong about 2-pass being so different. Usually, 2-pass doesn’t do more than 5%-6% better quality, and the difference above is more than that. And don’t forget that Vimeo uses more bitrate for 1080p, so even with a single pass it will still win.

For the Audience Attention, it is not a graphical curve, but it tells you how many people watched the full video or not. But it doesn’t really matter, what we compare here is quality, no other features. Youtube doesn’t offer the same quality, be it a commercial partner or not. And even if it did, you can’t just become a commercial partner, you’ll have to be picked up by them.

As for Philip Bloom, I mostly see his good videos to be embedded from Vimeo on his blog. He uses ExposureRoom for tests only I think, the way I use Youtube for my tests.


Karel Bata wrote on December 26th, 2010 at 4:42 AM PST:

You’ve convinced me (as far as quality is concerned).

At the end of the day it depends on what the video is, and who your target audience are.


Comments are closed as this blog post is now archived.

Lines, paragraphs break automatically. HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

The URI to TrackBack this blog entry is this. And here is the RSS 2.0 for comments on this post.