Why tying "video quality" to frame size and rate is dumb.

[-Reda-]

Y'all need to stop trolling cause not once have I said "upscale" or simply "LANCZOS TO 720P THEN ITLL BE PERFECT". See Godix's post for details.

[Vivaldi]

You can say that all you want, but it's really common knowledge that upscaling decreases quality. This thread isn't anything more than place to Baww about the fact that some people didn't like your video, under the thinly veiled guise of some sort of public service presentation.

Fact 1: I am NOT the original poster.
Fact 2:

good video quality and sound quality (not a 10 though because its not 720p or 1080p).

Let's break that quote down shall we:

"good video quality" and then "(not a 10 because it is NOT 720p or 1080p)"

SO.. in THAT regard.. the correlation is "unless the video is 720p or higher, then a video is not to receive a 10"

Therefore, he makes the strict correlation that only 720p or higher videos can be rated a 10.

The point of the thread is that resolution does not equal video quality. In fact, 720p or higher resolution DOES NOT EQUAL HIGH DEFINITION. This means, the expectation that because a video does not have a resolution of 720p or higher is ENTIRELY INCORRECT. -Reda- made no mention of the actual quality of the video itself, but rather judged it simply by a resolution. So, PAY ATTENTION.

On a side note, while I do care if people like my videos to some degree, and will often argue back about things I disagree with, I sure as hell don't try to disguise complaints about comments made on the videos I make. And tons of people don't like things I've made in the past and will continue to do so, so don't give me none of that lip. I'm fully aware that my videos aren't the greatest things ever (not even close and neither is anyone else's) and I have never claimed them to be.

I was referring to all the people involved in this, not just you, and not just trythil.

I was also not defending -reda-. In fact you'll see from my post that I think rating something for it's resolution is dumb Point for point I basically agree with you and trythil. I'm referring your (in the collective sense) attitude to the whole thing. I get the "point" of this thread, I'm saying the point is "no duh" and the only practical purpose of this thread is to bitch about -reda-'s comment.

Also, I'm not saying you think your video's are the great thing ever, or that everyone likes them. I'm not saying you shouldn't discuss things you disagree with. What I'm saying, is that dispite what you say, you (in the collective sense), in the testoROS thread, are bitching and moaning and bawwing about every single little thing someone says. And I just think that's bad ettequitte.

-if someone says they don't like it, that's an opinion. If you disagree, you could engender discussion about it.
-if someone says something that is factually wrong, you could calmly correct them.
Lashing out at everybody isn't making a stand for yourself, it's just rude.

The quality's a non-issue here. The issue is the huge bitchfest people are cooking up since they apparently can't take an opinion with an ounce of grace or complacency. That's my problem.

Would it have made a difference if I had dragged out the RoS3 introduction instead?

So you'd rather drag it out across the whole forum? This could've been solved in one post.

"O_o Not everybody had HD sources, and it would've been detrimental to upscale just for the sake of higher resolution."

I mean, it would have taken me longer to do: I don't have the files handy (they're archived, and may have broken paths that need to be fixed), whereas the molecule scene was in ~/prj/testosteros. But I can repeat this demonstration with those files, if you want. The result is the same.

The "maligned" adjective refers to my publicized opinion of the introduction sequence. You can find by reading the announcement thread for "that thing".

===

I really don't care whether or not people like "that thing"; it is irrelevant, and as Niotex pointed out, was merely used as a convenient example. What I do care about is correcting what I see as a erroneous mental construction.

You may recall that I once posted, on this board, a request to not confuse "free software" with "open source software", and described the differences from my point of view. That post was met with roughly the same level of accusation, but I did make some headway. Maybe.

(let me get philosophical here)

What people write -- what they say, the language they use -- is often a good window into how they think. (For you linguists: yes, I know Sapir-Whorf is unproven and (for all I know; I haven't kept up too well with language research) possibly even discredited, but linguistic constructions have to come from somewhere, right?)

I saw, in that opinion quoted way at the top of this thread, a link between frame size/frame rate and the judgment of "video quality" that, in my (hopefully reasonable) opinion, is wrong. (And dumb, but I will admit to using some invective there for theatrics.) I saw an opportunity to announce my position on that link, and invite discussion.

(end)

[/quote][/quote]
But given the fact that you used testoROS as an example and the thread's still active, surely you can see the negative connotation that has? Of course I can't prove that this thread exists only to argue in defense of testoROS, But the presentation seems rather shallow, especially considering that the basic information that an upscale in resolution does not equal an increase in quality, which is all that was imparted through this thread's original post, is common knowledge. The actually possible disscuscion of whether resolution should play into a video's ratings wasn't arrived at without needless bickering.

[blabbler]

I could subdivide the sphere meshes a couple times, which would help.

more subdivisions is all you need. the version i saw was sd and the low polycount was very obvious. you also seem to have an issue with colour matting - i'm not sure how you dealt with the alpha or what compositing you did, but you have a grey halo on the molecule geometry. did you render with straight alpha?

[trythil]

more subdivisions is all you need. the version i saw was sd and the low polycount was very obvious. you also seem to have an issue with colour matting - i'm not sure how you dealt with the alpha or what compositing you did, but you have a grey halo on the molecule geometry. did you render with straight alpha?

The halo was intentional; it was generated in compositing. (Node setup went edge detect -> dilate -> heavy blur -> additive blending, if I remember correctly.) It may have been better if I made it look less like an error and more like it was meant to be there.

All layers were rendered to OpenEXR files with alpha and Z channels, and later combined in Blender's compositor. Somewhere down the line we ended up throwing the OpenEXR sequences in After Effects; I can't remember why.

[trythil]

But the presentation seems rather shallow, especially considering that the basic information that an upscale in resolution does not equal an increase in quality, which is all that was imparted through this thread's original post, is common knowledge.

If that bit plus all its implications are indeed common knowledge, then I am glad that is true. That means that this post is unnecessary after all.

I was under the impression that it wasn't; that additional data point of an opinion was what pushed me to write this.

[blabbler]

heh. you took an exr sequence into ae and forgot to bling the crap out of it with exaggerated depth of field and haze? shame on you sir



=D

[Qyot27]

As far as I was concerned, the point enforced in the OP is that you can't link resolution or frame rate to subjective quality, not necessarily because there's an arbitrary standard such as Blu-ray available, but because the source being worked with doesn't dictate anything - output rendering and encoding params are what matter. Let's face it, the idea probably is well-etched that 'HD' footage has a high bitrate, and thus comparatively as high a quality as the original had or at least close enough, but nothing says it always will.

You could encode 1080p Blu-ray source in Cinepak and it'll look like posterized ass (I actually tested a similar example of this using one of the Star Trek trailers over on the Apple site; took even longer to encode than modern solutions would under sane conditions). Or you could royally screw the pooch by using MPEG-4 ASP or H.264 but set the bitrate way too low and/or smoothing functions way too high and mangle the output. And in such a situation, 480p output without all that damage would look far more impressive and deserving of high marks.

I'll admit, I started out in the era when VCD-level encodes were commonplace, and to be honest, I still think that those 352x240 MPEG-1s can look awesome, even if they are constrained. My eyes are not attuned to judging quality by resolution. I judge quality by artifacting, and even with those old encodes you can get some that look downright flawless.

[Beowulf]

This thread reminds me that I'm going to make a "nostalgia" version of my next video witch will be a fuzzy mpeg-1 at 320x240 resolution, that has been run through a capture card (at high quality).

[Castor Troy]

This thread reminds me that I'm going to make a "nostalgia" version of my next video witch will be a fuzzy mpeg-1 at 320x240 resolution, that has been run through a capture card (at high quality).

Roots?!

[inthesto]

WHY Y'ALL GOTTA BE JOCKIN' ON MY BROTHA TRYTHIL



In other news, Reda is a gigantic fagmo.