Compression Discussion

[Nded]

Re: User-requested enhancements & features?

I'm debating semantics with you!

Anybody who complains about Jon's rants is guilty of antisemanticism.

[HeadHodge]

Re: User-requested enhancements & features?

No. Hodgy will still be complaining about something then!

Hodgy - I love ya man!

Ditto.

I'll keep complaining until the world is as perfect as me!! I thought I made a mistake once, but I was wrong!!

[HeadHodge]

Re: User-requested enhancements & features?

The VOD services are more about compromise and convenience. I'm not expecting it to be as good as a BD/HD DVD, but in sacrificing that I get instant viewing of a title, no waiting. And I think that is a decent trade off.

I'm not expecting VOD to provide a lossless reproduction of the source content. Because, I just don't think with todays technology, it's feasable.

But what I do think is feasable and should be done is for someone to come up with a scientific (not subjective) metric or benchmark that measures the amount of loss that is introduced from transmission of the source content to the resulting destination content.

With some benchmark like that, we could at least put a stake in the ground and be able to set a goal and be able to measure the progress of acheiving 0% loss as technology improves.

Such a benchmark would be totally independent of how the content is transmitted or how large the destination content is, it would simply compare the loss of the resulting destination content from the source.

[Greg]

Re: User-requested enhancements & features?

Hey Hodge,

You do know that most of your questions have answers all over the internet right? (deep answers too, that take days to come up with )
Have you developed an allergy to search engines?

One of the first google results leads directly to http://pdiff.sourceforge.net/ .
The "video compression" search into wikipedia takes you directly to http://en.wikipedia.org/wiki/Video_compression ... which is very informative and has a link to your holy grail: http://en.wikipedia.org/wiki/Subjective_video_quality

-- Greg

[redwein]

Re: User-requested enhancements & features?

Why I am being specific in saying that there are no "lost" pixels has a purpose. The poster who kinda started this discussion suggested that Vudu's HD content might not really be HD. In reality that would mean that the pixel count of the content is not 1024 x 768 or higher or as in Vudu's case 1920 x 1280. The fact is that there is always 1920 x 1280 pixels. I felt that your statement of "losing" pixels indicates that maybe the initial 2,457,600 pixels is some number less than that. My point is that the resolution in pixels of the picture does not degrade. The quality of the image degrades but the pixel count stays the same.

I'm debating semantics with you!

Technically then, you could replace a vivid high detail image with a full pixel count white sreen and claim no "loss" of pixels. I still believe the terminology I used was correct. What I was trying to get across to that guy was that what Vudu does is far superior to an upscaling DVD player since they have to fabricate pixels out of nothing. Intelligent compression starts with a perfect source and renders it somewhat less perfect but in a very intelligent way so as to retain detail where it is important and allow it to be lost when it is less so. That's what I was trying to say.

[HeadHodge]

Re: User-requested enhancements & features?

Hey Hodge,

You do know that most of your questions have answers all over the internet right? (deep answers too, that take days to come up with )
Have you developed an allergy to search engines?
-- Greg

All I know is that everytime I go to the Rest, I come back to the Best. You!!!

(Thanks for the links)

[HeadHodge]

Re: User-requested enhancements & features?

One of the first google results leads directly to http://pdiff.sourceforge.net/

What did you enter for your search, Martinis or Meds??? I get something different. It's usually about someone called Timothy Leary.

[HeadHodge]

Re: User-requested enhancements & features?

Intelligent compression starts with a perfect source and renders it somewhat less perfect.

There you go with the Oxy-Morons again!!!

[HeadHodge]

Re: User-requested enhancements & features?

Hey Hodge,

You do know that most of your questions have answers all over the internet right? (deep answers too, that take days to come up with )
Have you developed an allergy to search engines?

One of the first google results leads directly to http://pdiff.sourceforge.net/ .
The "video compression" search into wikipedia takes you directly to http://en.wikipedia.org/wiki/Video_compression ... which is very informative and has a link to your holy grail: http://en.wikipedia.org/wiki/Subjective_video_quality

-- Greg

I quickly purused those links, but it still boils down that you have the source and destination content to compare and I don't. I only have the destination, so no matter how much time I spend on this, I will only be able to speculate.

So I guess I would call this this a "User-requested enhancement or feature" for VUDU to provide some sort of non-subjective video loss metric for it's service.

But I have a feeling it would be easier for me to fly to Thailand and back for a fun day on the strip than to get that!!

[Rusdude]

Re: User-requested enhancements & features?

Having a third tier of content that would be on the level of Blu-Ray/HD-DVD versions had been discussed previously (probably more than once too). Ideally, I'd like to see the day when we can get exactly duplicates of high-def discs (with multiple audio tracks, subtitles, extras, etc.) on demand. However, that's probably way off, but I don't see a problem with having a choice between Instant HD vs. BR-Quality Delayed HD.

A Blu-Ray quality movie is probably going to be, what, 25-40 GB once you remove all the extra stuff and keep high-bitrate video + DD5.1 soundtrack that Vudu can use, right? So even if that means waiting 6-10 times longer, it's still going to be quicker than doing Netflix. Plus, with 30-day rental windows, we can cue up the downloads and have 10+ movies available ("+" for when we get to expand storage using external HDDs).

Bottom line is that just like Rental vs. Purchase debate depends on personal preferences, so does Instant HD vs higher-quality delayed HD. And as long as a consumer is given a choice between those options, then I see no harm in having third tier of quality.

[redwein]

Re: User-requested enhancements & features?

Well it would take extra resources for Vudu and complicate the UI just for starters. I think, from a strategy standpoint, it makes perfect sense to pursue the instant gratification angle and become the king of it. Waiting 12-20 hours for a movie (using your 6 to 10 times longer guess) is certainly nothing I would ever do. I would guess most others wouldn't either, given the excellent quality available instantly.

The main benefit of Vudu for me is that I can decide what I want to watch when I want to watch it. That defines Vudu for me and I think that is what they are pushing as their reason for existance as well. My mood isn't static and I really don't know beforehand what it is that I will watch. I want Vudu to focus 100 perent of their effort on that.

[HeadHodge]

Re: User-requested enhancements & features?

which is very informative and has a link to your holy grail: http://en.wikipedia.org/wiki/Subjective_video_quality

-- Greg

I bought you a copy of this for your birthday.

VQLab 1.1

[NA9D]

Re: User-requested enhancements & features?

Technically then, you could replace a vivid high detail image with a full pixel count white sreen and claim no "loss" of pixels. I still believe the terminology I used was correct. What I was trying to get across to that guy was that what Vudu does is far superior to an upscaling DVD player since they have to fabricate pixels out of nothing. Intelligent compression starts with a perfect source and renders it somewhat less perfect but in a very intelligent way so as to retain detail where it is important and allow it to be lost when it is less so. That's what I was trying to say.

Very funny on that first sentence! Technically yes.

But I agree with you 100% on the rest there. That's correct!

[HeadHodge]

Re: User-requested enhancements & features?

I think you are wrong. Just look what happens with the blockiness of a low detail image. That seems to suggest that there is a bigger delta than a high detail scene where you can read fine print in a newspaper.

Here's another possible explanation to prove that you haven't yet proved I am wrong.

SSIM

SSIM is a more complex metric based on different properties of the Human Visual System (HVS) that is starting to replace PSNR as the most widely use metric. It estimates the quality of the video much better (a higher correlation with subjective assessment) by subtracting information from the frames like luminance, contrast and structure information, and comparing those estimated values instead of directly comparing pixels. The metric is based on the fact that the HVS is more sensitive to structural changes in videos than to luminance and contrast changes.


The HVS (Human Visual System) behaves in a similar way, being more sensitive to changes in the darker areas of a picture.

[redwein]

Re: User-requested enhancements & features?

Ok, Here is what I think is proof. If you take a solid image (every pixel is the same), compress it with a lossy format then decompress it, you should get the exact same image back since there is no reason for the decompression algorithm to invent any new pixel values. Obviously this will only be true for a very small number of images or the compression would be lossless or very near lossless by definition.

As long as there is one case where the data loss differs between 2 images, it would be proven that the rate of loss is not constant.

BTW, I just did an experiment. I went into MS Paint, created a solid color image and saved it as a tif file. I then reopened it and saved it as a jpg file. I then reopened that and saved it as another tif file. The first tif file and the second were identical. I then opened a photograph in paint and saved it as a tif file. I then opened that and saved it as a jpg file. I then reopened that and saved it as a tif file. Those 2 tif files had significant differences at the byte level. I was also able to visibly see differences when I looked at the 2 tif photos but obviously saw no difference in the solid image.

Good enough as proof?