Posts made by leeward

gt : Mele Kalikimaka

user : Hauoli Makahiki Hou

I find many people’s views regarding race to be generally very narrow.  Among the developed countries, Americans tend to have the highest levels of ignorance of foreign cultures coupled with a lack of curiosity and disinterest in learning about other cultures.   I read somewhere that most American high school seniors have difficulty finding Iraq on the world map.

David J. Keeling wrote in Global Ignorance – Geographic Illiteracy, Western Kentucky University 2/2003 : “Although most Americans are more aware today of distant lands such as Afghanistan, Iraq, North Korea and Venezuela due to recent media reports on political and economic crises in these countries, few have the ability to find these places on a world map or to discuss critically their cultural characteristics. This lack of geographic awareness permeates all segments of our society, from students to presidents, and from ordinary citizens to senators and businesspeople.”

This is the same kind, depth and scope of ignorance which led to America’s catastrophic involvement in the middle east.  I believe that George Bush did not even know the difference between shiite and sunni, much less cared about it, before invading Iraq.

Rather than accepting responsibility to learn about foreign cultures, most gay Americans appear to be quite bigoted about other gays of different nationalities, quite a poor mark for a subclass of people who themselves demand to be treated more equally by the straights in society.

Obama’s choice of Rick Warren of Saddleback Church in Lake Forest, California, to deliver the  invocation at Obama’s inauguration is a poor one.  And made so early in his term.  This does not bode well for LGBT rights issues, at least as regards support from this Administration.

Add :

Shelter
Trick
Mysterious Skin

@raphjd:

[Obama] has repeatedly said that his religious beliefs won't allow him to support gay marriage.

My belief is that Obama would like to support gay marriage (he is far too intelligent - editor of Harvard Law Review and note his roster of cabinet members and staff - to be too beholden to conservative religious beliefs, and he is from Hawaii, where its State Supreme Court was the first to strike down anti-gay state legislation on constitutional grounds), but he had to get elected with the support of middle America.  And after inauguration, if he is to promote any real change (even if it is on non-gay issues), then he has to govern through the middle.

Bill Clinton tried to make that kind of change in the US military soon after his election, but was shot down and had to settle for Don't Ask, Don't Tell.  He had difficulty getting things done after that, having alienated the middle.

Obama will not make that same mistake.

Regarding the DX50 codec (DivX 5/6) and the Xvid 1.1.2/3 codec used in converting the vob file to an avi container, there is a difference.  In videos shot using high definition cameras, such as the recent ones from Falcon, Bel Ami and RSS, there is sufficient video information within the standard definition format to confirm a visible difference.

In brightly lit, high contrast scenes, the better codec to use is the DX50.  DX50 maintains the sharp, snappy look, keeping the bright white details and minimal staircase edge artifacts.  Xvid 1.1.2/3 on the other hand, shows glare in the high whites and a greater decrease in the contrast range.  The Xvid codec makes what should be a bright outdoor shot look as if we are seeing it through dark sunglasses.  Surprisingly, the same advantage goes for dark scenes.  I find that DX50 maintains the black level details better than the Xvid 1.1.2/3.

Where Xvid 1.1.2/3 excels are in the mid-tone, indoor scenes with evenly lit, studio lighting.  In those scenes Xvid 1.1.2/3 brings out the creamy, high color-saturated look to skin tones.

Hence, you will see that I used the Xvid codec in Cross Country Part 1; whereas in Cross Country Part 2, where there are high-contrast scenes, particularly in the final bright-sun, on-the-boat scene, I used the Divx codec.

Using speedtest.net, stopping all programs and pausing all torrents :

DL : 281.3 kBps
UL : 60.7 kBps

My actual average speeds are

DL : 200-280 kBps approx, varying throughout the day
UL : 75-80 kBps approx steady throughout the day

I have not seen a thread in which Uploaders can voice their opinions about video quality vs file size.  The new rationator rules force me to discover some information as to the kinds of software available for efficient file conversions.

(And for the Leechers who download full-length videos, what are your preferences as to the optimal file size(s) for each video hour.  Do you watch on 15" screens or larger?  What are your preferences for sound quality, MP3 over AC3? Do you care?)

For the Uploaders, what program(s) do you use to convert full-length videos?  What are the advantages and disadvantages?

If you do care to post, let’s have some information:

Program          Min System Req              Supported Input Files                         Supported Output Files

Codec(s) used          Conversion time(s) per original video hour             Freeware or Payware

Output video quality : Resolution (sharpness)
        Color saturation
        Color balance (accuracy)
        Contrast range
        Black level
        Pixelation
        Blocking
        Staircase (edge) artifacts
        Judder artifacts (stutter)
        Panning artifacts

Output Audio quality Frequency range
        Dynamic range

I know, most of you don’t care, and will say “I know video quality when I see it,” just like (male) porn.  Most of you will also say, “What are you cra*y?  I actually have a life.”

But since so much effort is being put into this stuff, let's do some analytical survey.  And you can alert us of any new developments, too - like high definition conversions.

@geekguy:

I will leave it to others more knowledgeable about torrents to confirm the stuff in your post.

So reducing download speed becomes factor only once a full copy of the original file has been seeded.  correct?

No, reducing the download speed is important, if the downloading does not take place linearly - for example, I download parts of the beginning of the file from one seeder and the last parts of the file from someone else, both taking place simultaneously.

When I started this thread, I thought it would be simple, but the more I thought about it and wrote, the more I realized how complicated it is, because we need to know how the system of leeching and seeding works.

This is where Administration steps in, to tell us how the seeding and leeching processes actually work, or point us in the right direction where we can find out.

Thanks for the welcome.

B.  Let’s now break the caveat and assume that UL seeds the beginning of the file to PS and the ending of the file to AJ.  Both PS and AJ have the same download rate.

PS now has the choice of downloading from either UL or AJ or both.
AJ now has the choice of downloading from either UL or PS or both.

If PS has a seeding rate of 1MB/s over AJ’s rate of 25 kb/s, then potentially AJ will have downloaded from PS for a 1MB/s / 100 kb/s = a 10:1 ratio from PS:UL, all other things being equal, by the end of the download.
PS will have downloaded from AJ for a 25 kb/s / 100 kb/s = a 1:4 ratio from AJ:UL.

In this scenario, the bottom line very much depends on the download rate of PS, the ability of PS to acquire as much file data as possible in the quickest time possible, in order to realize the potential of his seeding rate.

In the GT real world, so long as there are leechers and enough of them, PS will eventually reach his seeding potential.

The conclusion would be that PS can add fairness to the AJs by reducing his downloading speed as well as by reducing his seeding speed.

Let’s take another approach, using the simplest three-peer example :
UL - who has received approval to upload the complete 1 GB file and has an upload capacity of 100 kb/s
PS - who has power seeder capability of an uploading rate of 1Mb/s
AJ- who has the average joe uploading rate of 25 kb/s

A. caveat: This example assumes that UL seeds linearly in time, starting with the beginning file data and ending with the last of the 1 GB file data.

For the sake of convenience, let’s assume that both PS and AJ download at the same rate, although usually PS can also download faster than AJ.
PS and AJ start leeching at the same time.
UL now seeds at 100 kb/s to both PS and AJ.
Immediately, UL starts seeding and continues to seed the entire 1 GB file for at least 1GB /100 kb/s = 10,000 seconds or 2.78 hours.
Both PS and AJ download at the rate of 50 kb/s each.

At no point does AJ have any downloaded data that PS could want or need and vice versa and both complete the leeching for 5.54 hours.

Let’s now assume, as is usually the case, that PS can download faster than AJ.
If the system permits PS to download faster than 50 kb/s, then :
1.  At all points in time, PS now has the upload advantage over AJ.  At all times both UL and PS have file data that AJ needs to download.
2.  At all points in time, AJ has no file data that either UL or PS could want or need.
In this scenario, AJ has performed 0 seeding, while PS has seeded to AJ.

In the real world of GT with multiple seeders and leechers, PS very much has seeding advantage to the detriment of the AJs, if PS can download faster than the AJs or if PS has commenced leeching earlier than the AJs.

I have read the thread  "New Ratio requirements" and have seen some of the light.  I came to the new rationator rules as a gambler comes to Las Vegas, seeking to beat the system.  I understand now the system is a community.  As such, my prior post is modified.

Regarding the ideal Xvid Mpeg-4 conversion, is there any software other than FairUse Wizard or Handbrake that will achieve excellent image quality with little noticeable motion artifacts?  I have an issue with FairUse because I read that it accepts only DVDs on disc or an ISO image on hard drives, but not DVDs on hard drives or simple stand-alone vob files.  I have an issue with Handbrake because the conversion time is too long on slower CPUs, like mine.

I have a low-power, single-core CPU, but use a conversion software that specifies a certain output file size.  If I have other programs running, it simply takes longer to convert but the file size remains the same.

I have seen other downloads, using different codecs like Xvid 1.1.0 and they all seem to hover around the .8 to 1.1 Gb per video hour and they all appear to be too soft in image quality at that rate, even if the motion artifacts seem to be minimal.  I find the image softness to be unacceptable.

The exception is the conversion using the Xvid Mpeg-4 codec, which results in a smaller footprint than my DivX 5/6 - Xvid 1.1.2/3 codec, and both the image quality and the decreased staircase motion artifacts are superior to what I have.  But I heard that the conversion time, using Handbrake, exceeds ten hours.  At that rate it is not worth the trouble.  I have checked the bonus points I receive for uploading, and they are so few that I may cease uploading and rely in seeding others' uploads instead.  The few bonus points I receive do not justify the conversion time and upload bother.

With the new rationator rules, are you still recommending a 1Gb per video hour conversion rate?  Do downloaders now prefer sacrificing video quality for the sake of maintaining their ratios?