Video Super Sampling ... SPOT please clarify

Comments

cheroxy wrote on 4/14/2004, 10:23 AM
Spot, just out of humourus curiosity...
if Skevos shows up...are you going to do the supersampling live or the size verifaction of your anatomy?

If the second option is what you were refering to see if Liam or someone can include that into the SCVUG video streams! J/K

;)

Cheroxy
Skevos_Mavros wrote on 4/14/2004, 10:39 AM
Spot said:

1. I did NOT use an MPEG from the site, and the two files now on the tutorial page were created yesterday and this morning from QT files downloaded from the LOC site.

Then please explain how frames containing unique information that is simply absent from the Quicktime (performer positions, scratches, other marks) made it into your larger MPEG if they weren't in the quicktime file you claim to have used as source.

Look Spot, my argument and evidence are clear, and unlike your position my results are repeatable by others.

And for the record, if it turns out I'm wrong, my reaction will be "Oops! My bad! Sorry for wasting your time everyone! Sorry Spot!", and I will then go on with my life. And I repeat my offer made to you in private email Spot - show me how I'm wrong and AU$100 goes to your nominated charity!

2. I did not call any names.

You most certainly did.

I said that "attacking me when I'm not around to defend myself is chickensh**.

This isn't calling names? You're indulging in sophistry now Spot.

I feel that kicking at someone when they're not around is indeed poor form and cowardly behavior.

Nice, so you repeat the original name calling and add calling me a coward to the list as well. I'm just starting to wonder if I've misjudged what kind of man you are Spot.

Tell me, Spot, if I said that putting misleading information on a website and refusing to change it was chickensh** and cowardly, would you feel like I had just called you a couple of names? I think you would.

On the other hand, your post contains false statements. I've not used the MPEGs

That was a suggestion or a theory as to what happened. Of course I have no idea of the truth of how those extra frames got into your MPEG file, I was making a very generous (to you) guess. If you have an alternative explanation I'm all ears.

don't have any of them on my server,

I never claimed you did. I'm suggesting you once had it on your hard drive.

but do have a number of the quicktimes because they were part of a disk we used in a school training situation.

I believe you on that point, but it's a tangent. I wasn't claiming that you used a Quicktime, I was claiming you used, must have used, one of the larger MPEG versions. All I'm suggesting is that you forgot. Or that you mixed up your files. I'm not putting words in your mouth, the suggestions are mine. I'm suggesting explanations for the contradictions I see in your tutorials. Feel free to supply me with better explanations (that make sense and are repeatable!).

Until you do I'm happy for people to read and try out both our explanations and consult the Vegas docs and reach their own conclusions.

Further, that doesn't grant anyone the right to call me "prairie nigg**" "Liar" "Thief" and any number of other names that came both privately and in posts.

Of course, you didn't mean to imply that I said any of those things, did you?

Post my mail to you, Skevos, I've nothing to hide.

Okay, they are posted at the bottom of the temporary web page, for the three people interested! Again, I'll let others decide which emails are the most courteous and professional.

All the best Spot,


Skevos Mavros
http://www.mavart.com
Skevos_Mavros wrote on 4/14/2004, 11:02 AM
(ADITIONAL MATERIAL)It's like standing in a room completely naked, and everybody commenting on the size of my di**. If I had a big di**, some would comment on how I used a patch or something else to make it big. If I had a little di**, people would be laughing at that. But no one else dares take their pants down. And instead, they make comments about how their own di** arrived at a particular size, even posting false, BS webpages about how they arrived at having a big di**.

So, let me see if I understand your dick analogy - when you post a web page making unrepeatable claims it means you have a big dick? And when I post a page making repeatable claims that means I have a small dick? Or I'm hiding my dick? Or attacking yours? :-)

As I said, I'm happy for people to compare our dicks... I mean pages and conclude for themselves who has the bigger... I mean better argument.

All you are doing is adding to the bullsh** by posting what you THINK I did.

I did indeed post a few theories, mainly so I could avoid hearing you claim again that I was accusing you of lying - okay so that idea was futile!. However, I also posted clear evidence, along with repeatable instructions, that allow people to test my observations for themselves. It's simple - the file you claim is created from the Quicktime contains information that is not in the Quicktime. Did you even bother to test it yourself?

Tell you what. At NAB I'm busy as hell. but if you'll find me at the show, I'll do this for you in person, live on the floor.

Nice offer, but I'm in Australia. Tell you what, head on over to the film school where I work, or even my home, and after sharing a beer or two I'll happily show you how what you claim is impossible. Though, in all honesty, I already have.

All the best Spot,


Skevos Mavros
http://www.mavart.com

EDIT: Minor typos/syntax.
DavidMcKnight wrote on 4/14/2004, 11:20 AM
They have a patch for that?





(and if you think I mean software fix, you're not reading the thread closely enough)
mark2929 wrote on 4/14/2004, 11:32 AM
Spot I believe you when you say,

While I may not have explained well, and maybe even have misunderstood the tool,

Perhaps in the thousands of instances you are right This is One instant that may have escaped.

LET It go"...... For now"

Wait untill your home then either make a repeatable experiment OR Learn for yourself where you may have gone wrong.... Its No bad thing...It can only blow out of proportion if you let it. I (Personally) dont care that much One way or the other... But I do care about people. I say forget this Post for now, And come back later ! with a well thought out response.

Dont be hurried or cajoled into defending your position. Right or wrong. Its only a Small part of the BIG Picture and what Vegas and you do.

Skevos I dont think Spot means to be agressive... He possibly feels "Pressured" And this needs all his Concentration WHEN his mind is being diverted by other things. I say give Spot the Time and space he needs to give a satisfactory response.
johnmeyer wrote on 4/14/2004, 11:42 AM
Skevos,

I really appreciate all the time you have taken on this -- several hours of your life it looks like,

The communication of your results has been hampered by a classic case of "multiple conversations." I first heard about this in a class thirty years ago, and it goes something like this:

There are five conversations going on in any communication:

1. What you want to say (your intent, what's in your own head).
2. What you actually said (what the tape recorder picks up).
3. What you thought you said (your recollection).
4. What the recipient hears (affected by noise, acoustics, etc.)
5. What the recipient perceives and remembers (changed by bias, perceptions, mental processes, etc.).

I think somewhere there was a crossed wired where someone got you confused with someone else and started reacting to something that you never intended to say and, in fact, never actually said.
Skevos_Mavros wrote on 4/14/2004, 11:47 AM
mark2929 said:

I say give Spot the Time and space he needs to give a satisfactory response.

Fair enough, that's good advice thanks. And I repeat once more - Spot rocks. On this forum and elsewhere. Spot is the man. A Vegas guru. This rare mistake, and his reaction to it, just proves he's human that's all.

We all make mistakes. Just a couple of weeks ago I was doing some "cable revision" with a classroom full of second years at the film school, and I was several minutes in to explaining some connectors when one student meekly put up his hand and pointed out that I had gotten male and female around the wrong way! Yeesh! (insert size of dick joke here) I'm pretty sure I went bright red. What was weird was that the student apologised to me after class! For spotting a really bad error! As I said to him, it should be me thanking him for providing the correction. Later that night I wondered why only one student spotted my very obvious mistake. Hmmm...

And if nothing else, all this supersampling brouhaha has allowed me to discover a server full of fascinating historical video clips! Check out this guy! :-) Thanks Spot!

All the best,


Skevos Mavros
http://www.mavart.com
Skevos_Mavros wrote on 4/14/2004, 11:53 AM
johnmeyer saidL

I really appreciate all the time you have taken on this -- several hours of your life it looks like,

THanks. Heh - I guess it's obvious that I'm on holiday this week! (though I should be working on some projects).

Nearly 5am here - time for bed!


Skevos Mavros
http://www.mavart.com
jboy wrote on 4/14/2004, 12:16 PM
Boy, I wish Zippy were still around to jump into this thread.
johnmeyer wrote on 4/14/2004, 12:18 PM
And if nothing else, all this supersampling brouhaha has allowed me to discover a server full of fascinating historical video clips! Check out

Was this guy the founder of Cirque du Soleil?
BillyBoy wrote on 4/14/2004, 1:29 PM
Nobody attacked you SPOT, certainly not me. Some merely pointed out the misleading incomplete, unable to duplicate results in your original tutorial and asked you for clarification by reposting parts of the orignal thread that had the original questions you never addressed so you didn't have to go hunting for it.

If that struck a nerve, I'm sorry. You have claimed to be thick skinned, obviously you very much the opposite and overly sensentive to even mild criticism even when that criticism is only meant to resolve a long standing issue that has never really been fully addressed either by you or any Sony representative.

You have said many times, come see what you do it. Well guess what, what you've been doing with regard to supersampling is mostly BS and now your backpedaling away from it as fast as you can.

Cheesehole wrote on 4/14/2004, 1:36 PM
This might be going back a long way for some, but I think everyone should be allowed at least one "GG" moment. ;D

I'm sure I'll have mine someday... Let's just hope Spot doesn't delete all his posts and leave on us!
winrockpost wrote on 4/14/2004, 1:58 PM
BillyBoy,, do you ever read your own posts or do you have short term memory loss,

OK ,Mpeg render by itself did one hell of a "fix" in my tests
Think you nailed it JohnnyRoy,, you may have also Skevos but my brain wore out on me before processing all your info,,, reverse engineering at its finest
JohnnyRoy wrote on 4/14/2004, 2:35 PM
> OK ,Mpeg render by itself did one hell of a "fix" in my tests

And this says a lot for Vegas’ ability to upsample content with good fidelity. I was just trying to be more scientific and only change one variable at a time. So it just made sense to me to upsample to 720x480 MPEG2 as a control test and then upsample to 720x480 MPEG2 with SuperSampling and then compare the two.

The difference is what SuperSampling did. You cannot compare the upsampled 720x480 MPEG2 file with SuperSampling to the original 160x120 Quicktime file and attribute all the results to SuperSampling alone because the native 720x480 upsample did a lot of the work as well.

Any way you cut it, Vegas ROCKS!

~jr
farss wrote on 4/14/2004, 2:45 PM
This is well defined methodology, one used over thousands of years. In the midst of all the hubris it is being used here although at times it's getting a bit clouded.
SPOT has published a 'paper' showing some rather remarkable results. We're not talking here just about a 'how to' kind of thing but a paper that claims some fairly remarkable results, perhaps a bit like cold fusion.
The scientific approach is that a group of his peers attempts to reproduce the results.
Now even IF some of his peers are able to achieve the same results this doesn't necessarily prove the original hypothesis. The hypothesis needs to be tested under other scenarios and if it doesn't hold up then further investigation of the original methodology is needed to resolve why it produced one set of results that cannot be extrapolated.
When you clear away all the hubris this would seem to be what has happened. What hasn't happened though is the degree of scientific detachment required in these sorts of matters. It matters not one iota who made the claim, where they were born, what language they speak or if everything else they've said was bunk. What matters are the FACTS.
The facts as I see them and I'll say upfront I haven't tried any of this myself.
SPOT is able to achieve a result, a repeatable result in front of many witnesses on many occasions.
Others are not able to achieve the same results under slightly different conditions.
Some have been able to achieve the same results BUT deeper investigation has revealed a flaw in how the results were evaluated.
Sony themselves have admitted that a bug exists that would explain all of the above.

This is good science, no reputations should be harmed by this. This process happens 100s of times a day in the scientific community. Where reputations get harmed is when we let personalities get in the way of facts and logic.
All I'm getting at is this could have been resolved with far fewer posts, much less angst and in a more dignified manner if personalities, grandstanding and speculation had been kept out of the discussion.

Bob
Cheno wrote on 4/14/2004, 4:00 PM
All I can say is I say Spot do this at the LA VASST last year.. took so much time to render we saw only a small portion, but low and behold, we were watching him do it. About 30 others saw it too.

I say we let it die... this string takes too long to load :) Lets talk about the Hawaiians flying Krispy Kreme's from island to island.. that's pretty funny! Of course I remember driving 30 min and waiting for 1 1/2 hours for 4 dozen when the first one opened in Utah.. hehe
Cheesehole wrote on 4/14/2004, 4:26 PM
farss> What hasn't happened though is the degree of scientific detachment required in these sorts of matters.

I'm sorry farss, I don't quite understand. Are you disputing Skevos's results? His analysis was dead right. You can verify it yourself if you want. If you have a point to make about his science, I think verifying his results would be more effective than simply implying it wasn't good enough because of personal bias. I think given what he was dealing with he did a remarkable job of sticking to the "facts".

I totally agree that what should have happened, did not happen, and has not happened. Personally, I would take one look at Spot's tutorials and say, "no way". But I can imagine someone who isn't as knowledgable about the science of image processing might take it for granted that they were accurate. I know there is probably more misinformation than accurate information out there on the web, but when it is about Vegas for some reason I find it offensive.

Those who wish to let this thread die - lobby Spot to fix the tutorials. As long as they are out there propagating false information this issue will not go away.

For those who don't understand what is going on, read Skevos's temporary page.
JohnnyRoy wrote on 4/14/2004, 4:53 PM
> All I can say is I say Spot do this at the LA VASST last year.. low and behold, we were watching him do it

You’re missing the point. I can demonstrate this live in front of a room full of people just as I did at home for myself and give the ”illusion” of the impossible because the preview will be pixilated garbage and the final result will be miraculously clear. As long as I don’t resize the original preview window to 720x480 (at which point the original file will no longer be pixilated and show the true details) no one will be the wiser. But that would be deceiving to do knowing what we know now. (I don’t think it was Spot’s intent to deceive)

The remarkable results are due to the way Vegas displays Quicktime files in preview mode. NOT because SuperSampling can make clear letters out of a blur of pixels which is what most of the arguments on this thread are about. I don’t think Spot realized what was going on at the time and was only reporting (and posting images of) what he saw. Now that we’ve pointed out this flaw, however, he probably should fix the tutorial because it is misleading.

An accurate portrayal would be if he rendered the Quicktime file to 160x120 uncompressed AVI file first, and show a snapshot of that in the Vegas preview. Then he would be showing a true comparison of before and after. At this point he could claim what both Vegas resize and supersampling can do.

In order to just show what super sampling can do, however, I would also show an unprocessed 720x480 MPEG2 file and a supersampled one just to make the comparison apples to apples at the same resolution (i.e., resize both with and without supersampling). The results will be subtler at that point.

Warning... Scientific method joke:

A scientist wanted to measure the results of amputating a frogs legs.
He yelled JUMP at the frog with four legs and the frog jumped 9 feet.
He cut of one leg and yelled JUMP and the frog jumped 7 feet.
He cut off a second leg and yelled JUMP and the frog jumped 5 feet.
He cut off a third leg and yelled JUMP and the frog jumped 3 feet.
He cut of the last leg and yelled JUMP and the frog didn’t move.
Once again he screamed at the top of his lungs JUMP and the frog didn’t move.

Conclusion: By amputating all of its legs the frog has appeared to have gone deaf!

;-)

~jr
jazzvalve wrote on 4/14/2004, 5:37 PM
SPOT showed it in quick time and vegas both i think. i saw it to but what do i know? i'm still a beginner. but it looked terrible in quicktime and looked great when vegas was done.
L25 wrote on 4/14/2004, 10:06 PM
rarely did zippy jump into a thread, more likely zippy would hijack the thread, but usually just troll the thread....
farss wrote on 4/14/2004, 10:12 PM
Cheesehole,
NO I'm not disputing anyones results at all. My comment was directed who think that just because an experiment fails or used incorrect methodology that it is a reflection on the person.
From a scientific point of view an experiment that gets the same results is actually rather boring, it just shores up what we already expected, it's when fails and we have to go back and change our theories of methodolgies that we advance our knowledge.
Of course history is full of people who stuck to their guns as though it was their reputation at stake and that's just silly.
Quite oftenly two conflicting results seem correct, that's an indication that deeper investigation is needed. I do think that has happened here which is great. But that doesn't make anyone right or worng. It means their methodolgy was wrong or something was overlooked.
What I'm objecting to is the attack on personalities rather than ideas.

BTW if nothing else it proves how great this forum and the user community is, I've yet to see anything like this level of investigation on another products forums.
Cheno wrote on 4/14/2004, 10:35 PM
"I've yet to see anything like this level of investigation on another products forums."

Amen... Skevos and Cheesehole win the "Sony Fine Tooth Comb" award!
TorS wrote on 4/15/2004, 12:24 AM
I urged for calmness when this thread was still young because quite a few of you people have rather harsh or indelicate ways of putting your truths across. Winning the argument seems more important than to reach a common (and true) understanding of the topic. That is a very unscientific approach, whether you’re right or wrong.

Supersampling was there for nearly six months and hardly any of us knew about it until Spot published the tutorial. Thank you for that, Spot. And thanks to the guys who saw that something about it was not right, and put in a lot of work to correct and to clarify. I needed that.

Usually inaccuracies or professional disagreements are dealt with in two or three posts here. Circumstance plus the fact that Sony’s documentation seems uncommonly vague made this one last a lot longer. Okay, we can live with that. I just hope that no one is hurt, nor hardened in the process.

This thread can die now, but if you want to take it further, think about the common man. Instead of proving each other wrong, prove yourself right in such a way that I and the silent majority can understand and benefit from it. Your opponents may make a lot of noise, but we (the people) are here in greater numbers, and we listen.
Tor
RexA wrote on 4/15/2004, 1:29 AM
>Instead of proving each other wrong, prove yourself right in such a way that I and the silent majority can understand and benefit from it.

I think that was the original intent in the thread and I think it happened. Several people who ultimately provided the most descriptive information, approached it only with the desire to understand how the features really worked. I also think, if one was paying attention to the real (technical) subject, that there was much information gained.

Along the way some people got upset, but it seems to me that there was benefit for the majority. A little perceived or real controversary in the process is just the way life works. This was a difficult and subtle issue that has been addressed before and never adequately closed.

Let's hope we never stop trying to find our way to real answers just so no one gets offended.

In this thread, it got messy at times along the way, but I think ultimately real answers were presented for anyone who listened with an open mind. Sorry that one person seems to have taken it as an attack.

I'm glad we had a few people willing to keep at the real heart of the subject until some new answers were generated. For them, I offer my thanks.