Jump to content


Video-Film Challenge


  • Please log in to reply
30 replies to this topic

#1 Charlie Seper

Charlie Seper
  • Guests

Posted 02 June 2005 - 08:25 PM

Starting Page One

There are 10 web pages linked together here. The url is for the first page and each link takes you to the next page in order. When you get to page ten it'll link back to the first page where you can start over if you like.

There are four frame grabs on each page. That's a total of forty stills. Of those forty, fifteen are from video. Your challenge is to pick out the fifteen video stills. All photos are the same size and were gathered pretty much the same way. All use the exact same jpg compression. They look nearly identical to the DVD's. I grabbed them at full screen with PowerDVD (17" monitor) and then brought them down to 960 x 540 using Ulead Photo Impact which also has the best jpeg compressor I know of.

Post your answers as:

Page 1?photo C

Page 1?Photo D

Page 4?Photo D

And so forth.

Everything is as random as I could make it. There may be any number of video stills on any page. I didn't use too many faces and I tried to stay away from known actors for the most part when I "did" use a person's face

I'll post the answers here next Thursday, May 16th.

Who knows, maybe some of you will surprise me.

PS, If you recognize anything here from a movie (I doubt if you will) please keep it to yourself until after next Thursday.
  • 0

#2 jeremy edge

jeremy edge
  • Guests

Posted 06 June 2005 - 03:44 AM

Starting Page One

There are 10 web pages linked together here. The url is for the first page and each link takes you to the next page in order. When you get to page ten it'll link back to the first page where you can start over if you like.

There are four frame grabs on each page. That's a total of forty stills. Of those forty, fifteen are from video. Your challenge is to pick out the fifteen video stills. All photos are the same size and were gathered pretty much the same way. All use the exact same jpg compression. They look nearly identical to the DVD's. I grabbed them at full screen with PowerDVD (17" monitor) and then brought them down to 960 x 540 using Ulead Photo Impact which also has the best jpeg compressor I know of.

Post your answers as:

Page 1?photo C

Page 1?Photo D

Page 4?Photo D

And so forth.

Everything is as random as I could make it. There may be any number of video stills on any page. I didn't use too many faces and I tried to stay away from known actors for the most part when I "did" use a person's face

I'll post the answers here next Thursday, May 16th.

Who knows, maybe some of you will surprise me.

PS, If you recognize anything here from a movie (I doubt if you will) please keep it to yourself until after next Thursday.

<{POST_SNAPBACK}>


hmmmm... best jpeg compressor?
Well I'd say if you really wanted people with a good eye to figure it out you'd have to provide more detailed photos .The distortion from the jpegs make it impossible to see video noise or grain ,which would be giveaways to whats what. A lot of the photos look a little blurry and digitized ...I'd say this is a pretty useless test.

The major benefits of using film over video is its high resolution and its frame rate and motion. followed closely by its dynamic range.none of these stills can really put that to the test. They're not clear and of course...they dont move.

If you want to know" Can I fool someone that doesnt know this is video with my dvd player on pause through a small monitor ?" Then you dont need a test ...the answer is yes you probably can.

Now fooling people with moving images on a larger or high resolution display would be a harder task.And no doubt JVC and Panasonic are making good gains to that end on the prosumer end of things. But its been long established by many that even super 8 has a multitude of advantages over standard 60i DV.Much less a 35mm feature you have on DVD!

I have an idea for a test : Why dont we see if we can tell George Lucas from Steven Spielberg from looking at them from behind a stained glass window?

Sorry for the sarcasm....but this video vs film thing really has been beaten to death and really doesnt deserve any more kilobytes.
  • 0

#3 drew_town

drew_town
  • Sustaining Members
  • 383 posts
  • Camera Operator
  • Southeast US

Posted 06 June 2005 - 09:49 AM

While Jeremy brings up some good points, I'll post the stills that I think are video. Honestly to do a good test you need to avoid using jgeg or any other form of compression.

Page 1A
Page 2C
Page 3B
Page 4A
Page 4D
Page 5B
Page 5D
Page 7A
Page 8D
Page 9A
Page 9B
Page 10B
Page 10D

Without a doubt this thread will turn into a shouting match before too long. So I can accept the fact that some of my guesses will be wrong and I'll leave it at that. Some of the images you have posted just aren't good images in terms of focus, exposure, and composition so those were most difficult to distinguish. Obviously I consider some of the images to be quality video images mainly because I was only able to list 13.
  • 0

#4 Tim Tyler

Tim Tyler

    Administrator

  • Admin
  • PipPipPipPip
  • 1291 posts
  • Cinematographer
  • Olympia, WA (US)

Posted 06 June 2005 - 01:06 PM

Without a doubt this thread will turn into a shouting match

It better not.

Those JPG's are only 20k each which is ridiculously small for any quality test.

Recompress them at 150k or higher (from the originals) if you?re serious.
  • 0

#5 Patrick Neary

Patrick Neary
  • Basic Members
  • PipPipPipPip
  • 873 posts
  • Cinematographer
  • Portland, OR

Posted 06 June 2005 - 01:50 PM

not to throw a wet blanket on this, but if these are all frame grabs from DVD's, then they're all video...and heavily compressed video at that!
  • 0

#6 Charlie Seper

Charlie Seper
  • Guests

Posted 06 June 2005 - 02:35 PM

Not bad Drew. You missed six.
  • 0

#7 Charlie Seper

Charlie Seper
  • Guests

Posted 06 June 2005 - 02:53 PM

hmmmm... best jpeg compressor?
Well I'd say if you really wanted people with a good eye to figure it out you'd have to provide more detailed photos .The distortion from the jpegs make it impossible to see video noise or grain ,which would be giveaways to whats what. A lot of the photos look a little blurry and digitized ...I'd say this is a pretty useless test.

The major benefits of using film over video is its high resolution and its frame rate and motion.

<{POST_SNAPBACK}>


Well, you're kind of making my point Jeremy. I've always said that the big difference between film and video was one of motion, and of clarity on a 30' screen. I also would add depth of field to that.

I think shooting video at 24P takes care of the motion. For the most part I think that a TV broadcast done on video at 24P such as "Joan of Arcadia" looks just like film. Of course that's on my standard def TV. And generally speaking, a good high-end hi-def cam will have a lense with decent DOP.

So the big difference between film and video is the amount of resolution when spread out on a 30' screen at the theatre. My comments have always been that video can look just like film--only it'll look softer (i.e.--less focussed) on the big screen.

As to the jpeg compression, I think you think to much of your eyesight abilities. I can see no difference between the original screen grabs and the compressed jpegs. Now if I printed them out...sure. On on a computer screen at 96dpi? Nope. I'll be happy to show an original at full resolution to compare though, but you'll have to wait a few days. I've gotta split for a while.

PS, I've also never been able to see any difference between the 4-2-2 pics taken by my still camera and a 4-1-1 compressed copy of it when viewing on screen. Again, if I print them out, then sure, but not on screen. And I don't think theatre projectors make films look any better. At least they shouldn't given the resolution of most of them and the distance to the screen.
  • 0

#8 drew_town

drew_town
  • Sustaining Members
  • 383 posts
  • Camera Operator
  • Southeast US

Posted 06 June 2005 - 05:36 PM

PS, I've also never been able to see any difference between the 4-2-2 pics taken by my still camera and a 4-1-1 compressed copy of it when viewing on screen.

<{POST_SNAPBACK}>

The difference comes into play when you start color correcting and keying. Less compression makes it a whole lot easier, and to some extent, possible. More resolution helps a lot too.

7 out of 13 huh? A few of them were really obvious, others weren't. In closeup it's harder to tell because you see a lot of the detail that would get lost in a wider shot. It's one of the techniques for shooting SD video for the big screen.

I want to see someone pick out at least 10. Come on. I know it's not ideal, but if I can find half of them some of you guys can find the rest.
  • 0

#9 jeremy edge

jeremy edge
  • Guests

Posted 07 June 2005 - 11:27 AM

"I can see no difference between the original screen grabs and the compressed jpegs. "

You can't?
look for the little blocks here and there....they almost have a multicolored tint to them.There actually quite large....bigger than the pixels on ntsc set and much bigger than film grain or video noise. You need to get rid of that to make a good judgement.

Go to filmshooting.com and look at some of the screen grabs from k40 on super 8.
Some are real grainy others look better than a few of the stills you posted.

You don't have to have golden eyes to know that 24p mini-dv is no comparison to 35mm film. You might have a fighting chance with super 8 but I still think film would win.

"Joan of Arcadia" I'm sure is not shot with a dvx.Although I think some of "Pimp My Ride" and some of "Chapelle's show' is.Some good examples of good looking mini-dv.

You have to understand that people cant go around saying that dv can look as good as film otherwise the manufacturers will get complacent. The 2 new cams from jvc and panasonic are a great step forward I bet ...but we have yet to see video equal film and if thats where things ae headed we should not be satisfied till it truly does.
  • 0

#10 Rik Andino

Rik Andino
  • Basic Members
  • PipPipPipPip
  • 783 posts
  • Electrician
  • New York City

Posted 07 June 2005 - 04:45 PM

I was bored so...

1D
2C
3B
4A
4B
5C
6C
7A
7D
8B
8C
9B
9C
10B
10D


After starting out I quickly realized how difficult this is...

1)
The images are not moving...it's difficult to judge the stills...

2)
The images are highly compressed...so you can't really see the quality...
In fact many images were extremely blurry to tell what it was.

3)
It's hard to concentrate when you're blasting the Foo Fighters music :)

So I conclude that looking a low-res stills cut from DVDs
Isn't the best way to judge the differences between Film and Video...
But I did it anyways....so how I do?
  • 0

#11 Annie Wengenroth

Annie Wengenroth
  • Basic Members
  • PipPipPipPip
  • 653 posts
  • Sound Department
  • Brooklyn, NY

Posted 07 June 2005 - 08:22 PM

I was blasting Ministry...
  • 0

#12 Sam Wells

Sam Wells
  • Basic Members
  • PipPipPipPip
  • 1751 posts
  • Cinematographer

Posted 07 June 2005 - 11:35 PM

They cancelled "Joan of Arcadia"

(See if they'd only shot it on Kodachrome............... :lol: )

-Sam
  • 0

#13 Jonathan Spear

Jonathan Spear
  • Basic Members
  • PipPipPipPip
  • 586 posts
  • Other

Posted 08 June 2005 - 07:38 AM

"Ministry"

Hell yeah B)
  • 0

#14 Charlie Seper

Charlie Seper
  • Guests

Posted 13 June 2005 - 04:08 PM

You missed 5 Rik.
~~~~~~~~~~~~

Yeah, I heard they cancelled Joan of Arcadia. Its a shame really. It was an especially good show for young people I thought.
~~~~~~~~~~~~

Jeremy,

If any of these stills look blocky its because they did on the DVD. But how else to judge the dif between the two? Even if I had these same still grabs from a hi-def DVD (which I don't think are out yet), you still wouldn't see any difference on a 96 or 72 dpi computer monitor. Not unless you doubled the size of the image or something. And on a hi-def Plasma screen using hi-def image sizes (720p or higher) then it'd be no contest. Obviously the miniDV format is done for going against hi-def on a hi-def viewing format. But that's not what I'm arguing. I'm saying that on a standard def TV, or for that matter, on a 30' theatre screen (which is actually a standard def format when you consider the projector image ratio and the distance it must cover to get to the screen, etc.) that video can look almost exactly like film if shot in progressive frames at 24fps. This is true whether its hi-def or standard def video. I would also say that part of this equation calls for consideration on part of the film projector since they can vary in quality greatly from theatre to theatre. My argument has always been that video can look just like film but that in a high quality viewing envorornment the video will probably look less detailed.

There's an awful lot of hype on the part of film industry people as to what film is really capable of. And as video gets better and better, the hype will get louder and stronger. As far as I'm concerned, the sooner the gravy train ride is over for the fat cats in the film industry the better. They've been socking it to people ridiculously for several decades now. Think about it. $10,000 for a measly bit of glass on the front of their film cams that costs them a nickel's worth of sand and a few hours grinding time. The purely insane cost of developing film. (The chemicals can't possibly cost "that" much). The cost of film stock itself. They've inflated their prices to the point of loan shark rates. They certainly didn't have our best interests in mind. I hate to say it because I've got a good customer that's a retired exec from Kodak. He's a good guy but one of a few.

Look, I came from the audio end of things. I've owned a recording studio near St. Louis for most of my adult life. And as you may know, the digital revolution hit there over a decade ago. And there was the exact same silly hype from analog tape people. "Digital recording will never sound like a good reel to reel tape". Most of the studios that still have reel to reel decks around these days only use them to remaster old material, or to dub it to a DAW unit. Almost nobody records analog anymore. But what was worse was the way these idiot people who either owned or worked off commission in music retail stores would come into these dopey Internet forums for music recording claiming how this product was crap or that one and how you had to buy something REALLY expensive instead. They created a silly amount of hype in order to sell more expensive products. Ask yourself this: If you were an unscrupulous person working off commission sales in a music store, say Guitar Center for instance, and you could make a $100 commission off a POD amp modeler that sold for $500 or a $15 commission off a Berhinger V-Amp that had 95% the same sounds, but was made cheaper in Germany, wouldn't you go into stupid Internet forums and start blasing the hell out of Behringer products. Because that's exactly what they did. But we're wise to them by now.

Some of you really need to think more about the whole film verses video thing and ask yourself who's really benefitting from all this film hype. MiniDV may not quite be up to film yet and likely never will be. But hi-def video looks so good that no one could tell the difference. And now we're waiting to see what HDV will be capable of. The stills I've seen from the Sony units look marvelous. The HD100 from JVC looks to be much better yet. Personaly, I'm tickeled pink.
  • 0

#15 jeremy edge

jeremy edge
  • Guests

Posted 13 June 2005 - 08:06 PM

I actually use the v-amp!
You can hear it at myspace.com/backstreetlaw

I dont think the tech is good enough yet to stop using tubes altogether though!

Here's my point.....
SACD and dvd audio came too late.
We settled for 16 bit cd and now we have people settling for mp3 and mini disc.

If we let the manufacturers know that we think video is "Close enough" to film then the camera manufacturers will get lazy.Believe me i want a video camera that will do the same job as film.

If someone handed me the money I'd by the new jvc.

But I kinda feel like these new camcorder technologies are overpriced .(not talking about the pro lens etc just the tech on silicon)
Show me a sub $1000 that does 24p,true 16:9 , good manual focus etc and I'm in.
maybe Behringer should get into THIS business too!

I just let my buddy borrow a k3 for 2 weeks.He owns an xl1.he thinks the footage we've seen so far rips apart the xl1 in standard definition!

Tell you what, post a short (can be really short) and lets compare it to some of the shorts done in 16mm on this board and see how many get fooled. You'll have me convinced then that were are finally there and kodak needs to close up shop.
  • 0

#16 jeremy edge

jeremy edge
  • Guests

Posted 13 June 2005 - 08:15 PM

For the record I'm one of those crazy people that think Analog is still the bomb.

If I ever can...I'll record on 2 studer 24 tracks and mix in pro tools hd.

We recorded on adat. ADAT sucks!!
When that came out you were probably throwing out the tape machines.

Only now with the 24/96 machines do we have something that really competes with analog.

But you still need a ton of outboard gear to try to emulate the natural compression of analog tape.

Sorry for the off topic.
  • 0

#17 Joshua Provost

Joshua Provost
  • Basic Members
  • PipPip
  • 71 posts
  • Cinematographer

Posted 14 June 2005 - 12:26 PM

I'm saying that on a standard def TV, or for that matter, on a 30' theatre screen (which is actually a standard def format when you consider the projector image ratio and the distance it must cover to get to the screen, etc.) that video can look almost exactly like film if shot in progressive frames at 24fps.


Charlie,

Have you ever seen SD projected on a 30' theater screen? Or even a 20' theater screen, for that matter? I have, and it's pretty bad, even progressive 24p SD. There is no contest. SD can hardly stand up to the largest rearprojection home units, nevermind large theater screens.

Also, a computer LCD screen is the ideal way to critically compare the images. It is progressive, of better quality than any LCD TV set, and won't be doing any rescaling or resizing like a set would. It's the best way, not the worst. However, heavily compressed JPEG images aren't a valid test.

Josh
  • 0

#18 Charlie Seper

Charlie Seper
  • Guests

Posted 16 June 2005 - 02:12 PM

Jeremy,

I've also been using two Behringer mixers for a couple of years with no problems what-so-ever. They're dead silent and sound great. I also just got one of their stereo compressors--Composer Pro-XL. But I've got a very expensive DBX-160 mono unit that runs around $500 that I've had for many years. This is the one people rave about so much. Its a fine compressor but I've been doing a lot of stereo miking acoustic guitar the past year for some solo fingerstyle stuff and needed a stereo compressor. The Behringer units are only a little over a hundred dollars so I figured it couldn't hurt to try one. It honestly sounds as good as my DBX unit. It's automatic functions aren't too great, but as long as you know what you're doing and can set things up manually it sounds unbelievably good. I post some settings, sound samples and whatnot over at Cakewalk's site for the young guys now and then. I just did one for acoustic guitar miking in stereo along with the settings I use for this compressor.

Stereo Mic Technique Thread

If you keep an eye on the same forum I'll be posting some settings for the Mesa Boogie MarkII in the V-Amp2 after the first of the month.

I would have to disagree with you about 24/96 recording though. The only thing 24 bits give you is a tiny bit more headroom that most people will never use unless they're recording a symphony with some steep volume dynamics, and it gives just a slight bit more stereo seperation. But if you dither down to 16-bit afterwards you lose that extra stereo seperation. With DVD audio you can of course keep things in 24-bit and keep the stereo seperation too. Its hardly worth it though. You have to be wearing headphones or have your head right between the speakers to notice it because its very slight. When 24-bit first came out, people were hearing that extra bit of stereo seperation and fooled themselves into thinking that their tracks sounded better sonically. Tell you what though: Record something in mono at 24-bit and then record the same thing in mono at 16-bit and then tell me you hear a difference. There isn't any. And very few people record at resolutions higher than 44.1 or 48. It just doesn't sound any different at all.

24-bit does have a few advantages though. You can make several more changes to a 24-bit wave file without degrading it than you can with a 16-bit file. Although I can make at least 5-changes (effects, hard limiting, normalization etc.) to a 16-bit file before I notice any damage to the sonics. What a lot of us do though is to record all our tracjs at 16/44.1 or 16/48 and then mix down to a 24 or 32 bit file. This is a lot easier on your machine and allows you almost unlimited tracks and effects with a decent computer. And its the stereo mixdown file that undergoes most of the changes so we just bump it up to 24 or 32 bit before we dicker with it. The basic tracks don't generally undergo much editing to speak so 24-bit isn't really needed.

24-bit has a better noise floor, but we already had a s/n of over 90db in 16-bit, which is so close to dead silence that no improvment in s/n ratio can be heard.

When the first 24-bit workstations first came out a friend of mine who worked at MacMurray Music in St. Louis told me he couldn't hear any difference. Neither could I. We both thought it was quite a coinsidence that after recording in 16-bit for over a decade, music manufacturers suddenly decided we needed 24-bit. The fact is that retail sales in music workstations had dropped to almost a halt because everybody already had one. They had to do something to pick sales back up, thus 24-bit workstations....

Here's a 16-bit recording (wma file) that came out of my studio 2 or 3 years ago. I played everything except the drums. And the vocalist is awful! But I defy anyone to find any noise, distortion, or unwanted artifacts anywhere on it. It has around 28-tracks I think. And by the way, we were just horsing around with computer recording at the time. This was done on one of those little SBLive cards that people said so many bad things about. There was a lot of flack about SB cards for a few years (which was mostly Creative's own fault for bad advertising and not giving enough info about the cards), but if people had been using those cards correctly they would have sounded as good as this. We had "yet another" thread about the SB cards at Cakewalk just a month ago. I'll post the url to it below the music url.

Me and My Arrow

Cakewalk SBLive Thread
  • 0

#19 Charlie Seper

Charlie Seper
  • Guests

Posted 16 June 2005 - 02:28 PM

Charlie,

  Have you ever seen SD projected on a 30' theater screen?  Or even a 20' theater screen, for that matter?  I have, and it's pretty bad, even progressive 24p SD.  There is no contest.  SD can hardly stand up to the largest rearprojection home units, nevermind large theater screens.

  Also, a computer LCD screen is the ideal way to critically compare the images.  It is progressive, of better quality than any LCD TV set, and won't be doing any rescaling or resizing like a set would.  It's the best way, not the worst.  However, heavily compressed JPEG images aren't a valid test.

Josh

<{POST_SNAPBACK}>



It depends totally on the projector, screen, and number of times a film has been run. You can watch a movie at one theatre and think it looks awful and see the same one at another theatre and think it looks great. They show a lot of amatuer films shot on miniDV here at Webster University in St. Louis in their own theatre (they have a film school and a film festival), and I've never heard anyone say that the DV films looked bad at all. In fact, most people are awed at how great they look on a big screen because they've heard all this misinformation on the stupid internet about DV looking bad on a big screen. So they're generally schocked to see it looking so good instead.

Heavilly compressed jpeg images are a perfectly valid test if you do it right. I did. They look exactly the same as the DVD's they came from. If you can find a better jpeg compressor than Ulead's or Fireworks I'd like to see it. And I'm showing them at full scale. Now if you want to boost them to 200% or something before compressing them, sure there'd be a difference. But we're not testing to see which one we can blow-up the biggest. We're trying to simulate real word viewing conditions. Very few people are using hi-def viewing systems. And a movie theatre screen in most cases is far from hi-def too.

A hi-def viewing test is a completely different thing altogether. Obviously DV is done for in a test like that. But hi-def for the masses is a long way off.
  • 0

#20 Charlie Seper

Charlie Seper
  • Guests

Posted 16 June 2005 - 02:46 PM

Here are the answers to the video/film samples. The 15 stills that came from video were:

1-D
2-C
3-B
3-C (This is me BTW)
4-A
4-B
5-A
6-D
7-A
7-C
8-C
9-A
9-B
10-B
10-D

I really thought this would be an easier test than it apparently was because I used 4-stills from a well known film shot on video--"Fast Runner". Two of the stills were obviously of Eskimos! The headshot of myself, I just played with the color and contrast, and then blurred it a bit. Nobody got that one. I'm telling ya, film is blurry! If I would have left it looking sharp you would have thought it was video for sure. Those sharp lines are things we mentally associate with video. Film doesn't have to look blurry BTW. But with all the filtering film makers do, it ends up that way because they want to soften skintones.

Anyway, I had fun doing it. See ya in a couple of weeks.
  • 0


Media Blackout - Custom Cables and AKS

Technodolly

Broadcast Solutions Inc

Willys Widgets

Ritter Battery

Wooden Camera

Gamma Ray Digital Inc

Abel Cine

The Slider

CineLab

Tai Audio

Aerial Filmworks

Rig Wheels Passport

CineTape

Visual Products

Glidecam

rebotnix Technologies

Paralinx LLC

Opal

Metropolis Post

FJS International, LLC

Technodolly

Paralinx LLC

Aerial Filmworks

Wooden Camera

Abel Cine

CineLab

Metropolis Post

Broadcast Solutions Inc

Glidecam

FJS International, LLC

CineTape

Tai Audio

Rig Wheels Passport

Willys Widgets

Gamma Ray Digital Inc

Opal

Visual Products

The Slider

rebotnix Technologies

Media Blackout - Custom Cables and AKS

Ritter Battery