View Full Version : What would you chose as a trade off?


King Goldfish
10-06-2004, 06:07 PM
I was gonna post this in Shaws post about the new Sony HDV coming out but felt it is to broad of a question.

The Panasonic AG100a (Or whatever the actual model is) has alot of awesome features from what everyone tells me.. from being able to pick up very low light to the cool film emulation of 24p to the 16:9 mask of 720x480 to give it HD Quality. but the fact is that its not HD makes me wonder if I should get it. Its becoming a bit outdated and Im seeing more and more places online selling them for under 1800 bucks.


Now that Sony is coming out with the new 3 Color chip HDV camera for a prossumer price but lacks alot of the decent functions like low lux and 24p but offers higher resolution Should I go with it.

I dont know what I want to do right now.. I would like to do ID filming as a hobby for a while, Im getting into Special Effects and hope to learn MAYA once I master 3ds Max and learn how to character model.


3DS Max renders in true HDTV and the new sony being under 8000 bucks would make a great companion for doing some cool Animations with real background shots.


It would be nice to *post edit* everything in Vegas or Premier and have the size show as 1440 pixels x 1080 lines.

but wouldnt 1080i be more like 1080x720? that was the other question I want to ask. Is this 3500 dollar camcorder by Sony going to offer true HD or will it be a lower end.


I plan on Buying a nice HD widescreen I saw from WalMart for 989 bucks which is 51 inches.. i dont need a reciever because i just want it more for viewing DVDs and Home Movies.

So here are all my questions.

1) How does 720x405 (720x480 masked) look on a HD Monitor?
2) Will a HDV Camcorder look better on a HD Monitor (51 inches) then a Panasonic AG100a camera using masked 16:9 ratio?
3) Is the new Sony HDR-FX1 1440x1080 or 1080x720 interlace? and can I render it to progressive afterwards?
4) Do the more expensive HD Cameras have progressive capability? Is the 3 Color chip on a higher end camera make a difference then a single on a higher end HD camera. (I Know the JVC thats out around 3500 bucks only has a sinigle chip but is a HDV 1080i camcorder)



Im just trying to understand what im really getting for the money.. the specs are a bit vague. one site says 1440x1080 but another said 1080x720 and is not true HD quality.

King Goldfish
10-06-2004, 07:26 PM
OK.. after talking to a sales rep on the phone I sort of got a better understanding of how HD works.

His way of explaining it was Progression is Doubling the line or mirroring the line of resolution below to double the sharpness. sort of feathering the black line. it is so microscopic that the human eye doesnt see it.

but if you were to have a 50 foot diagnal screen, you need a much higher resolution.


1080i is not 1440x1080 he said.. he said its 1080x720.. or 50% larger then standard 720x480

it is also 4:3 and not 16:9 it basically crops or masks it to 1244x720 or whatever the HD res is.

So basically, 1080 is not used on most small HD projection or LCD/Plazma screens but rather for the high end projectos in the thousands of dollars.

which I guess is why people of Independant filming want it.. it will be a slight improvement for resolution when post edited to progress but not much. He told me that the Panasonic 100 or 100a will still have a better lenz and better color even with a single chip at 720 vs the 1080i coming out by Sony (FX1).

Does anyone agree with this or is this his sales pitch to dump old stock off on me? either way he could have sold me either cameras. He said I could get a Panasonic 80 for about 2200 bucks but he said the 100 is a much better camera. I dont think 3200 bucks is a good price for yesterdays technology. Maybe 2000 bucks.

indietalk
10-06-2004, 09:18 PM
OK.. after talking to a sales rep on the phone I sort of got a better understanding of how HD works.

You mean HDV (hybrid HD/DV), not HD.

Shaw
10-06-2004, 11:18 PM
Goldfish: :)

This is hopefully semi-coherent. If it reads like Greek please let me know! I'm trying to capture all my thoughts at once while trying to answer your questions so it might not be very coherent. :)

About HD, HDV, resolution and HDTVs:

ALL versions of HD are 16:9! Don't listen to anyone who tells you HD is 4:3!

The high definition spec requires specific frame sizes of 1920x1080 and 1280x720. Those are the *only* pixel counts that can be considered "HD" for use on a home system (pro HD cinema cameras are a different story as they are not restricted to HDTV requirements).

1920x1080 is often refered to as 1080p/i and 1280x720 is refered to as 720p/i. Each of these can have progressive frames or interlaced "frames." Progressive frames are full frames scanned from the CCD(s) - precisely the way film is. Interlaced footage is composed of two fields (for historical reasons) that are each half resolution and both are played together to create a full frame. This is, however, not without its disadvantages. Interlaced footage displays each field separated by a small time interval (historically this was necessary as TVs could not scan an image fast enough resulting in the image dissipating towards the top of the screen before the next frame was scanned). This is the largest contributer to the "video look." Because of have 60 different fields being played rather than 24 or 30 frames you get a much much smoother motion rendering resulting in "hyper-realistic footage."

the High Definition spec allows for:

1920x1080 interlaced
1920x1080 24 progressive
1920x1080 30 progressive
1280x720 60 progressive
1280x720 30 progressive
1280x720 60 interlaced
1280x720 24 progressive

Because of the smaller frame size 720 footage can increase the possible progressive framerate while staying without broadcasting bandwidth limits.

What does this matter for our discussion?

The JVC creates 1280x720 30 progressive frames per second
The Sony Creates 1920x1080 60 interlaced fields per second

30 progressive frames cannot be transfered to film and look good. 60 interlaced does much better. If you want to "go to film" then the Sony is what you are looking for. If, however, you can do without 24 frames per second footage and want a look close to film the JVC is the better choice. The DVX shoots only standard definition 720x480 but shoots in true 24 progressive, 30 progressive and 60 interlaced modes.

*Please note that both the JVC and Sony fit the high definition specifications. They are just different forms. Both should look great on a large screen.

Progressive footage in NOT line doubling! Line doubling refers to a feature of HDTVs which upresses (increase the resolution) of lower resolution footage so it can play on an HDTV. Progressive scan is a full frame scan of the CCD (image sensor). Hence the reason it looks more like film (as film is a sequence of whole frames).

Interlaced footage can be rendered to progressive footage. This is often called "deinterlacing" as you are combining 2 interlaced fields into one full frame. There is some great software out there that will do this wonderfully.

So where does 1440x1080 come from? Well the current Sony model only captures 1440x1080 and then converts this, in camera, to 1920x1280. Thus your final image will be 1920x1080 but the amount of actual data captured will representative of a 1440x1080 image.

All the professional HD cameras I am aware of use a single chip so you can get good footage from this set up. The difference is that these are usually _large_ chips and not the 1/3" chips you will find in either of the "prosumer" HDV cameras. In general 3 chips will produce much better color. With one chip each set of four pixels records a separate color - two green, one blue and one red. The problem with having each pixel capture a color is that you effectively mutilate the amount of color data you could have. With a 3 chip camera you have a separate chip for each color. Each of these three images are then combined to create the final product.

Vegas will edit high definition footage. It will use a lossless compression (ie just manages the data better, you don't loose anything) to allow faster editing but that's it.

The guy was right about one thing though. Not very many HDTVs will play true 1920x1080 footage. It is usually downconverted slightly. Plasma TVs will display the full 1920x1080 image unmanipulated. It sounds like what you are looking at is a projector with a separate screen? If so check the projector and see what it is capable of displaying. It should say in the specs.

So bringing it all together.

The JVC does 1280x720 30p
The Sony does 1920x1080 60i
The DVX does 720x480 24p, 30p, 60i.

If you want to project this onto a wall the HDV cameras will hold up better. You CAN "upres" the DVX footage to fit the HD specs. It will hold up well. The DVX will be softer though as you cannot create more detail in the image just blow it up so it does not pixilate. The HDV cameras will be MUCH more detailed.

The Panasonic MAY have a better lense. It certainly is better, IMO, than the JVCs. The Sony however has a lense designed by a very high quality lens crafter... only time will tell how good it is.

Hope this made some sense. With all of the round about discussion I think I may have made myself sick trying to read it! :D Please ask questions if you have any!

Shaw
10-06-2004, 11:30 PM
Oh, forgot to mention:

The JVC is officially rated at 35 lux
The Sony is officially rated at 3 lux

These are the manufacturers specs

3 Lux is damn good. In fact I believe 3 lux would put it second in low light ratings with normal DV cameras! Of course, we still have to see how it does but the test footage I have seen looks like it has decent dynamic range and low light abilities :)

King Goldfish
10-07-2004, 11:21 PM
Shaw.. it just kind of spun everything told to me.. but I believe what you're saying.. but its like i had to reconstruct my thinking to follow this.

This is off the subject a bit.. but remember when I made that test video a few months back with my new 3 CCD camcorder? how the picture was pullilng apart like taffy and you could see the black res lines.. You or someone told me to go into vegas and use Progressive render to fix it.. and it did. So Progressive you're saying is not mirroring the horizontal res line below to double it but it is just one big scanned frame? while Interlace is 2 images woven together? like image one is line 1,3,5,7 etc while image 2 is 2,4,6,8 etc?

So how does vegas make a 2 image picture become one whole image? assuming this is what you are saying. the DV Tape is just raw data that tells the editor what to do before encoding it to playable video?



PS.. Indie I meant to Say HDV sorry/

btw guys.. thanks for sharing your knowledge with me. Alot of people forget what its like to crawl first.

In the future I want to share my knowledge with people once I have a full understanding.. so people know when a Sales Person is bullshitting them or not. LOL

Shaw
10-08-2004, 01:07 AM
Mr Goldfish,

I just hope I am making some sense. This stuff can be insanely confusing - even for those who are greatly experienced with video! Don't worry about asking questions :). That's what we are all here for - sharing knowledge!

I tried looking for the post that discussed your problems with the footage. I honestly can't seem to find it though :(. I do remember it being a field issue.. not sure what the exact problem was or why the fix worked.

But anyway, yes true progressive footage is composed of a series of full frames. Just like pictures from a digital still camera (only lots of em!).

For a very good explanation of interlaced vs. progressive footage take a look at the following website:

http://www.mandarinpictures.com/stephenzinn/

(Please note that this website was created by a PAL user so he/she talks about 25fps and 50fps instead of 60 as we have in the US. The general principle though does not change)

Interlaced footage is composed of two half resolution fields (see the link).

Canon uses what they call a "Frame movie mode" which mimicks a fully scanned image. What the camera does is capture both interlaced fields at the same instant and then combines them (you get a resolution loss). A true progressive image does not loose resolution (it even appears to our eyes to be higher res!) as the image sensor is scanned in one pass from top to bottom.

Interlaced fields are not usually captured at the same instant. Usually there is a slight delay. This becomes problematic because fast movement will cause each of the fields to be different from each other! Deinterlacing then becomes a very tricky situation (there is great software that will take care of this for you though) as you can't easily combine the fields as they don't make a coherent single frame.

See the link above though, it's much more helpful than I am and has lots of pictures

Of course, we can always answer any questions you may have as well :D

King Goldfish
10-09-2004, 10:40 PM
Mr Goldfish,

I just hope I am making some sense. This stuff can be insanely confusing - even for those who are greatly experienced with video! Don't worry about asking questions :). That's what we are all here for - sharing knowledge!

I tried looking for the post that discussed your problems with the footage. I honestly can't seem to find it though :(. I do remember it being a field issue.. not sure what the exact problem was or why the fix worked.

But anyway, yes true progressive footage is composed of a series of full frames. Just like pictures from a digital still camera (only lots of em!).

For a very good explanation of interlaced vs. progressive footage take a look at the following website:

http://www.mandarinpictures.com/stephenzinn/

(Please note that this website was created by a PAL user so he/she talks about 25fps and 50fps instead of 60 as we have in the US. The general principle though does not change)

Interlaced footage is composed of two half resolution fields (see the link).

Canon uses what they call a "Frame movie mode" which mimicks a fully scanned image. What the camera does is capture both interlaced fields at the same instant and then combines them (you get a resolution loss). A true progressive image does not loose resolution (it even appears to our eyes to be higher res!) as the image sensor is scanned in one pass from top to bottom.

Interlaced fields are not usually captured at the same instant. Usually there is a slight delay. This becomes problematic because fast movement will cause each of the fields to be different from each other! Deinterlacing then becomes a very tricky situation (there is great software that will take care of this for you though) as you can't easily combine the fields as they don't make a coherent single frame.

See the link above though, it's much more helpful than I am and has lots of pictures

Of course, we can always answer any questions you may have as well :D


see what i mean Shaw.. the jerk got it backwards.. he claimed Progressive scan was 2 layer.

It didnt seem to make sense since the very word "Interlace" would imply the woven of 2 objects.

King Goldfish
10-09-2004, 10:47 PM
Holy crap shaw... Thats a load of info to absorb

lol

but I sort of get it now.

I hope this doesnt sound lame.. but the 3:2 effect I always notice, I thought was some sort of transition of video. An Effect purposly put there to change scenes.

So That is the by product of 3:2?

King Goldfish
10-10-2004, 12:03 PM
update


I was slightly intoxicated last night while responding to this forum.


i've done that before.


Who wants a beer? :D