DV Info Net

DV Info Net (https://www.dvinfo.net/forum/)
-   Canon XL1S / XL1 Watchdog (https://www.dvinfo.net/forum/canon-xl1s-xl1-watchdog/)
-   -   Frame mode/16 X 9 Resolution Loss (https://www.dvinfo.net/forum/canon-xl1s-xl1-watchdog/14726-frame-mode-16-x-9-resolution-loss.html)

Hugh DiMauro September 19th, 2003 07:19 AM

Frame mode/16 X 9 Resolution Loss
 
So, I'm reading a DV INFO.NET article about progressive scan by Steve Mullen, and he describes how the XL1s frame mode delays the row two green elements by one time line and then added to the row three green elements etc etc yadda yadda yadda which results in reduced vertical resultion only giving 320 lines of resolution. Another article I read stated that using the 16 x 9 function further reduces vertical resolution.

Here's my question: If this is the case, why bother? Doesn't that just defeat the whole purpose of the Canon's superb lens optics and 3 ccd capability? So I should just always shoot interlaced, using the 16 x 9 guides and de-interlace in post?

Edwin Quan September 19th, 2003 02:53 PM

i'd love to hear the answer to this questions. anybody?

Jeff Donald September 19th, 2003 03:53 PM

It depends what medium you're shooting for, broadcast, internet, VHS, DVD, film etc. Can you be more specific as to your footage's intended use?

Don Berube September 19th, 2003 04:04 PM

Perhaps you should direct your question to people who are actually bonafide shooters, who actually shoot and make their living with the cameras being talked about.

There is no absolute way to state the absolute resolution from any model of camera as it varies with each application, depending upon what lens is used, filtration used and detail settings used. The glass in front of the image plane is what makes the most difference. Quality glass is key. It can be said that the only true way to measure camera output resolution is by using a professional sharpness indicator chart and literally counting the lines resolved over a high quality broadcast monitor. If you do not even have a high quality professional broadcast monitor to begin with, then you are literally wasting your time obsessing about resolution. In any case, the overall image which people will percieve as high quality and professional is a combination and sum total of many things, not just resolution.

There are many professional shooters here who use the XL1S. Some of these people use it for a certain percentage of their work, while some use it exclusively. In any case, most shooters who use the XL1S will agree that they use Frame Mode more often than not. Everyone I know who shoots with an XL1S uses Frame Mode most of the time.

Personally, I utilize many different cameras, including the XL1S. It all depends upon the intentions of the Director and the needs of the client. When I select a camera to shoot with, it is for various reasons which usually differ from project to project. Many times the XL1S does a wonderful job at fulfilling these requirements. When I do choose to shoot with the XL1S, 99.97% of the time I shoot in Frame Mode with the 16X manual/ Servo lens, Century Optics adaptors and occasionally a PL-XL adaptor with a nice film lens and such...

A good friend of mine who specializes in wildlife footage and shoots for very high profile clients as well also shoots in Frame Mode 99.97% of the time. It's the glass used that matters the most.

As our good friend Jeff Donald suggests, whether or not to use Frame Mode depends upon how your footage will ultimately be utilized and viewed. If you shoot news/ ENG footage then shoot in normal Movie Mode/ 60i. That is what news stations are used to.

Frame Mode has a certain look that many people love. It lends itself well to the overall "cinematic look" sought after by many DV filmmakers and content creators who typically produce long-form projects.

If your footage will be presented via streaming video over the web or CD-ROM, go right ahead and shoot in Frame Mode - it looks great.

How exactly is YOUR footage being used?

- don

Peter Moore September 19th, 2003 08:11 PM

"So, I'm reading a DV INFO.NET article about progressive scan by Steve Mullen, and he describes how the XL1s frame mode delays the row two green elements by one time line and then added to the row three green elements etc etc yadda yadda yadda which results in reduced vertical resultion only giving 320 lines of resolution. Another article I read stated that using the 16 x 9 function further reduces vertical resolution.

"Here's my question: If this is the case, why bother? Doesn't that just defeat the whole purpose of the Canon's superb lens optics and 3 ccd capability? So I should just always shoot interlaced, using the 16 x 9 guides and de-interlace in post?"

I don't understand the problem in answering such a simple question. My answer is, no it's not always better to shoot 4x3 60i. Sometimes it is, sometimes it isn't.

If what you want is a film-like 16x9 look at 30p, you have several options. For 16x9 you can use s a 16x9 anamorphic lens. It's pricey, but makes a huge difference in resolution.

For framerate one possibility is to deinterlace the footage yourself. The nice thing about 60i is that you can fairly easily go to 24p. But if you want 30p, you're probably better off with using Canon frame mode. 60i to 30p conversion will necessarily blur the image a little bit, or show blocky movement.

A lot of people (including myself) are skeptical of Steve's article. While the technical explanations seem sound, the real world results do not match up. I've seen absolutely no resolution loss in frame mode. None. Yes, there is obviously resolution loss in 16x9 mode, but it's not 25%, it's somewhat less. That's because Canon cameras in 16x9 mode still give you a 480 line image but use less of the CCD to do it. The CCD is actually more than 480 lines, so you're somewhat better off using the camera's 16x9 mode.

The lowest quality option will be to do your 16x9 in post, almost guaranteed. That's not true of certain other cameras, but it is of Canon XLS and GL1-2. However certain programs (Vegas for one) can do a good job of this. You may want to do this to have the flexibility of altering the framing of your shots later on (just like filmmakers do!)

Josh Bass September 20th, 2003 04:49 PM

So wait, now we like the anamorphic mode in the XL1s? I've been doing it in post. . .it doesn't make sense though, now that I think about it. How could a method that simply chops off 25% of your image have more resolution than a method squeezes all the pixels into the same area?

Don Berube September 20th, 2003 06:49 PM

errrrrrrr,,, ummmmm
 
I was talking solely about Frame Mode in my last post in trhis thread.

I do not recommend using the built-in, electronically interpolated 16:9 from any make or model of camera - it is simply too soft.

The only time I would recommend using built-in electronic 16:9 in any camera would be if you are only purposing your footage to a streaming video or half or quarter-size encode (QuickTime, WMP, Real, etc).

To clarify, I do not recommend using built-in electronic 16:9 (from ANY camera) if you are planning to present your footage full-frame, be it from a video tape, DVD, broadcast over the air or cable or output to film. You can most always get away with the loss of resolution if you are only presenting your footage in a frame size smaller than that of full-frame - such as 320x240, 240x160, etc. If you do shoot this way, make sure you back off on the filtration and diffusion, your image will be soft to begin with. Go ahead and still use Neutral Density filters as needed.

If you are using a camera that does not offer true 16:9 CCD's then your best route to acquire in 16:9 would be as follows:

1) shoot in 4:3 and use Mini35Digital adaptor with a true anamorphic 35mm cinematography lens (well, that WOULD be sweet... if only...) and monitor with a true 16:9 monitor

2) shoot in 4:3 with a Century Optics Anamorphic adaptor and monitor with a true 16:9 monitor

- - the Sony PVM-8044Q will resolve in 16:9 mode, so will the Panasonic/ NebTek NEB70XL LCD panel - -

3) shoot in 4:3 with a Century Optics .6x or .7x Wide Angle adaptor and frame for 16:9, monitoring via a high quality monitor masked with gaffers' tape for 16:9 (or mask your LCD screen with tape for 16:9)

NOTE: the Century Optics Wide Angle adaptor is helping to give you a wider and more 'majestic-looking' field of view which will help in framing for 16:9 to get that 'widescreen' look. Century Optics makes these adaptors for most every camera out there. Keep in mind that Century Optics glass is of the utmost finest quality and more often than not their lens adaptors have more resolving power than the actual lenses on most camcorders. This makes a big difference. Do not substitute the Century Optics adaptor for a lesser brand like Kenko (or any other adaptor not manufactured by Century Optics). I know this sounds like a huge and shameless plug for Century Optics, but I am speaking from experience. Their glass rules and definitely will add to your image rather than being a bottleneck.

- don

Boyd Ostroff September 20th, 2003 07:58 PM

Don Berube writes:
> - - the Sony PVM-8044Q will resolve in 16:9 mode, so will the
> Panasonic/ NebTek NEB70XL LCD panel - -

I think Century now offers an anamorphic LCD screen magnifier which will stretch the image into the correct 16:9 proportions. Another option (on the Mac) is a little shareware program called BTV Pro which will take firewire from your camcorder and let you display the full frame in any size and proportion on a powerbook screen.

Don Berube September 20th, 2003 08:38 PM

I agree that the BTV solution is handy if you do not have access to a good monitor that can resolve 16:9. However, you really do benefit from using an NTSC monitor that displays both fields and a true NTSC color space. A VGA screen doesn't do this.

I guess if your footage will never end up on an NTSC screen, then it doesn't matter as much.

- don

Boyd Ostroff September 20th, 2003 08:55 PM

Agreed, but just wanted to point out an inexpensive ($45) solution to displaying anamorphic video in the correct proportion. I find it very handy personally. It lets me preview my footage with a director by loading it on my laptop, and I don't need to re-render it in the correct proportions. You are correct that the colors won't be quite right and the resolution is limited a bit. But it does show the full frame, and it's well worth the shareware fee. The program can also do some other cool things, like capture directly to your hard drive and record time lapse video.

Josh Bass September 20th, 2003 10:34 PM

Okay wait. . .that Ben Syverson guy says that doing in camera 16:9 (even without the real 16:9 chips) is better than letterboxing in post (obviously not better than buying the anamorphic lens, but screw that). . .you all are saying he's wrong?

Boyd Ostroff September 21st, 2003 06:45 AM

It is camera-specific and has to do with how the DV compression comes into play. From what I've read, the Canon's do a better job with in-camera 16:9 while the Sony's are better to crop in post. I don't have a Canon, but did run tests with my VX-2000 and there is a noticeable improvement when you crop in post (see this example).

Rather than trying to conceptualize and theorize and ask everyone else's opinion, why not run your own tests? It really isn't that hard. You can download a resolution chart at bealcorner. Shoot a test with your own camera, export some frames and look at them in photoshop. That way you can form your own opinion and not be colored by the prejudice of others. This is absolutely not intended to question any of the great advice people are giving in this forum! But we all have our own personal styles and values, and in the end YOU will need to be happy with the results.

Peter Moore September 21st, 2003 08:44 AM

"I do not recommend using the built-in, electronically interpolated 16:9 from any make or model of camera - it is simply too soft."

Well I do, only on Canons, if you're sure you want 16x9 and don't need the freedom to reframe your shots later. Canon's 16x9 mode is better than doing it in post. I don't think you'll find any comparison where doing it in post is better, and if you do, I'd love to know what software you're using.

That is ONLYL true of Canon cameras. Sony's 16x9 mode, for example, is no different than doing it in post, and some say it's worse.

Don Berube September 21st, 2003 12:03 PM

I stand by my statement. It applies to Canon as well. Note: This is by no means meant to denigrate the usefullness of the XL1S, you all know that I still love to shoot with the XL1S! It's just the reality until we see true 16:9 CCD's incorporated (such as the new Optura Xi with a 2 megapixel "True 16:9 High Resolution" CCD)

I can say that I am speaking from experience. I have tried them all, including the DVX100 and much more expensive GY-DV700. You have to define what price point and what output venue you are referring to at this point. I am simply referring to what is acceptable as professional broadcast full frame or film out. It could be said that this also dictates that you should be shooting with a larger 2/3" CCD - but obviously here we are talking about cameras "under $4000", often referred to as the 'minimum acceptable standard' for professional broadcast full frame or film out...

As I stated, built-in 'electronically interpolated' 16:9 Mode (from any camera not using true 16:9 CCD's) is acceptable and useable enough for anything less than full-frame broadcast or film output. Go ahead and use it if you plan to output to a 1/2 or 1/4 frame size MPEG-1 (or any other encode that is 1/2 or 1/4 frame size). It is simply not detailed enough for professional full-frame broadcast or film out. Sure, it may be viewable via full frame broadcast,,, but it just doesn't look as good as it would had the footage been acquired with a higher-end professional broadcast camera with larger, true 16:9 CCD's.

One camera that does not offer true 16:9 CCD's, but does seem to have enough detail and resolution to make shooting with built-in 16:9 worth considering is the DVX100. Most everybody will agree that the DVX100 does technically output the most detail and resolution out of all of the cameras under the $4000 price point. Note: This is by no means meant to denigrate the usefullness of the XL1S, you all know that I still love to shoot with the XL1S! Now, we all know that resolution is not everything,,, but consider it important if you have to output 16:9 internally with a camera that does NOT offer true 16:9 CCD's... This camera has a built-in 16:9 mode that still outputs 4:3 video, but with a letterbox mask applied - much like you would get if you had shot in 4:3, framed for 16:9 and applied a letterbox mask in post. I would still maintain that you should use a Century Optics .6X or .7x wide-angle lens adaptor in front of the lens to create a wider, more cinematic and 'majestic looking' field of view. The extra resolving power of the CO adaptor really does add to the image. This overall combination looks pretty decent and is definitely worth considering if you have limited post-production hours available and you need to absolutely limit the amount of rendering that you do before your output deadline. Of course, when the upcoming 72mm anamorphic adaptor is made available by Century Optics, then that will clearly be the way to go for anamorphic 16:9 video on the cheaps (with either the XL1S or DVX100). The Panasonic Anamorphic is not resolute enough, exhibits some chromatic aberrations and is not rectilinear (too much barrelling). The Panasonic Anamorphic is quite good enough for home and some corporate video applications though. Just not quite good enough for higher-end professional broadcast or film out.

Now let's be clear that we are referring to outputting to a 4:3 video frame with a 16:9 letterbox mask applied. If you want to output and resolve to a true 16:9 frame (with no letterbox mask), then you simply need to shoot with either an anamorphic adaptor or use true 16:9 CCD's.

If you still maintain that built-in electronically interpolated (or stretched) 16:9 is good enough for full-frame professional broadcast or film out, then I would say that we just have different ideas of what is acceptable and what isn't... it may look good enough for you on a smaller 9" or 14" screen,,, but when you blow it up to a big screen it simply dopesn't cut it. It may be 'good enough' for home or corporate video use, but it isn't high enough in detail and resolution for professional broadcast or film out.

1) If you want to output 16:9 anamorphic video but you do not have true 16:9 CCD's, then: shoot in 4:3 using a Century Optics anamorphic adaptor and monitor with a quality 16:9 NTSC monitor. If you have a flip-out LCD screen on your camera, then also consider using the Century Optics Widescreen Magnifier
http://www.centuryoptics.com/products/dv/lcd/lcd_ws.htm

2) If you want to output in 16:9 Letterbox (in a 4:3 video frame, viewable on any 4:3 monitor), then:
a) Shoot in 4:3, use a Century Optics Wide Angle adaptor in front of the lens and frame for 16:9 and resolve to an anamorphic 16:9 frame in post via Final Cut Pro
http://www.kenstone.net/fcp_homepage...ding_16_9.html

b) -or- shoot in 4:3, use a Century Optics Wide Angle adaptor in front of the lens and frame for 16:9 and apply a letterbox matte to the 4:3 footage in post.
http://www.kenstone.net/fcp_homepage...ding_16_9.html

This is all changing of course and will inevitably be moot point. Most every new camera model being offered from this point on will provide true 16:9 Modes (either true 16:9 CCD or via megapixel CCD's) so that it will be upwards compatible with the upcoming HDTV broadcast frame size standard(s). Do you think we will ever see one standard fully and universally adaopted world wide? In this lifetime? Hmmmmmm...

- don

Peter Moore September 21st, 2003 03:15 PM

"but it just doesn't look as good as it would had the footage been acquired with a higher-end professional broadcast camera with larger, true 16:9 CCD's."

I believe the original question was whether using the XL1S's 16x9 mode was better than shooting with an XL1S in 4x3 and performing the stretching in post. I certainly was not comparing the XL1S 16x9 mode to a higher end camera or to the use of an anamorphic adapter.

If all you have to choose from is built-in 16x9 and post-16x9, and you're using the same camera either way (a Canon) I think the choice is definitely to use built-in 16x9. I have compared using the GL2's 16x9 against taking a 4x3 shot from the GL2 and stretching it myself. The result is definitely better with the in-camera 16x9.

I don't endorse using it over an anamorphic lens or a true 16x9 CCD, of course. Nothing beats a true 16x9 CCD.

As for whether it's good enough for broadcast, I happen to think it is, but it's very subjective.


All times are GMT -6. The time now is 11:46 PM.

DV Info Net -- Real Names, Real People, Real Info!
1998-2024 The Digital Video Information Network