Go Back   Steve's Digicams Forums > Digicam Help > What Camera Should I Buy?

Thread Tools Search this Thread
Old Apr 22, 2008, 12:09 AM   #1
Junior Member
Join Date: Apr 2008
Posts: 4

I am looking into the waterproof camera Sanyo Xacto VPC-E1. However, just last Friday, Sanyo announced their new upgrade to their waterproof line (DMX-CA8 ).

Can you help me decipher the details to help me decide if I should wait (and pay much more) for the newer version? Thanks in advance.

VPC-E1 ("older" model) --> 6 megapixel, CCD, 30 fps video

DMX-CA8 (newer model) --> 8 megapixel, CMOS, 60 fps video

These are the main differences. I know that with regular digital cameras, the 6 vs 8 MP wouldnt make that much of a difference for regular use, but how about with camcorders, especially with CCD & 30fps vs CMOS & 60 fps (both in underwater conditions and dry)

I am a newbie, dont need super great quality, stills and video will probably not be printed or converted to DVD, just shared online (not youtube compression, but left as original)

gobears29 is offline   Reply With Quote
Sponsored Links
Old Apr 23, 2008, 11:04 AM   #2
Nagasaki's Avatar
Join Date: Jan 2004
Posts: 897

Regular US TV is30 fps so I guess this wouldn't be a particular advantage. Movies are 24 fps.As for the resolution I think a standard TV is less than 1 Mp so 6 or 8 is still a lot more than you need. Even high definition TV is only 1920×1080 pixels ie 2Mp.

There's no clear cut winner between CMOS and CCD although more manufacturers seem to be switching to CMOS.

Nagasaki is offline   Reply With Quote
Old Apr 23, 2008, 2:15 PM   #3
Junior Member
Join Date: Apr 2008
Posts: 4

Thanks for replying. After more research, I came to the same conclusion that for my needs, it doesnt make that much of a difference.
gobears29 is offline   Reply With Quote
Old Apr 23, 2008, 8:47 PM   #4
Senior Member
VTphotog's Avatar
Join Date: Mar 2005
Location: Extreme Northeastern Vermont, USA
Posts: 4,296

The only reason I can think of for 60fps over 30fps is that you would have better stills captured from the video, if that is what you want to do. (usually-not always) One major disadvantage is larger file sizes. Of course, I don't do video, anyway, so someone who does may have better information. The 6mp vs 8mp issue would only apply for camera still pictures, and there is not likely to be a lot of difference. CMOS generally has lower power consumption than CCD, so you may get longer battery life.

VTphotog is offline   Reply With Quote
Old Apr 24, 2008, 4:33 PM   #5
Senior Member
Join Date: Jul 2007
Posts: 116

The 60fps over 30fps is an advantage if you film alot of sports or fast moving activities.
herbs is offline   Reply With Quote
Old Apr 29, 2008, 1:10 PM   #6
Senior Member
Join Date: Dec 2006
Location: Davenport, IA
Posts: 2,093

The latest HD televisions do offer a 60fps framing rate.
ac.smith is offline   Reply With Quote
Old Apr 30, 2008, 11:24 AM   #7
Join Date: Jul 2006
Posts: 91

Double check the resolution the camera offers at 60fps.

My cameralest me take video at640x480 (full standard TV)@30 fps or 320x240 (1/4 standard TV)@60 fps.

If the only wayto get the video out of the camera is usinga standard (yellow; composite) video jack, then there will be considerable additional compromising of the picture notably losing crispness of color edges.

ajaynejr is offline   Reply With Quote
Old Jul 2, 2008, 3:36 PM   #8
Junior Member
olivia_k's Avatar
Join Date: Sep 2005
Posts: 12

Experience with the CA8? Any advantage over the previous model?

The new waterproof camera is on sale now, I want to replace my CG6.
olivia_k is offline   Reply With Quote
Old Sep 22, 2009, 4:41 AM   #9
Junior Member
Join Date: Oct 2007
Posts: 11

Originally Posted by Nagasaki View Post
Regular US TV is30 fps so I guess this wouldn't be a particular advantage. Movies are 24 fps.As for the resolution I think a standard TV is less than 1 Mp so 6 or 8 is still a lot more than you need. Even high definition TV is only 1920×1080 pixels ie 2Mp.
I disagree. 60P is a significant advantage. Regular TV is 60I, not 30fps. The interlacing hurts it badly. HDTV in US is 720/60P or 1080/60i. It should be 1080/60P. 1080/60I is so poor that 720/60P often looks better. Movies are 24P, but that isn't the same as 24P in TV terms. 24P would have been a disaster on old TVs. 30P wasn't fast enough refresh, which is why they went to the 60I hack. The dwell time is different (on CRT, a good LCD can improve this). 24P film also reedems itself by compensating for the inferior time resolution with better spacial resolution. Video should be 60P.

1080/60P transcodes well into other formats. Except 24P but it still transcodes substantially better into 24P than transcoding from 60i or 30P.
60i transcodes very poorly into 30P, and 30P transcodes poorly into 60i.

To transcode 60P into 30P, you either throw away every other frame or average two frames together (the first simulates a higher shutter speed and is more stroboscopic, the second a lower one and more blurry). Each gives a result similar to shooting native in that format with the corresponding shutter speed. To transcode into 60i, you take the odd lines from odd frames and the even frames from even frames or average the odd lines from frames N and N-1 and the even lines from frames N and N+1 (detail: I may have odd and even field order reversed here). Like before, this gives essentially the same results you would have obtained shooting in 30P or 60i natively.

To transcode 30P into 24P, you throw away one frame out of every 5. This means that objects in motion move further every forth frame than they did in the preceeding 3 frames, giving a jittery motion. 60i -> 24P gives a little more time resolution bu has the usual problems of transcoding from interlaced to progressive. To convert 60P to 24P
you throw away, or average, 3 out of 5 frames. The jitter is still there, but less. 120P would be ideal for transcoding to 24P, 60I, or 30P because it is a multiple of all the frame rates.

60P provides more data for processing as well. Chromakey, Rhotoscoping, motion tracking, etc. And it provides better stills.

Thus, if you are trying to produce quality results, you should generally shoot in 1080/60P instead of 30P, even if your video will be delivered in inferior formats because you can deliver in multiple different inferior formats while minimizing the quality degradation and you have high quality video available when it is possible to deliver it. If your primary deliver format will be 24P, you might shoot in 24P, though these days the primary 24P medium is film where higher resolution than a 1080 camera can deliver and 24P is often transcoded to 30P when delivered digitally (such as DVD -> CRT television) and thus would be better delivered at 30P in the first place (DVD supports 24P or 30P).

60P also lets you do slow motion effects better than 30P.

As for your comments about HDTV only being 2MP, the extra pixels in the camera can still help. Look up Bayer Pattern. A 1920x1080 "2MP" camera sensor is usually not really 1920x1080. It has a total of 1920x1080 sensing elements when it should have 1920x1080x3 sensors (one for each color). So, the real resolution is less. Now, the camera more or less throws out the same information that compression will throw out. And the compression throws those away because they eye is has less spacial resolution for chromenance than luminance (i.e. different number of rods and cones in the eye). So, video formats such as NTSC/ATSC/MPEG throw out half the color information while sending the lumance at full resolution. But a bayer pattern sensor doesn't really have the full spacial resolution it claims for luminance. If you point the camera at a black and white image, it does. But if you put a green, red, or blue filter in front of the camera, you lose 50%, 75%, or 75% respectively of the pixels. So a bayer pattern sensor produces a lower quality output. The image has been degraded before it has even been compressed. The way you get around this, is by using a bayer pattern sensor of 4 times the resolution (2x horizontal, 2x vertical). Then you have two green, one red, and one blue pixel for each pixel in the resulting image. Now, you really don't want to throw away any of this color information before editing, even if you will be throwing it away when compressing for delivery. The lower color resolution really hurts chromakeying. And it hurts edits which change the resolution and interpolate. Suppose you have something in the edge of your video you need to get rid of or the image wasn't composed as well as it should be. You can crop and enlarge slightly. Suppose you want to run a deshaking filter - you need extra pixels at the edges. Unfortunately, most digital camcorders compress which means you are throwing away information before editing which is a bad thing. You don't need to make it worse by throwing away information before compression. But on a better camera that did not do 4:2:2 pulldown (throwing away the color), the extra pixels would make more of a difference. Professionally edited video also goes through color correction and other color processing; the extra color information should be thrown away after this stage, not before.

So, 1080/60P with an 8MP bayer sensor or a real 2MP tristimulus sensor should be the MINIMUM for shooting HD video. You should really have a bit more than 1920x1080 as well, even if you will never go to film or cinema4K, so you can adjust the cropping of the frame. cinema4k would be a good format to shoot HDTV. If you are going to shoot cinema4K, you should really have more than cinema4K for your raw footage. And even 35mm is a compromise - consider IMAX. But IMAX is too expensive even for most professional movies. You also have UHDV, which is 7680x4320 50P/60P, which the Japanese hope to roll out around 2020, which is about 33MP. And if that seems like a lot, being there is more than an order of magnitude higher resolution. Ever read an (grown up, normal sized text) book or magzine on film or movie viewed so it takes up the same portion of your field of view as if you were holding it in your hands with the screen filling your field of view? You can't. The resolution isn't good enough.

And while movies may be shot on film at 24P, it is generally done with cameras that can be "overcranked" to higher speeds when necessary for slow motion effects (or special effects that might need higher temporal resolution). And many people find 24P to be annoyingly jerky, even on film.

Video goes through many processing stages and it loses quality at each stage. If you start at less than HDTV quality, your final result will be less than HDTV quality. If you start at bayer pattern "would you believe 1920x1080 resolution?" then it certainly won't.
Quality does matter. Now, if you are just shooting stuff for personal use, inferior quality might be worth the money savings. If you are shooting pro-am video, quality matters but you may not have the budget for the good stuff. If you are doing professional work, inferior quality definitely doesn't cut it. So, the answer is not that quality doesn't matter, just in some cases it doesn't matter enough in comparison to the cost. Even the best professional cameras on the market today are not good enough, if you are picky. But if you are on a budget, you have to aim a bit lower. And there are diminishing returns, above a certain point you get less bang for each buck spent.

The VPC-FH1 can take 8MP stills while shooting 1080/60P video - so the 8MP can be used in that way as well, if the camera allows.

CCD vs. CMOS, is another can of worms.

Last edited by whitis; Sep 22, 2009 at 6:55 AM. Reason: accidental premature submit
whitis is offline   Reply With Quote

Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off

All times are GMT -5. The time now is 2:16 AM.