So, should we wait for the huge price drop?

spencer said:
Man, do all you guys talk about resolution all the time? I think I saw only one post on this thread regarding color and the way the image looks.

Call me crazy, but I'd rather take something that looks more filmic and has way better colors and less resolution than a camera that has great resolution but lacks in the actual image.

I don't have much loyalties towards one or the other, but all of the Canon stuff I've seen (mind you, not H1, just the XL-1 and XL-2) didn't look as pleasing to the eye as the DVX footage I've seen (if you can somehow find the short film HUSK, you will see really what a DVX could do.) Maybe that subconsciously pledges my loyalty to HVX (from what i've heard, a hi def DVX) and makes me suspicious of the H1 (i'm going to assume a hi def XL2 with some bells and whistles). Resolution be damned.
Click to expand...

Oh, right?

28 Days Later. Did I say enough?

If it's a Canon Wild-Card? Then whatever, it's still a testament to what time, experience, and effort can do with a Canon camera.

You're subconciously suspicious ideals are suspect to Brand Loyalty. Hell, you should've just came out and said "I'm on Panasonic's side.", you contradicted yourself.

As I said in another thread, though-- I wouldn't shoot a film with the XL2 for free. I'd rather rent a DVX100a/b, and that's no exaggeration.

I hold no brand loyalty, but too many people are judging the cameras before anyone even produces anything worth watching.

Pft. Said it once, I'll have to keep saying it I see.
 
Pascal_Parvex said:
Well, you call 25 and 50 frames variable frame rates? It's nothing new apart from the fact that the 50 frames are progressive.
Click to expand...
Pascal, start reading before you bash others:
Framerate: variabel 12/18/20/23/25/27/30/32/36/48/50 fps (Bilder/s)
From the German HVX website.
Convinced?

Best Pe.

P.S.: Apart from that you are NOT a comical troll if you ask if it makes sense to spend 6000 k€ for a camera that comes out last in a 4 cam shoot-out. But ya know,
do not get into a dispute if religious belief is stronger than the proven facts of life.
 
Kholi said:
Oh, right?

28 Days Later. Did I say enough?

If it's a Canon Wild-Card? Then whatever, it's still a testament to what time, experience, and effort can do with a Canon camera.

You're subconciously suspicious ideals are suspect to Brand Loyalty. Hell, you should've just came out and said "I'm on Panasonic's side.", you contradicted yourself.

As I said in another thread, though-- I wouldn't shoot a film with the XL2 for free. I'd rather rent a DVX100a/b, and that's no exaggeration.

I hold no brand loyalty, but too many people are judging the cameras before anyone even produces anything worth watching.

Pft. Said it once, I'll have to keep saying it I see.
Click to expand...

****, see, I associate canon stuff with low budget indies and whatnot. I don't think I found out that 28 Days Later was shot on an XL-1S (right?) until about a month or two ago, and I was like "no ****ing way..."

So, yeah, ok, Canon has made some kickass images. But in all honesty, I still like the best DVX footage I've seen over 28 Days Later. I guess we have to take in to account post production as well though. So, since I'm without any experience (and any sight of raw footage) I suppose I have to concede any opinions and loyalties I have and wait unti I see the cameras' raw footage for myself.
 
Excuse my ignorance but if all the under 10k HD cameras (I assume they all can) recorded straight to a firestore device or cineporter would the softness during motion still be a problem? Would the sound still be compressed? Would the colour saturation be higher? Or is it the bit rate difference (25 vs 100) that influences these most? Thanks.

R.N.
 
Ralph, firewire has nothing to do with image quality, it's only a transfer protocol.

The camera and codec (DVCPRO100 , HDV, ect....) effect the image.
 
Stevet said:
Ralph, firewire has nothing to do with image quality, it's only a transfer protocol.

The camera and codec (DVCPRO100 , HDV, ect....) effect the image.
Click to expand...


What is it about the HDV codec that makes it inferior?
 
Ralph...Well, as HDV is a codec designed to compress an HD image onto DV tape...through a variety of protocols including creating blocks of frames that share the mejority of visual info, ect....without using the 'inferior' word, it's easy to assume there are some drawbacks!

Anybody who has seen stuff like the canon footage from sundance (the couple being hit with a photo flash...with beautiful texture both at the moment of the flash and the darker moment afterward) can see that HDV is pretty damn fine in the right hands!

There are criticisms of DVCPRO100 as well, of course (people have descibed it as noisy), but the issues there are less graspible for an HD newbie like myself.

My assumption was that HDV was inferior because it had no be, in order to fit on a DV tape in the first place. But the deeper you get into this stuff, the more alice in wonderland it gets.
 
Brand loyalty is a ridiculous thing. I would chop in my FX1 at the drop of the hat for something better. If the HVX200 proves to be a much better camera its bye bye to the FX1. And in a years time if Sony bring out someting better than the HVX200, then its bye bye to the Panasonic.

As far as I am concerned I don't care for brand names. It's the camera that counts, whether it's made by Sony, Panasonic, JVC or Acme.

There are so many folk on here who poo poo the FX1/ZU1, and have never seen the image it gives or even held one in their hands. Its a stupid way to be, although very common.

I am just glad I don't fit into that catagory. I just want the best in that price bracket. End of.
 
Technically, pixel shifting can give you 50% resolution advantage. I read in Hdforindies that Panasonic's Jan actually said on Creative Cow that they use both horizontal and vertical pixel shifting on the HVX200.

So, if you start with lets say a 720x576 dvx pal sensor, you have to crop it to 16:9 which would be 720x432 active pixels. Adding the advantage of pixel shifting in both directions, this would be 1080x648pixel for luma. In lines, that is 1080*9/16= 607.5 horizontal lines and 648 vertical lines. So, from an engineering point of view it is possible that you can get about 600x650 lines or something similar out of pixel shifting a cropped PAL sensor in a 3xccd system.

Pixel shifting in a valid technique and actually increases luma resolution up to 50%. The canon XL1 uses pixel shifting in horizontal axis only to get 270k sensor to 410k luma result. That is 410/270 or about 50% increaze.

http://consumer.usa.canon.com/ir/controller?act=ModelFeaturesAct&fcategoryid=115&modelid=8274&pageno=1

Since Jan says they do pixel shifting, we only have to figure from what and to what they pixel shift on the hvx200. DVCPROHD is maximum 1280x1080 at 1080p. If pixel shifting gives 50% increase and since they wouldn't want to pixel shift to something higher than the recorded format capabilites to not loose light with smaller pixels for no reason, they would have to start with 1280/1.5 x 1080/1.5 sensors. That is 853x720. Jan will probably release the actual sensor resolution one day and explain how it is implemented in both axis.

If the camera does use cropped PAL sensors, it will have, as calculated above a maximum of 607 horizontal and 648 vertical lines at reasonable sharpness settings. So it should:

1) Appear inferior in horizontal resolution to cameras with more than 1080 true pixels in horizontal axis, such as JVC HD100 and also Canon XL 1H.

2) Have similar vertical res with 720p cameras.

3) Have similar horizontal resolution or slightly inferior compared to pixel shifted cameras with 960pixel sensors as the Sony FX1.

The problem with all this is that we wouldn't be able to tell, even if the camera had a true sensor with lets say 1920 pixels horizontally, since the DVCPROHD format has 1280 horizontal pixels, less than HDV cameras at 1080 (1440 pixels) and same at 720p (1280 pixels), it would appear lower horizontal res compared to other 1080 cameras and similar res with 720p. Simply put, the horizontal capability of DVCPROHD 1080p is 1280 and is the same with the 1280x720 formats on the market and smaller than the other 1080 formats..
 
Inferior/superior, sharper/softer, cleaner/nosier, dvcpro100codec vs hdv codec. It all looks so new and they all sound so close in performance that it seems like it comes back to the subjective: price, look, and what you're going to use it for.

I hear people complain about the compressed sound in the hdv codec. How is it compromised and what does that mean when I hear it?

Oh I found the answer to my own question earlier about bit rate: "a lower bit rate means that there is less information in the video signal which will still result in a*signal of poorer quality." My question is how much pooer quality 5%, 10%, 25%, 50%, and in all areas or just some like colours, sound, and/or sharpness?
 
limo991 said:
Technically, pixel shifting can give you 50% resolution advantage. I read in Hdforindies that Panasonic's Jan actually said on Creative Cow that they use both horizontal and vertical pixel shifting on the HVX200.

So, if you start with lets say a 720x576 dvx pal sensor, you have to crop it to 16:9 which would be 720x432 active pixels. Adding the advantage of pixel shifting in both directions, this would be 1080x648pixel for luma. In lines, that is 1080*9/16= 607.5 horizontal lines and 648 vertical lines. So, from an engineering point of view it is possible that you can get about 600x650 lines or something similar out of pixel shifting a cropped PAL sensor in a 3xccd system.

Pixel shifting in a valid technique and actually increases luma resolution up to 50%. The canon XL1 uses pixel shifting in horizontal axis only to get 270k sensor to 410k luma result. That is 410/270 or about 50% increaze.

http://consumer.usa.canon.com/ir/controller?act=ModelFeaturesAct&fcategoryid=115&modelid=8274&pageno=1

Since Jan says they do pixel shifting, we only have to figure from what and to what they pixel shift on the hvx200. DVCPROHD is maximum 1280x1080 at 1080p. If pixel shifting gives 50% increase and since they wouldn't want to pixel shift to something higher than the recorded format capabilites to not loose light with smaller pixels for no reason, they would have to start with 1280/1.5 x 1080/1.5 sensors. That is 853x720. Jan will probably release the actual sensor resolution one day and explain how it is implemented in both axis.

If the camera does use cropped PAL sensors, it will have, as calculated above a maximum of 607 horizontal and 648 vertical lines at reasonable sharpness settings. So it should:

1) Appear inferior in horizontal resolution to cameras with more than 1080 true pixels in horizontal axis, such as JVC HD100 and also Canon XL 1H.

2) Have similar vertical res with 720p cameras.

3) Have similar horizontal resolution or slightly inferior compared to pixel shifted cameras with 960pixel sensors as the Sony FX1.

The problem with all this is that we wouldn't be able to tell, even if the camera had a true sensor with lets say 1920 pixels horizontally, since the DVCPROHD format has 1280 horizontal pixels, less than HDV cameras at 1080 (1440 pixels) and same at 720p (1280 pixels), it would appear lower horizontal res compared to other 1080 cameras and similar res with 720p. Simply put, the horizontal capability of DVCPROHD 1080p is 1280 and is the same with the 1280x720 formats on the market and smaller than the other 1080 formats..
Click to expand...

Good explanation there. So it is feasable that the HVX COULD be using pixel shifted DVX CCDs, I'm not saying it is. At any rate the HVX is not a 1080p camera but rather a 720p that can upsample to 1080.

Also I hope that res chart test was wrong at DV.com as the HVX looked really bad in the jiggle test. I mean it looked worse than a decent SD camera.
There was no comparison between the HVX and the others it just flat out lost.
I was shocked at how bad the res was and again hope it was a bad setting on the camera or the focus got knocked out or something.

The Canon held its own very well with the CineAlta and it is definately the res king in the $10k camera range.

The Sony changed a lot with the different sharpness settings. Perhaps the HVXs sharpness setting was set too low (-3)?

Anyway next time anyone gets 4 cameras together shoot something real so we can see what the cameras really look like. I would like to see heavy motion as in sports footage or anything moving.
 
js33 said:
Anyway next time anyone gets 4 cameras together shoot something real so we can see what the cameras really look like. I would like to see heavy motion as in sports footage or anything moving.
Click to expand...

js33, are you an engineer or a shooter? Sounds like you work for Canon?? Ask Gary Adcock about his 3 cam comparision. In the real world the HVX is clearly the image winner. I would not buy a camera based on a static test chart. I cannot say that they are as many shooters excited about a camera more than the HVX.
 
I am most concerned with image quality and that includes resolution after all isn't that what HD is all about? More resolution than SD? If the cameras resolving powers are barely better than a DVX you might as well shoot the DVX in 16:9 mode and just blow the image up in post.

Also I'm not a big fan of HDV's meg2 compression and I think the HVX will win only because it is using a superior recording format that will handle motion better than HDV. But there is no arguing from what we have seen that the HDV cameras with their native 1440 x 1080 have the upper hand in resolution.
 
If the camera is 540 lines vertical lines it doesn't matter if it is recording twice that in luma and twice that in chroma (4:2:2) since I still get low resolution. It could be a Sony at 25p at half vertical res and achieve the same real resolution.

On the other hand, the camera still solves SD's main problem, chroma resolution. Even if it was sd it would have a lot more chroma:)

IF we take 550x540 lines of resolution, this is 977x540 luma resolution. The format also does 640x1080 lines chroma so it can get 640x540 chroma. So this is still a lot better than 720x480 SD luma and 180x480 SD (4:1:1) chroma, isn't it? Almost 4:4:4 after sd downsampling.
 
Salatar has a good point about waiting, but you have to dive in at some point.

Improvements will never stop. If they did, we would not buy, it's that easy.
 
You must log in or register to reply here.
Back
Top