Sony HVR-Z1U vs. Panasonic AG-HVX200

Radek Svoboda · June 25th, 2005, 08:32 PM

Quote:

Originally Posted by Barry Green

I don't know how many would have predicted the level of purple/green chromatic aberration we're seeing in the FX1 lens...

XL2 with latest Canon lens has it, the DVX has it. If those cameras had resolution of Sony, it would look lot worse, so actually Sony's chromatic aberation is not too bad at all. I would be very surprised if HVX200 would not have it.

Sony is about 5,000 USD camera, same as HVX200. The Zeiss lens performance actually excellent and color fringing you refer is not noticable at all in practically all moving image situation.

Radek

Thomas Smet · June 26th, 2005, 11:10 PM

Barry thanks for those tests. I wanted to test the different formats but I do not use an Apple so I do not have a DVCpro HD codec to use.

Have you thought of using some 3D animation for testing? You could render a scene with a lot of detail to mimic a real world scene and add in lots of object and camera motion. Depending on the program you use you could even render interlaced and with a softer filter to mimic shooting with a video camera.

I understand this isn't a great test and this doesn't tell us how the cameras will work. The format is important however. You could have the greatest camera in the world but if it records to a not so good format you are stuck with that footage. At first I didn't like DVCPro HD for it's lower resolution but loved it for it's higher chroma. The real example I have had of any DVCPro HD samples is from the test on the One Rive Media website and in my opinion they looked very bad. Of course it is really hard to tell without comparing those same samples to HDV.

Even if your digital camera doesn't take raw photos could you still do a test for us? Even with jpeg compression you should get a nice RGB image to compress. Shoot a high megapixel image, slighlty blur it and down sample to all the HD resolutions. This should get you at least a little closer to a real world softer video test.

Barry Green · June 27th, 2005, 08:28 AM

I thought of using 3D-rendered footage, yes, but it'd have to be exceptionally photorealistic 3D footage, and I don't know where to get ahold of that.

I used the OneRiver picture as well, and there were definite artifacts in both, if you look in the earlier message I posted you will see examples of extractions for HDV and DVCPRO-HD from the oneriver picture. That's a picture that reallly stresses the codec, pushing it way beyond the bounds a normal picture would (as it was designed to do), and I think the DVCPRO-HD version shows a lot of artifacts, but is clearly and obviously quite superior to the HDV version.

As for not being able to test DVCPRO-HD on a PC -- you can, if your computer supports QuickTime. Avid has a codec pack available for free download and use (on systems that don't have Avid software installed). One of the codecs is Avid DV100, which is a DVCPRO-HD codec. Since running those tests I've downloaded and installed the codec and used it from Vegas -- it's very very slow, because Vegas' integration with Quicktime is quite inefficient. However, it's there, it does work, and it lets you create/export DVCPRO-HD files in 1080/50i and 1080/60i. There's also an option for 720/59.94p, but I haven't gotten that one to work properly yet. But the 1080/60i definitely works as it should.

And, it's different. Very different from Apple's implementation. I ran the same test with both pictures from both codecs (Apple's and Avid's) and the original, and stacked 'em all in a Photoshop document so I could selectively toggle layers on and off to compare them against each other. There were surprising differences -- one was overall softer, but with less mosquito noise, the other was quite a bit sharper, but with a noticeable increase in mosquito noise... in some areas of the OneRiver picture one looked decidedly better than the other, but in other areas the reverse was true...

... makes me want to run the HDV test again using a different platform. I figure Sony Vegas should be about the best for 1080i, figuring that Sony developed 1080 and would share the tech with their software editor, but maybe Apple or Avid native HDV would produce different results too...

Thomas Smet · June 27th, 2005, 01:10 PM

Barry thanks for the info on the DVCPro HD codecs. I am a 3D Studio Max user so I will try and do some tests with the codecs and see what I come up with. I did one small test on a still image I made in 3D Studio Max and the DV100 1080 version looked much better than the HDV version. I am also having odd problems with 720 but I will do more tests.

Jeff Asmar · June 28th, 2005, 12:23 PM

Thanks for all the info guys!
What kind of on camera light would you guys recommend for the Sony HVR-Z1U. Also, I 'm also looking for a wide angle and a telephoto lens. Any recommendations?

June 26th, 2005, 11:10 PM	#17
Thomas Smet Trustee Join Date: Mar 2004 Location: Milwaukee, WI Posts: 1,719	Barry thanks for those tests. I wanted to test the different formats but I do not use an Apple so I do not have a DVCpro HD codec to use. Have you thought of using some 3D animation for testing? You could render a scene with a lot of detail to mimic a real world scene and add in lots of object and camera motion. Depending on the program you use you could even render interlaced and with a softer filter to mimic shooting with a video camera. I understand this isn't a great test and this doesn't tell us how the cameras will work. The format is important however. You could have the greatest camera in the world but if it records to a not so good format you are stuck with that footage. At first I didn't like DVCPro HD for it's lower resolution but loved it for it's higher chroma. The real example I have had of any DVCPro HD samples is from the test on the One Rive Media website and in my opinion they looked very bad. Of course it is really hard to tell without comparing those same samples to HDV. Even if your digital camera doesn't take raw photos could you still do a test for us? Even with jpeg compression you should get a nice RGB image to compress. Shoot a high megapixel image, slighlty blur it and down sample to all the HD resolutions. This should get you at least a little closer to a real world softer video test.

June 27th, 2005, 08:28 AM	#18
Barry Green Barry Wan Kenobi Join Date: Jul 2003 Location: North Carolina Posts: 3,863	I thought of using 3D-rendered footage, yes, but it'd have to be exceptionally photorealistic 3D footage, and I don't know where to get ahold of that. I used the OneRiver picture as well, and there were definite artifacts in both, if you look in the earlier message I posted you will see examples of extractions for HDV and DVCPRO-HD from the oneriver picture. That's a picture that reallly stresses the codec, pushing it way beyond the bounds a normal picture would (as it was designed to do), and I think the DVCPRO-HD version shows a lot of artifacts, but is clearly and obviously quite superior to the HDV version. As for not being able to test DVCPRO-HD on a PC -- you can, if your computer supports QuickTime. Avid has a codec pack available for free download and use (on systems that don't have Avid software installed). One of the codecs is Avid DV100, which is a DVCPRO-HD codec. Since running those tests I've downloaded and installed the codec and used it from Vegas -- it's very very slow, because Vegas' integration with Quicktime is quite inefficient. However, it's there, it does work, and it lets you create/export DVCPRO-HD files in 1080/50i and 1080/60i. There's also an option for 720/59.94p, but I haven't gotten that one to work properly yet. But the 1080/60i definitely works as it should. And, it's different. Very different from Apple's implementation. I ran the same test with both pictures from both codecs (Apple's and Avid's) and the original, and stacked 'em all in a Photoshop document so I could selectively toggle layers on and off to compare them against each other. There were surprising differences -- one was overall softer, but with less mosquito noise, the other was quite a bit sharper, but with a noticeable increase in mosquito noise... in some areas of the OneRiver picture one looked decidedly better than the other, but in other areas the reverse was true... ... makes me want to run the HDV test again using a different platform. I figure Sony Vegas should be about the best for 1080i, figuring that Sony developed 1080 and would share the tech with their software editor, but maybe Apple or Avid native HDV would produce different results too...

June 27th, 2005, 01:10 PM	#19
Thomas Smet Trustee Join Date: Mar 2004 Location: Milwaukee, WI Posts: 1,719	Barry thanks for the info on the DVCPro HD codecs. I am a 3D Studio Max user so I will try and do some tests with the codecs and see what I come up with. I did one small test on a still image I made in 3D Studio Max and the DV100 1080 version looked much better than the HDV version. I am also having odd problems with 720 but I will do more tests.

June 28th, 2005, 12:23 PM	#20
Jeff Asmar New Boot Join Date: May 2005 Posts: 7	Thanks for all the info guys! What kind of on camera light would you guys recommend for the Sony HVR-Z1U. Also, I 'm also looking for a wide angle and a telephoto lens. Any recommendations?