Do codecs make a difference in image quality?

Daniel Epstein · June 20th, 2014, 01:17 PM

I agree with most of what has been said before and you should remember that your display may not be able to show you the difference between codecs. Usually the big improvement from lower bit rates to higher ones is not as much in static resolution as in color information, dynamic resolution and motion artifacts. You might see a difference in quality if you use green screen or color correction between the lower bit rate codecs and higher ones. I would try shooting the same shot in different codecs and combining them in an edit to see if you find a visible difference and why.

William Hohauser · June 20th, 2014, 07:04 PM

You can pixel poke and view footage at extreme magnification to get an idea of what the codec may be doing. However it can get to be a waste of practical time depending on how you expect to use your footage. Some people have said that movie projection on a large professional screen is a good arbiter but I have found that projection actually covers fine artifacts that might be annoying on a monitor. I made a short doc years ago in HDV for BluRay and it ended up being projected last year at a cinema. Despite it being almost entirely handheld and filming detailed city street shots, it looked great and I couldn't see any artifacts that were previously noticeable while editing.

Ty Ford · June 21st, 2014, 07:49 AM

Daniel,

Sounds like a good idea. Do you suppose the type of shot with this HM650 would make a difference? e.g. locked down, average light, hand held, hi or low light, high contrast scene, etc?

Thanks for your help.

Regards,

Ty Ford

Daniel Epstein · June 21st, 2014, 08:02 PM

Hey Ty,
Yes for test purposes the types of shots could reveal issues. I would mostly use locked down shots with a lot of motion in the subject if possible for comparison It depends on what you have access to . Lots of detail would probably be a good test. Brick walls always seem to reveal issues. High contrast scenes should show more issues. Daylight vs studio lighting would be nice. Gradients are always good to shoott to see if banding starts showing up. Handheld might be useful if you can repeat the shot. As many different situations as you can deal with would be the best but time and budget can limit how many different scenarios one can do. There was a very nice test of the Arri Alexa and Canon C500 but that may be too ambitious for this. You just need to get a sense of the difference if any between the compression schemes and the cameras performance.

June 20th, 2014, 01:17 PM	#16
Daniel Epstein Trustee Join Date: Jan 2006 Location: New York, NY Posts: 1,267	Re: Do codecs make a difference in image quality? I agree with most of what has been said before and you should remember that your display may not be able to show you the difference between codecs. Usually the big improvement from lower bit rates to higher ones is not as much in static resolution as in color information, dynamic resolution and motion artifacts. You might see a difference in quality if you use green screen or color correction between the lower bit rate codecs and higher ones. I would try shooting the same shot in different codecs and combining them in an edit to see if you find a visible difference and why.

June 20th, 2014, 07:04 PM	#17
William Hohauser Inner Circle Join Date: Dec 2005 Location: New York City Posts: 2,650	Re: Do codecs make a difference in image quality? You can pixel poke and view footage at extreme magnification to get an idea of what the codec may be doing. However it can get to be a waste of practical time depending on how you expect to use your footage. Some people have said that movie projection on a large professional screen is a good arbiter but I have found that projection actually covers fine artifacts that might be annoying on a monitor. I made a short doc years ago in HDV for BluRay and it ended up being projected last year at a cinema. Despite it being almost entirely handheld and filming detailed city street shots, it looked great and I couldn't see any artifacts that were previously noticeable while editing. __________________ William Hohauser - New York City Producer/Edit/Camera/Animation

June 21st, 2014, 07:49 AM	#18
Ty Ford Inner Circle Join Date: Nov 2004 Location: Baltimore, MD USA Posts: 2,337	Re: Do codecs make a difference in image quality? Daniel, Sounds like a good idea. Do you suppose the type of shot with this HM650 would make a difference? e.g. locked down, average light, hand held, hi or low light, high contrast scene, etc? Thanks for your help. Regards, Ty Ford __________________ Ty Ford's Audio Bootcamp Field Guide Ty Ford's Audio Video Blog

June 21st, 2014, 08:02 PM	#19
Daniel Epstein Trustee Join Date: Jan 2006 Location: New York, NY Posts: 1,267	Re: Do codecs make a difference in image quality? Hey Ty, Yes for test purposes the types of shots could reveal issues. I would mostly use locked down shots with a lot of motion in the subject if possible for comparison It depends on what you have access to . Lots of detail would probably be a good test. Brick walls always seem to reveal issues. High contrast scenes should show more issues. Daylight vs studio lighting would be nice. Gradients are always good to shoott to see if banding starts showing up. Handheld might be useful if you can repeat the shot. As many different situations as you can deal with would be the best but time and budget can limit how many different scenarios one can do. There was a very nice test of the Arri Alexa and Canon C500 but that may be too ambitious for this. You just need to get a sense of the difference if any between the compression schemes and the cameras performance.