Correct V1 Sharpness setting

[Steve Mullen]

I'm still working on the issue of the correct V1 Sharpness setting. Until one has their monitor sharpness correct -- it's impossible to evaluate their camera's setting.

So I download several 1920x1080 test clips -- played them on my MBP and fed my HDTV via HDMI -- and found that to fully eliminate EE on my HDTV, I had to dial Sharpness down to 8. In theory, this is the correct setting. However, at this setting, the test clips looked softer than does ATSC HD.

Which opens the question is "some" amount of EE needed on an HDTV to have the picture look good?

You can get these HD samples at: http://www.w6rz.net/

You can also use the "vertical rez. with moving bar" clip to test your deinterlacer.

[Tom Roper]

Originally Posted by Steve Mullen

Until one has their monitor sharpness correct -- it's impossible to evaluate their camera's setting.

It may be impossible to decide what level of sharpening you like but it's entirely simple to evaluate sharpening in the objective sense by taking the monitor out of the equation and doing the slant edge test with Imatest. It will report the image as under or over sharpened relative to the pixel radius of your choice, usually 1 or 2 pixel radius is generally accepted as a standard sharpening level for most comparisons.

[Daniel Boswell]

This is 30p clip with sharpness dialed down to 4. I think I found a sweet spot with the settings

http://homepage.mac.com/epiphany2002/30p.wmv

[Steve Mullen]

Originally Posted by Tom Roper

It may be impossible to decide what level of sharpening you like but it's entirely simple to evaluate sharpening in the objective sense by taking the monitor out of the equation and doing the slant edge test with Imatest. It will report the image as under or over sharpened relative to the pixel radius of your choice, usually 1 or 2 pixel radius is generally accepted as a standard sharpening level for most comparisons.

How many people have the charts and the Imatest software?

I know only one reviewer that uses Imatest to measure resolution. From what I can see, the numbers he gets don't agree with any of the reports by people who do it by eye. I'm not saying it's right or wrong, but I see a problem reporting numbers that can't be compared with the vast majority of camcorders that have been tested with during the past decade not using Imatest.

Plus, one needs to calibrate their monitor anyway. The problem is that now we have two paths into an HDTV -- HDMI and analog component. I'm going to burn the HD DVD test clip so I have one source of HD test signals. Too bad the HD networks don't broadcast test patterns any more. :)

[Tom Roper]

Originally Posted by Steve Mullen

I know only one reviewer that uses Imatest to measure resolution. From what I can see, the numbers he gets don't agree with any of the reports by people who do it by eye.

Forget about the resolution numbers, the slant edge test will tell you when a camera sharpening setting equates to a standard sharpening level which would generally be accepted as a "no sharpening applied" if you chose a 1 pixel radius for a DSLR, or 2 pixel radius for a point and shoot or video camcorder.