Jump to content

Why does the ArriScan oversample by 50%, per axis?


cole t parzenn

Recommended Posts

Chris M's suggestion that Arri traded 4× time for 4× space (~cost) overlooks that the larger sensor would be slower. Sensors take time to unload the image. I don't know if this time is proportional to the number of pixels, but if so there's no time lost in making four exposures, each with 1/4 as many pixels, excepting the time for the half-pixel motions by the piezo drivers. (How fast are they?)

 

Well, piezo drivers are 'pretty damn fast' in general - considering the distances involved if it's open loop then it'll be a very small and known duration - well tuned closed loop (feedforward) can probably approach that also.

 

But yeah, no idea of the relative scales of magnitude between sensor ADC offloads and the piezo micro-adjustments we're talking about - maybe they're comparable.

 

I agree though, it's off topic - just shooting the breeze :)

Link to comment
Share on other sites

  • 1 month later...

Arri does the sub-pixel shifting method of scanning.

 

They used to have this great animated .gif picture on their website which demonstrated how the sensor would shift via the piezo actuators and perform "micro-scanning". Sadly, I can no longer find it, even by way of the Internet Archive Wayback Machine. Perhaps someone else saved it?

 

EDIT: Wow! I found it! It was actually on Zeiss's website, which I conflated with Arri's. :P

 

microscanning.gif

Edited by Charles Zuzak
Link to comment
Share on other sites

  • Site Sponsor

Remember that the Arriscan is a true RGB scanner and uses a Monochrome version of the Arri sensor, so a 6K scan can be either 6 passes (RGB + RGB) or 7 Passes (RGB+IR + RGB+IR) or 12-Passes (for HDR Two Flash 6K) or 13-Passes (for 6K HDR +IR) the Area panel is moved by two precision Pizeo actuators for the 1/2 pixel re alignment.

Link to comment
Share on other sites

  • Site Sponsor

In general digitally sampling an image is best done at 2X the target resolution to improve aliasing performance.

 

http://en.wikipedia.org/wiki/Nyquist_frequency

 

So for a 4K scan you would really want to scan the original at 8K (The Northlight-2 and Imagica ImagerXE with the "10K" Kodak Tri-Linear CCD do this) the Arri scan is based on Arri's Alev(3? 4?) 3K area panel so in order to get oversampled (the 50%) scans at 4K they use the Pizeo motor 1/2 pixel shift.

 

These are true RGB scans where each color channel is scanned at the full frame size i.e. 8K x 6K, 6K x 4K, 3K x 2K etc. the performance of a oversampled scan even when down sampled is better than if the scan is at the target resolution and not down sampled.

Link to comment
Share on other sites

The ARRISCANNER is pretty old at this point. I think it arrived in early 2004. Back then I think Arri had just launched the D-20 or maybe not. In any case, the ALEV III is used in the Alexa, which is two generations removed from the D-20. My guess is the ARRISCANNER uses the original ALEV sensor or some technology related to it.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...