Big Resolution News...

Jannard · 02-11-2010, 05:27 PM

Well... even we were lulled to sleep with conventional thinking. It was the industry's perception (before RED) that the most measured resolution you could get from a Bayer Patter sensor was 75%. We showed that with Graeme's magic, you could get 80%. We believed for the past 2 years that was the theoretical limit.

We were wrong. What new techniques in the new sensor, new OLPF, new debayering technology, better compression and a wicked sharp RED Pro Prime... that "limit" has been blow by.

Here is a test of the new M-X sensor shot 4K 16:9 at REDCODE 42 with a 100mm RPP. It shows that 4K now yields 3.5K measured resolution (as compared to 3.2K in old tests). We also will show (as soon as I post the next chart) that 4.5K from the same sensor and lens yields a true 4.1K measured resolution.

I guess the "RED doesn't really deliver 4K" is old news.

Jim

4.5K showing measured resolution of 4.1K. (85mm RPP)

Michael T O'Connor · 02-11-2010, 05:30 PM

Wow. Really cool. I wonder what the theoretical limit really is... It can't be much higher.

Rob Ruffo · 02-11-2010, 05:41 PM

I wonder if it would even be possible to manufacture 35mm glass that resolved much more than that using current glass cutting/polishing methods. Anyone know the answer to this?

jbeale · 02-11-2010, 05:49 PM

It is certainly possible to produce (for example) a diffraction-limited f/2.0 lens that covers S35 (at least for one color, and for resolution measured in the center... probably could do it broadband and full-frame, but I'm not so sure about that spec). What you might not like is the price. ...and up til now, not many cameras could have taken advantage of such a lens anyway.

Noah Kadner · 02-11-2010, 05:52 PM

Wait- the pundits were wrong about RED- when has that ever happened? :) Congrats to Jim, Graeme and the rest of the RED folks. :)

Noah

Peter Plevritis · 02-11-2010, 05:52 PM

You got to leave some big announcements for Red Day.

Kwan Khan · 02-11-2010, 05:54 PM

Super Cool...

Gavin Greenwalt · 02-11-2010, 05:57 PM

I haven't looked at much of the new M-X footage but in the M-X reel there seemed to be more checker boarding in the individual channels when I looked at the uncompressed TIFFs.

Does this increased measurable resolution also introduce more potential artifacting?

Andy Jarosz · 02-11-2010, 05:58 PM

This is huge guys! Good job!

Michael Stanmore · 02-11-2010, 05:58 PM

LOL. It's good news... and not to vomit on your favourite rug or anything... but it still looks like things get a little shaky around the 3500 mark on the 4.5k graph...

Is that wrong? I'm not trained to read these charts in any way so I'm seeing with layman's eyes a smooth gradient from left to right, all looks fine until about 3100 where you get spikes both up and down which are at that point smashing through the gradient by about 100%.... and then by about 3700 things start to get properly weird...

Obviously if you reply and say that that's all normal, and that's how the values are expected to be graphed when you're getting 4K resolution... and that the seemingly wrong values between 3100 and 4000 are actually fine... then I will happily yield to your expertise...

But I was just wondering is all.

02-11-2010, 05:27 PM	#1
Jannard Red Leader Join Date: Dec 2006 Posts: 5,438	Big Resolution News... Well... even we were lulled to sleep with conventional thinking. It was the industry's perception (before RED) that the most measured resolution you could get from a Bayer Patter sensor was 75%. We showed that with Graeme's magic, you could get 80%. We believed for the past 2 years that was the theoretical limit. We were wrong. What new techniques in the new sensor, new OLPF, new debayering technology, better compression and a wicked sharp RED Pro Prime... that "limit" has been blow by. Here is a test of the new M-X sensor shot 4K 16:9 at REDCODE 42 with a 100mm RPP. It shows that 4K now yields 3.5K measured resolution (as compared to 3.2K in old tests). We also will show (as soon as I post the next chart) that 4.5K from the same sensor and lens yields a true 4.1K measured resolution. I guess the "RED doesn't really deliver 4K" is old news. Jim 4.5K showing measured resolution of 4.1K. (85mm RPP) __________________ "Everything in life changes... including our camera specs and delivery dates..." We reserve the right to refuse service to anyone with a bad attitude. Last edited by Jannard; 02-11-2010 at 06:18 PM.

02-11-2010, 05:49 PM	#4
jbeale Senior Member Join Date: Jan 2007 Posts: 1,666	It is certainly possible to produce (for example) a diffraction-limited f/2.0 lens that covers S35 (at least for one color, and for resolution measured in the center... probably could do it broadband and full-frame, but I'm not so sure about that spec). What you might not like is the price. ...and up til now, not many cameras could have taken advantage of such a lens anyway. __________________ -John Beale

02-11-2010, 05:52 PM	#5
Noah Kadner Senior Member Join Date: Dec 2006 Location: Los Angeles Posts: 1,703	Wait- the pundits were wrong about RED- when has that ever happened? :) Congrats to Jim, Graeme and the rest of the RED folks. :) Noah __________________ Filmmaker Training at www.CallBoxLive.com featuring the Sony EX1 Guidebook, DVX100, and Panasonic HVX200. DSLR Cinematography Blog , Canon 5D Mark II and Canon 7 D training! Check out my book: RED: The Ultimate Guide to Using the Revolutionary Camera. Freelance Writing Services.

02-11-2010, 05:54 PM	#7
Kwan Khan Senior Member Join Date: Jan 2008 Location: NYC Posts: 1,057	Super Cool... __________________ Kwan www.finalfootage.com RED# 2897> EPIC, RED Pro Primes (on the way),Tokina 11-16mm, Nikon 17-35mm, 50mm, 80-200mm, Redrock MB, Pancro Budget Kit, Panasonic 17", Reflecmedia Chroma Flex and SEGWAY. Green Screen Studio at Time Square, Kino 8'4/4'4/2'4' Kit, Lite Panle, Varizoom Remote Head, EZ JIB, Indie-Dolly kit, Glidecam S.Shooter, Mac Book Pro , Mac Pro, Final Cut Studio 2, Shake. Can be rented any where in NYC

02-11-2010, 05:57 PM	#8
Gavin Greenwalt Senior Member Join Date: Dec 2006 Posts: 3,625	I haven't looked at much of the new M-X footage but in the M-X reel there seemed to be more checker boarding in the individual channels when I looked at the uncompressed TIFFs. Does this increased measurable resolution also introduce more potential artifacting? __________________ Gavin Greenwalt \|\| im.thatoneguy im.thatoneguy[at]gmail.com

02-11-2010, 05:30 PM	#2
Michael T O'Connor Member Join Date: Nov 2008 Location: Los Angeles, CA Posts: 59	Wow. Really cool. I wonder what the theoretical limit really is... It can't be much higher.

02-11-2010, 05:41 PM	#3
Rob Ruffo Senior Member Join Date: Dec 2009 Location: Montreal Posts: 651	I wonder if it would even be possible to manufacture 35mm glass that resolved much more than that using current glass cutting/polishing methods. Anyone know the answer to this?

02-11-2010, 05:52 PM	#6
Peter Plevritis Member Join Date: Mar 2008 Posts: 60	You got to leave some big announcements for Red Day.

02-11-2010, 05:58 PM	#9
Andy Jarosz Senior Member Join Date: May 2009 Location: Chicago Posts: 224	This is huge guys! Good job! __________________ Reel

02-11-2010, 05:58 PM	#10
Michael Stanmore Senior Member Join Date: Apr 2007 Posts: 324	LOL. It's good news... and not to vomit on your favourite rug or anything... but it still looks like things get a little shaky around the 3500 mark on the 4.5k graph... Is that wrong? I'm not trained to read these charts in any way so I'm seeing with layman's eyes a smooth gradient from left to right, all looks fine until about 3100 where you get spikes both up and down which are at that point smashing through the gradient by about 100%.... and then by about 3700 things start to get properly weird... Obviously if you reply and say that that's all normal, and that's how the values are expected to be graphed when you're getting 4K resolution... and that the seemingly wrong values between 3100 and 4000 are actually fine... then I will happily yield to your expertise... But I was just wondering is all.

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode