Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Leica Lens Resolution test...

Thanks for the new chart Graeme. Very helpful visual. And thanks for all your posts.

KETCH ROSSi said:
If the chart looks this good on Leicas, I am pretty sure it will look exactly the same on Master Primes, as they are still the absolute Sharpest and best Cine Lenses ever made,
if the new Cine Leicas were to be any sharper then the Master Primes, then I must say that they have done nothing short of an insane Miracle a la RED... ;)
Click to expand...

I am waiting to see charts or compare them side to side, but I am sure they have done just that Ketch. All the more reason for Jannard and Leica to join forces, as they recently announced. (!)

Sounds like there is a little bit of work still left to do before release. But even if they were "only" just as sharp as MPs, if you consider the larger image circle and think about future sensor upgrades for Epic the Leicas are already an insane miracle a'la RED. Don't worry Ketch, that doesn't mean those Master Primes you own will stop being highly-coveted lenses everyone will want to use. ;)
 
In terms of the "came off the truck" performance of Pro-Res vs ArriRaw on the Alexa I am actually not surprised at how close they measure out to 1,400 lines or so. AFAIK the codec is engineered to allot more bandwidth to the more prominent image metrics at the expense of the more subtle characteristics (high frequency detail, flat field nuance, etc). FWIW the softer look of the Alexa is actually a good match for the MPEG-2 encoding step and typical sharpness added by consumer displays themselves (particularly 8 bit LCD panels). IMO as display and codec technology races forward, the compromises inherent in the Arri system will become more of a handicap.

Don't get me wrong, shooting a higher performance camera like the Epic and doing a proper virtual telecine step can yield a more detailed image with better dynamics than the Alexa even if it has to be blurred slightly prior to encoding for distribution (this is an MPEG-2 limitation, details of which are beyond this discussion). That said, if your mantra is "good enough" then shooting Alexa to Pro-Res using Log-C is a compelling option - as evidenced by the number of TV shows using it.

IMHO as the availability of Epics increases, the enabled feature set fills out and the back end tools mature (FilmLight, Assimilate, etc have more time to implement the new SDK) we will see DPs push more shows over to Epic when the new season starts shooting this summer. The compact size of the Epic, the desire to future-proof content, integration of native r3d files into popular post tools and the interest of established DPs in doing their best work will drive the shift.

For theatrical; the resolution advantage combined with the RAW color space, bit depth, HDRx, etc will make Epic the choice of all but the most loyal Arri fans - as evidenced by the slate of films that have chosen Epic of late.

Cheers - #19
 
Graeme Nattress said:
View attachment 13988
I've produced from Jim's data an MTF% graph that makes the data easier to understand. On the Y-axis we have %MTF normalized to the MTF at zero resolution, and I've also marked in the resolutions at where Jim noticed the aliasing. What's interesting is either shooting RAW or ProRes doesn't alter the MTF50% point at around 1100 lines, whereas the MTF50 on the Epic is way out at 2500. MTF50 is a very good guide to the overall perception of sharpness. You can see how going for a greater maximum resolution not only leads to lower aliasing, but also greater perceived resolution due to much better MTF50 performance. If you think that the MTF50 governs what you see on any size or resolution screen, and the detail beyond MTF50 is what you'd see on image projected at full detail, you can gain benefits not just with optimum projection, but all type of projection or display.

Graeme
Click to expand...

Thanks for this Graeme. For those of us still working to wrap our minds around this technical stuff, every little bit helps.
 
Roberto Lequeux said:
Thanks for the new chart Graeme. Very helpful visual. Thanks for all your posts on this thread.



I am waiting to see charts or compare them side to side, but I am sure they have done just that Ketch. All the more reason for Jannard and Leica to join forces, as they recently announced.

Sounds like there is a little bit of work still left to do before release. But even if they were "only" just as sharp as MPs, if you consider the larger image circle and think about future sensor upgrades for Epic the Leicas are already an insane miracle a'la RED. Don't worry Ketch, that doesn't mean those Master Primes you own will stop being highly-coveted lenses everyone will want to use. ;)
Click to expand...

Trust me, I will have no problem selling the Master Primes, if I see that the Leicas are actually superior in every way, I am all about the Uncompromise Quality in things,
I don't attach my self to any single piece of technology, unless it seats at the very top of technological advancements, and best for its intended use.

Even if the Leicas were in deed to be superior in sharpness, and many other things, the Master Primes, even so larger and heavier, still retain an incredible design,
that the Leicas do not compare to, especially the rubberized large ring for Manual focus, nothing else in the market is quiet likely, not to talk about the insanely smooth focus travel.

BUt I am already seating here, looking to get few of those EPIC/LEICA sets any way... ;)
 
Blair, I think the major difference on the Arri between compressed ProRes and Raw is not the codec, but the downsampling in camera to 1920x1080 before recording. The over-all low detail is from the sensor resolution of 2880 recorded pixels across, and the low MTF50 is most likely to be the low-con filter they use as part of their OLPF stack.

Graeme
 
Hi guys.

maybe you were waiting but unfortunately I still didn't open my promised brand new tread about Alexa -The New German Pragmatism

it's coming soon!

All I must say at the moment that wine always makes you thinking simply just better and always in a right direction...

Stay tuned...
 
Curiosity...

Curiosity...

My sense of the "look" of the Alexa is that the low-con filter combined with signal processing matched to it, is both improving and degrading IQ in different ways. I do agree that some of the softness on the zone plate is likely tied to the low-con filter. I also believe that it is a key enabler of the Alexa's extended DR.

Is it your assertion that the in camera downsample is more conservative (biased against aliasing on virtually any scene) than one can choose based on actual high frequency detail and observed moire when decoding ArriRaw in post?

I understand that as a part of the RedTeam you may have to decline to speculate on such matters, but my curious side had to ask ;-)

Cheers - #19
 
KETCH ROSSi said:
Trust me, I will have no problem selling the Master Primes, if I see that the Leicas are actually superior in every way, I am all about the Uncompromise Quality in things,
I don't attach my self to any single piece of technology, unless it seats at the very top of technological advancements, and best for its intended use.

Even if the Leicas were in deed to be superior in sharpness, and many other things, the Master Primes, even so larger and heavier, still retain an incredible design,
that the Leicas do not compare to, especially the rubberized large ring for Manual focus, nothing else in the market is quiet likely, not to talk about the insanely smooth focus travel.

BUt I am already seating here, looking to get few of those EPIC/LEICA sets any way... ;)
Click to expand...

Roberto and Ketch, while these new Leicas could be a tad sharper -- it seems like historically the appeal of Leica over Zeiss is not just sharpness but "character." Of course, this is not not the right place for that debate, on a thread centered around an MTF chart, but to me it's not as important if one brand is 2-5% sharper than the other. I guess I'm just one of those irrational people who believes in the fairy tale of the "Leica look" ....
 
This is a great post graeme - thanks so much - really appreciate it.
 
I'm not seeing any signal processing especially matched to the low con, and I do indeed agree the low-con contributes to the perceived DR. I'd not say the downsampling is not conservative as it's aliasing performance, especially with the in-camera recording, is not what I'd call good. All I can do is look at the images and say what I see. Of course, if you like what a low-con filter does, it's very easy to add the right amount for the shot as a filter in your matte-box, which to me, is the correct way to use such a filter, as a choice, not a permanent fixture.

Downsampling will never remove or stop aliasing happening - it can't close the barn door after the horse has bolted. All a good downsample can do is to preserve resolution and not introduce too much aliasing or ringing. If there's aliasing in the raw, there's nothing much any following stage of processing can do about it. That's why an OLPF is important, and that's why a high resolution image with negligible aliasing is a very important starting point.

Graeme
 
With analysis like this, we're getting right to the heart of "Why 4k?". It's why Arri recommend a 4k scan of 35mm film. It's why Sony make a 4k projector and why many others do so too. It's why Sony made the F65. However, it's even-more-so why we keep pushing resolution even when some people think 2k is enough. It's because there's more to resolution than resolution, and this discussion about resolution, aliasing and MTF goes some way to explaining that. The simple answer to "why 4k?" is more is indeed better. This thread goes some way towards giving a more detailed answer to that question. In the end, it's about making a very naturalistic, and unforced and un-electronic image.

Graeme
 
Graeme Nattress said:
With analysis like this, we're getting right to the heart of "Why 4k?". It's why Arri recommend a 4k scan of 35mm film. It's why Sony make a 4k projector and why many others do so too. It's why Sony made the F65. However, it's even-more-so why we keep pushing resolution even when some people think 2k is enough. It's because there's more to resolution than resolution, and this discussion about resolution, aliasing and MTF goes some way to explaining that. The simple answer to "why 4k?" is more is indeed better. This thread goes some way towards giving a more detailed answer to that question. In the end, it's about making a very naturalistic, and unforced and un-electronic image.

Graeme
Click to expand...

a simplification of that is it's always better to start with more samples and go down than...
 
Graeme Nattress said:
With analysis like this, we're getting right to the heart of "Why 4k?". It's why Arri recommend a 4k scan of 35mm film. It's why Sony make a 4k projector and why many others do so too. It's why Sony made the F65. However, it's even-more-so why we keep pushing resolution even when some people think 2k is enough. It's because there's more to resolution than resolution, and this discussion about resolution, aliasing and MTF goes some way to explaining that. The simple answer to "why 4k?" is more is indeed better. This thread goes some way towards giving a more detailed answer to that question. In the end, it's about making a very naturalistic, and unforced and un-electronic image.

Graeme
Click to expand...

So you're saying that less is really not more.

(Sorry Graeme, but someone had to say it.)
 
D Fuller said:
So you're saying that less is really not more.

(Sorry Graeme, but someone had to say it.)
Click to expand...

In the context of scientific and technical numbers: no. Less will never be more.
 
Steve Wake: I give gold stars out to only those who deserve them. Your assistance really helped. You stepped up, you get the star--but don't let that go to your head. You need another good deed for tomorrow.
 
Tim: Aliasing is fake detail at higher resolutions than what a sensor can physically resolve (constrained by number of pixels, fps, audio sampling rate).
Think of a moving car with rims seemingly being still or rotating backwards instead of being blurred (temporal aliasing) or moire patterns (spatial aliasing)
Usually, to combat aliasing, a kind of low pass filtering is deployed (OLPF, motion blur, mic response falloff)
 
View attachment 13988

It is often said that high limiting resolution - the reproduction of the finest detail a camera system can cope with, is not as important as how a camera deals with courser levels of detail. This is true enough. You can see in the chart how the Epic retains more contrast for longer (the graph stays higher, longer) so it deals with contrast in course details really well indeed.

But how is that achieved?

Well.... We have the goal of high contrast at lower detail levels, and the pit to avoid of horrible aliasing. To achieve high contrast at lower detail levels we aim for high limiting resolution with negligible aliasing. You cannot achieve high contrast at low detail levels with a small pixel count or you're into the pit of aliasing. Optics and sensors don't work to allow you to do that, not least because OLPFs have a very smooth rolloff.

So, a well engineered high resolution camera system will by nature of achieving high resolution and low aliasing have really good contrast at lower detail levels. I wouldn't even know where to begin to engineer a high resolution camera system to perform poorly at low detail levels.

On the other side, if you have too low a limiting resolution you're fighting aliasing if you want good contrast at low details, or you've got good aliasing control, but poor low detail contrast. Either way, you've got an image that not just measures soft, but feels soft.

That's why we're plotting the full MTF curve of the lens / OLPF / sensor / camera system here - to illustrate how engineering a high limiting resolution may appear excessive to some, but the result isn't just high resolution, but much better contrast at lower detail levels, and that improved contrast makes for a better looking image on everything from a postage-stamp sized web-movie all the way up to the largest high resolution projection.

Graeme
 
Great post Graeme, and couldn't agree more (Otto Schade would be pleased as well.) ;-)

Looking forward to seeing this to action at tomorrow's screening. Curious if we could see some of these MTF tests as well? Do you know if they are recorded in any way? Projectable?
 
I'm not sure what they're going to show, only that you'll enjoy what you see! I'm sure Otto Schade would indeed be pleased.

Graeme
 
You must log in or register to reply here.
Back
Top