Canon 7D Zone Plate Shots

[Daniel Browning]

If you look at the wedge pattern you linked me to, at 100% it moirés in the center: http://www.bealecorner.org/red/test-patterns/star-chart-bars-full-600dpi.png

Thanks, I see what you mean now. The linked images aren't really an ideal comparison because they themselves aren't anti-aliased. That's not a problem for their intended usage because it occurs at frequencies far below that which they are supposed to be sampled, but it does make them less useful for your purposes (as a comparison). For example, the Siemens star is supposed to be used at a magnification (print size, distance, and angle of view) that will not allow the camera to resolve to the core.

At first glance of course this is a ridiculous nitpick, but it's a proof of concept that shoehorning any pattern onto a pixel grid will eventually moiré,

Not necessarily. As long as "shoehorning" includes the removal of any spatial frequencies that are too high for the grid, moiré is avoided. For sufficiently low resolutions, that may mean nothing more than a single circular blob, or diffuse gray blur, but even that is preferable to aliasing.

and when framing the same object with different cameras, lower resolutions are mathematically more prone to moiré.

In practical reality, lower resolutions are indeed more prone to moiré. Not because of mathematics, but because the "practical" sources of blur and reduced contrast are smaller in relation to Nyquist. That includes scene detail, motion blur, aberration, diffraction, etc. If we only consider the pure mathematics (spherical cows in a vacuum, etc.), then the propensity for aliasing is the same.

 

[Roberto Lequeux]

It's good to know the facts and all.

I love Canon and Red products, but it is worth mentioning that Canon's and Red's marketing is written "very differently".

So while one of the two puts out "facts" about their products' specs, it makes sense that the other put out facts about both. : )

 

[Neil W. Smith]

spherical cows in a vacuum? ...

spherical cows in a vacuum? ...

In practical reality, lower resolutions are indeed more prone to moiré. Not because of mathematics, but because the "practical" sources of blur and reduced contrast are smaller in relation to Nyquist. That includes scene detail, motion blur, aberration, diffraction, etc. If we only consider the pure mathematics (spherical cows in a vacuum, etc.), then the propensity for aliasing is the same.

Daniel,

I understand Nyquist effect , I understand the the Poynton effect and to a lesser extent I understand the Graeme effect, but not sure I understand the 'propensity for aliasing is the same effect' .... could you please explain to a simple minded Trojan like me what you're getting at?.

Thanks,
Neil

 

[Eisen Feuer]

In case this has the potential to become a lengthy discussion, we might want to move this to emails/PMs. Your call, your post count would indicate you're 100x more invested in this forum than I

Thanks, I see what you mean now. The linked images aren't really an ideal comparison because they themselves aren't anti-aliased.

I zoomed in on them beyond 100% to see the edges... (the wedge specifically) how are they not anti-aliased?

In practical reality, lower resolutions are indeed more prone to moiré. Not because of mathematics, but because the "practical" sources of blur and reduced contrast are smaller in relation to Nyquist.

Aren't you making a mathematical statement by bringing Nyquist in?

 

[Daniel Browning]

The first images are too small to really comment,

Sorry I wasn't clear about this earlier, but for the purpose of discussing aliasing, one is meant to treat these as if they were 100% crops (i.e. just one guy cropped out of a very large group photo). So there is no larger image size (except to blow it up to a 200% crop).

Many would prefer the first image, describing it as sharp, crunchy, high microcontrast, with lots of fine detail, such as stubble. Others would see it as fake-looking, with harsh transitions, jagged edges, and lots of false detail, such as stubble that should be too small to see and jagged edges on the ear.

Some would prefer the second image, describing it as smooth, natural, with the appropriate amount of detail for its size. Others would see it as mushy, hazy, low contrast, and lacking in fine detail.

I can see that a filter ideally should be used - I think in some cameras it is too strong and over blurs the image which has been the case on some DSLR models

Well, essentially the filters have to be too strong; in order to knock out the aliasing you have to accept some collateral damage. But for me, reducing aliasing is so important that I'm glad to sacrifice a little contrast to do it; and most of the time I can get most of that contrast back through sharpening. (But not when the shot is too noisy for sharpening, for example.)

I dont think these VDSLRs are inhibited by the filter but by skipping

Agreed.

Im using the phrase HD 1080 in quotes with the EX1 because I dont think that is 1080 either (just a guess)

Check this out:

Assessment and settings for Sony PMW-EX1 and EX3

It's got a pretty good balance between aliasing and resolution.

I would define a proper 1080 image as one that looks as good as a 1080 image dowscaled from a Raw still - the EX1 does not do that

Now that's my idea of proper!

 

[Daniel Browning]

I understand Nyquist effect , I understand the the Poynton effect and to a lesser extent I understand the Graeme effect, but not sure I understand the 'propensity for aliasing is the same effect' .... could you please explain to a simple minded Trojan like me what you're getting at?.

:smile: The pixel-level contrast determines how much aliasing will occur (i.e. 80% MTF at Nyquist will result in very bad aliasing, whereas less than 10% MTF will result in none). Achieving high contrast for a small number of pixels (say, a 0.5 MP digicam) is easy: you can shoot handheld, on a rickety tripod, with soft lenses, at f/32, with a rough eyeball focus, and still get a very sharp shot. Change that 0.5 MP for a 50 MP camera, though, and now you find that shooting handheld causes a blur that spans almost a dozen pixels. Nailing focus is much more difficult. The rickety tripod is no longer sufficient to suppress that blur. Lenses that were sharp enough to cause aliasing on a 0.5 MP camera cannot do the same on a 50 MP camera. So on and so forth. It is those practical, everyday factors that make aliasing less and less likely with each increase in resolution.

 

[Sanjin Jukic]

How the hell did you get hold of such a beautiful lens, Sanjin? .... was your father part of the Zeiss company? ... is that why you care so much about visual imagery?

You need to come to LA, dude ... bring your quirky dual RED/Canon rig with you and your most excellent glass .... Rodney thinks you're very cool so you could join our crazy team.

We need an Austrian with a sense of humour to offset German seriousness and OC style ... come and see us soon .... but bring your glass with you ... wrapped up carefully.

Neil

Neil,

thanks for your invitation and I'm there in LA as soon as I can.

Btw, my father was an architect...and I started to play with a photo camera at the age of two-three years old kid...

but I was the most impressed with blown up picture with a slide projector at that time of of my childhood...

Also as you know that all Austrians are a bit "crazy" compared to Germans and I do not need to quote now the famous names... :laugh:

Here below is the old-timer lens Carl Zeiss Sonnar Olympia 180mm f/2.8 which you could still find to buy cheap @ vintage Leica shop in Vienna

and you could ask maybe @ Kevin Cameras vintage lens shop in LA:

Carl Zeiss Sonnar Olympia 180mm f/2.8 on 7D.

Carl Zeiss Sonnar Olympia 180mm f/2.8 on 7D.

 

[Graeme Nattress]

To add in some comments....

The charts we use are very large... Jim printed mine out for me on a large format printer and I mounted it onto a large sheet of thin ply to allow me to shoot it. I think it's like 5ft across or something. That is because to get a sin wave sufficiently small, the overall chart has to be large, or else you'd get aliasing from the print process.

So what does the perfect zone plate look like - well, it looks like that printout and you can't easily see that on a web image! The one I shot on the RED is close to perfect from a camera shot chart point of view, which is nice. Daniel's links to Alan's articles will show you how many other cameras perform, and Alan's words tell you what you can learn from these tests.

I do think you need slightly better aliasing performance at HD than SD and at 4K than HD because of the increasing likelyhood that you'll be downsampling the image, and having excessive aliasing early in the process has an increasing negative impact as you downsample.

Graeme

 

[Steven Caesare]

I agree with you Tom. It's good to know the facts and all. I just find it a little strange that this test is coming from inside Red HQ. If it was coming from a Reduser, like Sanjin, then it wouldn't matter, but the fact that Red is doing this test to blow the doors of Canon is what's weird to me. Red doesn't need to prove anything, they already have, 100-fold. The Canon is less than $2,000. The Red 8X Scarlet will be over $3,000 with a 2/3" sensor and fixed lens. The Red One is over $30k. I don't see the point. I don't see Canon shooting zone plates of the the R1 and claiming that it isn't true 4k so why is Red trying to prove how 'sucky' the 7D is?

Well, you gotta remember that these guys are shooters and gear-junkies themselves. Jim's owned just about every camera made.

Red started because Jim couldn't find the camera he wanted, yet KNEW could be made, so he decided to make one. Opting to sell the rest of us one was almost a side effect.

So if Red didn't exist, they'd be doing this stuff anyway... we just happen to get to see it now because of this forum full of people following Jim's "for-profit hobby".

-sc

 

[Neil W. Smith]

Carl Zeiss Sonnar Olympia 180mm f/2.8 ...

Carl Zeiss Sonnar Olympia 180mm f/2.8 ...

Neil,

thanks for your invitation and I'm there in LA as soon as I can.

Btw, my father was an architect...and I started to play with a photo camera at the age of two-three years old kid...

but I was the most impressed with blown up picture with a slide projector at that time of of my childhood...

Also as you know that all Austrians are a bit "crazy" compared to Germans and I do not need to quote now the famous names... :laugh:

Here below is the old-timer lens Carl Zeiss Sonnar Olympia 180mm f/2.8 which you could still find to buy cheap @ vintage Leica shop in Vienna

and you could ask maybe @ Kevin Cameras vintage lens shop in LA:

Carl Zeiss Sonnar Olympia 180mm f/2.8 on 7D.

Carl Zeiss Sonnar Olympia 180mm f/2.8 on 7D.

Architects have a strong sense of visual style ... so that's where you get it from.

Agree about crazy Austrians .... been drunk many a time in Wien - once an Austrian has had a drink or two, they really loosen up ... much friendlier than your cousins to the West.

Sweet lens set-up for your 7D ... what lens adaptor are you using? ... I have a request into Kevin cameras - they're just down the road from us. If they don't have the Carl Zeiss Sonnar Olympia 180mm f/2.8 in stock, could I ask you to see if your local Leica shop has a good one in stock? ... I'll get my sister to meet you there and she'll pay for it and ship it back to LA ... its not that I don't trust my sister to buy it own her own .... it's just that, well, she's my sister and she has no idea about technology :-(

See you in LA soon I hope ... maybe NAB in Vegas next year?

Thanks,
Neil

 

[Neil W. Smith]

Zone Plate TIFF?

Zone Plate TIFF?

To add in some comments....

The charts we use are very large... Jim printed mine out for me on a large format printer and I mounted it onto a large sheet of thin ply to allow me to shoot it. I think it's like 5ft across or something.

Graeme,

Do you have the TIFF file of the Zone Plate chart you use that you could send me? .... I'll trade you for one of the original 76 MB TIFFS from those first prototype images you and Jim tweaked ... still think the dolly shots of Jim's red and white 959 Porches were the best.

Last three years have gone past like a blur ...

Neil

 

[jbeale]

vector patterns and raster pixel-maps

vector patterns and raster pixel-maps

If you look at the wedge pattern you linked me to, at 100% it moirés in the center: http://www.bealecorner.org/red/test-patterns/star-chart-bars-full-600dpi.png

I think Daniel was originally referring specifically to the zone plates, for example
http://www.bealecorner.org/red/test-patterns/Zone720-sine-g22-A.png
http://www.bealecorner.org/red/test-patterns/Zone720-sine-g22-b.png

which on my screen, do not have visible aliasing when viewed at 100%.

Elsewhere on that page, the Siemens star chart does have moiré effect in the middle due to inadequate antialiasing. That pattern, of course cannot be perfectly represented by any finite pixel map since there is a singularity in the center (what should the value of the center pixel be?) I can add a star with a center cutout, if that will help anyone.

That pattern might be better presented as a PDF or SVG file, in which case the aliasing decisions are left up to your own software rendering it for display or printer.

 

[Graeme Nattress]

Graeme,

Do you have the TIFF file of the Zone Plate chart you use that you could send me? .... I'll trade you for one of the original 76 MB TIFFS from those first prototype images you and Jim tweaked ... still think the dolly shots of Jim's red and white 959 Porches were the best.

Last three years have gone past like a blur ...

Neil

Unfortunately, the file is not mine to give out. I still have a good bunch of early Frankie raw files - they're alway fun to look back at!

Graeme

 

[Sanjin Jukic]

Architects have a strong sense of visual style ... so that's where you get it from.

Agree about crazy Austrians .... been drunk many a time in Wien - once an Austrian has had a drink or two, they really loosen up ... much friendlier than your cousins to the West.

Sweet lens set-up for your 7D ... what lens adaptor are you using? ... I have a request into Kevin cameras - they're just down the road from us. If they don't have the Carl Zeiss Sonnar Olympia 180mm f/2.8 in stock, could I ask you to see if your local Leica shop has a good one in stock? ... I'll get my sister to meet you there and she'll pay for it and ship it back to LA ... its not that I don't trust my sister to buy it own her own .... it's just that, well, she's my sister and she has no idea about technology :-(

See you in LA soon I hope ... maybe NAB in Vegas next year?

Thanks,
Neil

Neil,

you are right and also that's a reason they needed psychoanalyst like Sigmund Freud to help them when they were sober :laugh:

For that Zeiss Sonnar Olympia 180 f/2.8 that became now a pretty rare lens to find and also was converted to Nikon F I do have an ordinary Nikon F to Canon EOS mount found @ Adorama.

If you want that I can have a look for that lens together with your sister @ Leica shop Wien no any problem and I'll find a time for that.

 

[Deanan]

Many would prefer the first image, describing it as sharp, crunchy, high microcontrast, with lots of fine detail, such as stubble. Others would see it as fake-looking, with harsh transitions, jagged edges, and lots of false detail, such as stubble that should be too small to see and jagged edges on the ear.

Personally, what bugs me about the aliasey one is it makes the skin tones harsher and makes people look older (not something actresses like , the hair is a jumbled mess rather than well defined, the logo on the shirt looks sharper but it's actually lost detail/ability to resolve what it is, and the shirt texture becomes harsh rather than natural.

I mentioned before that aliasy images also drop the apparent perceptual depth of field quite a bit which for motion makes it really hard to pull focus.

 

[Graeme Nattress]

And when you see the image in motion, aliases stand out worse than they do in stills too.

Graeme

 

[Wesley Scoggins]

I think perhaps there isn't any false color information in the image at all..

It just has such a high resolution it can see into another dimension. I think that is clearly the most logical conclusion. :dupe:

 

[chuck colburn]

Neil,

you are right and also that's a reason they needed psychoanalyst like Sigmund Freud to help them when they were sober :laugh:

For that Zeiss Sonnar Olympia 180 f/2.8 that became now a pretty rare lens to find and also was converted to Nikon F I do have an ordinary Nikon F to Canon EOS mount found @ Adorama.

If you want that I can have a look for that lens together with your sister @ Leica shop Wien no any problem and I'll find a time for that.

There is one of those Olympia Sonnars on ebay right now.

 

[Eisen Feuer]

Elsewhere on that page, the Siemens star chart does have moiré effect in the middle due to inadequate antialiasing.

So ultimately what is adequate antialiasing to you? If the wedge were correctly antialiased, would it just become a grey blob?

As I look at my own model, the color of each pixel defined only by statistical representation of the values of the sine wave, it becomes clear that while the purely mathematical model is technically 'perfect', It is not the ideal image you would wish to capture with your camera. So, Graeme's indication that the printed chart, not some computer model, is the 'perfect' representation to be matched makes sense:

The final judge is the eye.

 

[Andrew Thomas]

Thanks to those who continue to be civil during the discussion of the technical abilities of their various cameras out there. I have learned a heck of a lot about camera design in the last year as we've seen more and more companies put out their flavor of the week.

But it just continues to amaze me how so many Redusers, mostly a group of low budget filmmakers, who have experienced the internet "oppression" from the elitists in the industry in regards to their choice of camera would now turn around and treat the no budget filmmaker who is going out and shooting on their 5D/7D/GH1/etc in exactly the same disrespectful way.

Really shameful behavior.