5D vs Red aliasing etc fun analysis PICS.....

PappasArts

PappasArts

Veteran
In another thread there was questions of how the aliasing on the 5D is vs RED etc.

So I did some extraction from the RED vs 5D Mark II test done recently done by Richard Darge.

It's just for fun and not an exact nor even approximate let alone perfect analytical scientifically approved by a committee of our own educated peers of unattainable stature......

It's just a fun test.. Anyhow; here it is........




RVSM.jpg


RVSM2.jpg


RVSM3.jpg


Files sourced from: http://vimeo.com/5160288
Courtesy by Richard Darge
http://www.PappasArts.com



.
Michael Pappas
http://www.pbase.com/Arrfilms

*****Anamorphic DSLR Lens Test Images... links:
http://www.dvxuser.com/V6/showpost.php?p=1638289&postcount=97
http://www.dvxuser.com/V6/showpost.php?p=1637705&postcount=84
Facebook & Myspace:
http://www.facebook.com/people/Michael-Pappas/573417404
http://www.Myspace.com/PappasArts
http://www.PappasArts.com
AOL AIM { PAPPASARTS2 }

.
 
Last edited:
Red is still king when it comes to image quality. It will be interesting to see this test in a few months once the next batch of vdslrs come out.
 
There is no doubt the RED shines in this department (and hopefully the Scarlett with be equally robust), but as a side note, I have simply never seen footage as bad as these 5DII examples, not from my own 5DII (and I have shot literally hundreds of clips since early December 08 when I first got the 5DII.) nor from any other users.

I have seen nothing even remotely close to the poor quality of the 5DII examples used here.


As the sensor size on these cameras are different, I take it you are not enlarging the 5D footage to match the FOV of the RED ?

That would of course simply reduce the resolution of the 5D footage and would render your comparison somewhat absurd.
 
Last edited:
Isaac_Brody said:
Red is still king when it comes to image quality. It will be interesting to see this test in a few months once the next batch of vdslrs come out.
Click to expand...

Outside of the 60D from Canon coming and maybe a Nikon , I dunno if it will be a few months. I was thinking more like a year or more before we see that above gap get closer... I could be wrong though.


.
 
Lee Wilson said:
I take it you are not enlarging the 5D footage to match the FOV of the RED ?

That would of course simply reduce the resolution of the 5D footage and would render your comparison somewhat absurd.
Click to expand...

I'm not new to doing these strip tests. I used to do it in film!

No, if I did that. The 5D would look even worse.
 
Lee Wilson said:
So with the first clip (FU'S PALACE) - the RED would appear to have a wider FOV than the 5DII ?

And yet the sensor is smaller ? How did you achieve this ?
Click to expand...

He was zooming back an forth in that shot... This just shows how either camera would handle the same given linear lines etc.

Let us not forget, as above stated: It's just for fun and not a perfect analytical scientifically approved by a committee of educated peers of unattainable stature......

It's just a fun test..
http://www.PappasArts.com
 
PappasArts said:
He was zooming back an forth in that shot... This just shows how either camera would handle the same given linear lines etc.

Let us not forget, as above stated: It's just for fun and not a perfect analytical scientifically approved by a committee of educated peers of unattainable stature......

It's just a fun test..
http://www.PappasArts.com
Click to expand...


Better hide your neck! Gonna get yourself torn a new one. Hehe. 60D I want to see.
 
PappasArts said:
He was zooming back an forth in that shot... This just shows how either camera would handle the same given linear lines etc.
Click to expand...

Ah ! I see, that's why the FOV is so similar.

PappasArts said:
Let us not forget, as above stated: It's just for fun and not a perfect analytical scientifically approved by a committee of educated peers of unattainable stature......

It's just a fun test..
Click to expand...

. . . . but with footage that poor it is hardly even a test ! :2vrolijk_08:

Keeping to the same image (FU'S PALACE) - is that a 1:1 crop ?
 
I tried to find some stuff I had shot with lines just off the horizontal (about the hardest test for the 5D's not particularly great binning technique and lack of decent anti-aliasing) and whilst not being directly comparable to your examples - I could find nothing even close to the poor quality show in your clips - in fact they look more like SD scaled up to HD - it really is that bad - if I were to get footage like that even from my HV10 I would be compelled to send it in for a check over ! (and possibly burial).

Anyhow, this is what fine lines tend to look like on the 5DII - not as well rendered as the RED1, but certainly nowhere near as bad as your terrible examples (all images except obviously for your own JPEG at the top - are grabbed straight from the H.264 / straight out of the camera and I simply grabbed a horizontal section of interest - 1:1 crop)


http://img13.imageshack.us/i/horizontals.jpg/


Click^^^^^^
 
Last edited:
that one of your fav artists would be w/ wilson.. seems you're not comparing apples w/ apples.. mean.. same FOV

this and about settings.. sharpness etc including focus

that's why

apart that.. the tests are welcome.. no matters who wins
 
Ok !

One more quick test.

To get anywhere neat the ropey quality of your 5DII samples, I took a 1:1 crop of section of an unaltered H.264 clip - I then (using photoshop / interpolation set to bicubic) scaled it down to 50% wide and 50% high (effectively throwing away 75% of the information in the image) - I then scaled it by 200% (back to it's original size) - and even after this pixel beating it still does not look as bad as your 5D clips - although it is starting to get there.

Top: Your 5D clip.

Left: My 1:1 5D clip from the original H.264

Right: My 5D clip after it's sordid journey. (as you can see it is starting take on the appearance of FU'S PALACE !!)



65992979.jpg



My conclusion ? Your examples are not representative of the 5DII, and certainly not anywhere good enough to judge any qualitative issues such as comparative aliasing quality, this is not to say I disagree with your conclusion - that the RED beats the Canon when it comes to aliasing - that is so apparent it hardly needs to be pointed out, my HV10 beats the Canon in this department.
 
Last edited:
filmmaker's gang said:
that one of your fav artists would be w/ wilson.. seems you're not comparing apples w/ apples.. mean.. same FOV

this and about settings.. sharpness etc including focus

that's why

apart that.. the tests are welcome.. no matters who wins
Click to expand...


I don't understand what is being said here, the last two lines make perfect sense, but the first two are a little too obscure for me ! :(
 
As I wrote:

It's just for fun and not an exact nor even approximate let alone perfect analytical scientifically approved by a committee of our own educated peers of unattainable stature......

It's just a fun test..


Michael Pappas
http://www.pbase.com/Arrfilms

Anamorphic DSLR Lens Test Images... links:
http://www.dvxuser.com/V6/showpost.php?p=1638289&postcount=97
http://www.dvxuser.com/V6/showpost.php?p=1637705&postcount=84
Facebook & Myspace:
http://www.facebook.com/people/Michael-Pappas/573417404
http://www.Myspace.com/PappasArts
Arrfilms@hotmail.com
http://www.PappasArts.com
 
Last edited:
Lee Wilson said:
I don't understand what is being said here, the last two lines make perfect sense, but the first two are a little too obscure for me ! :(
Click to expand...

I thought I had to much coffee when I first read that.....

I don't know either.. Drunk posting possibly, or he was having a stroke while writing; not good!

Drop the bottle sir, and step away from the keyboard.....



.
 
PappasArts said:
I thought I had to much coffee when I first read that.....

I don't know either.. Drunk posting possibly, or he was having a stroke while writing; not good!

Drop the bottle sir, and step away from the keyboard.....



.
Click to expand...

Maybe a harrowing coffee+drunk+stoke combination . . . :shocked:
 
Last edited:
PappasArts said:
As I wrote:

It's just for fun and not an exact nor even approximate let alone perfect analytical scientifically approved by a committee of our own educated peers of unattainable stature......

It's just a fun test..
Click to expand...


My conclusion was equally fun:

"Your examples are not representative of the 5DII - and certainly not anywhere good enough to judge any qualitative issues such as comparative aliasing quality"

:2vrolijk_08:
 
You must log in or register to reply here.
Back
Top