Is it just me . . .

[Tim Carroll]

Not sure how many of you have seen this yet. It was done by the folks at Zacuto here in Chicago this past year. A camera "shootout" where they compared a number of cameras, filming the same scene.

 

Zacuto's Great Camera Shootout '08

 

They shot all the footage of the same scene, trying to match the lighting and then took everything to Filmworkers Astro here in Chicago to color correct it all and try to match it all up so the look was the same on everything and attempting to make it all look like the 35mm film.

 

Here are screen grabs of a few of the different cameras. The first was shot on 35mm Kodak 5219 Vision 3 500T film w/Zeiss Superspeed 50mm

 

 

Now the Red camera shot w/Zeiss Planar T* 1.4/50mm ZF

 

 

Now the HVX-200 with Letus35 Ultimate shot w/Zeiss Planar T* 1.4/50mm ZF

 

 

And finally the HPX-170 shot w/Stock Zeiss Lica Lens

 

 

After they show the clips, the guys sit around and discuss them and kind of agree that they all look pretty much the same. Not to my eyes. I think they look quite different.

 

Naturally the film footage has great latitude and great color and narrow depth of field, as you would expect. I was shocked by the RED footage which frankly looks pretty awful when set side by side with the film. I also found the HVX-200 with the Letus Extreme looking pretty bad. The HPX-170, aside from the huge depth of field, looked pretty decent. I am just surprised at how different they all look, and how lacking most of them are when compared to the 35mm film.

 

In particular, look at the female musician's hair and blouse. The film brings out many more of the colors in her hair and does a much better job of bringing out the texture in her satin blouse.

 

There were other cameras in the test as well, I just picked these four because we discuss them on this forum quite a bit.

 

Just wondered what others thought of this test.

 

Best,

-Tim

[Tom Hall]

35mm isn't dead yet thankfully.

 

It's interesting to see how awful the Red looks in comparison, I guess we forget how far off digital is from 35 when we keep looking at the RED in comparison to things like the HVX200.

 

It doesn't help that the focus looks off in both the RED and the HVX200 letus, I wouldn't be surprised to hear the back focus was off. They both are focusing about 3 inches too close. (see the violin edge is in focus while her eyes are not)

 

Looking at the highlight on her arm the narrow latitude of the HVX200 and HXP-170 is very clear.

[Mike Lary]

Aside from the bad focus in all the digital camera images, the last two photos clearly show a different exposure level on the violinist's arm. It looks like she was moving in and out of a spotty light. The comparison isn't valid unless the conditions were the same.

 

I don't see how anyone could think those images look the same unless they aren't wearing their prescription glasses. The Red footage is clearly out of focus, gain-y, flat and desaturated. The color of the girls hair and the candles is clearly shifted in all of the photos. The detail level in the bricks changes vastly from one photo to the next as well.

[Saul Rodgar]

Is it just me or the woman's face looks fatter on video than on film?

 

I wonder if any big stars would realize that shooting on video renders them less attractive and would start writing in their deal memos that video is out when it comes to acquisition formats.

 

Definitely, film looks better all around to me, as well as most of us film-heads here, no surprise there. But to the untrained eye, they may look the same. Just like the woman in the picture may pick out a violinist who has 10 years experience versus someone who has 20. Most of us probably could never tell, but a professional in that field would, and right away. The same thing may be happening here.

[Francis Kuhn]

Thanks for posting, Tim

 

Take a look at the horrible magenta fringing on the RED candle flames. That's what I've been battling with digital stills for the past few years. It's a real mess on sheet metal, it can sometimes be partially corrected by using the chromatic aberration sliders in RAW processing, but it usually works on only horizontal or vertical lines (and usually not both at the same time).

 

-Fran

[Tom Hall]

I thought the magenta was caused by the lack of a hot mirror when using the red?

 

http://www.reduser.net/forum/showthread.php?t=13598

Edited February 23, 2009 by Tom Hall

[Saul Rodgar]

I thought the magenta was caused by the lack of a hot mirror when using the red?

 

http://www.reduser.net/forum/showthread.php?t=13598

 

The issue of digital color aberration as evidenced by the two above posts, makes me prefer film from that point alone. In the test, the two digital cameras that I had higher hopes for, the RED and the 5DmkII, looked the worse. If I had to choose by going for what I see on the Zacuto test only, due digital color aberration, I would just shoot 35mm film for the one million feature, and 16 mm in the Amazon, but I would take a smaller EX3 / HPX170 too, for impromtpu situations.

 

Since the paint menus of all the different digital cameras would hardly match easily, let alone out of the box, film still would win because film stocks are virtually identical _color rendition wise_ regardless of format. Lens introduced color rendition, grain / resolution and DOF would change on different film formats, true.

 

If one used 7201 with color matched lenses, footage shot on an Arri 416 would equal footage shot with an Aaton Xtera. Aside from pitch, wave and other mechanical aberrations, of course. Personally I hate to be fiddling with menus, and a DIT is not something I probably would be able to afford.

[Jim Hyslop]

Is it just me or the woman's face looks fatter on video than on film?

I think it's a combination of the bow's shadow on her left cheek, and she appears to be farther away from the film camera.

 

--

Jim

Edited February 23, 2009 by Jim Hyslop

[Jim Hyslop]

After they show the clips, the guys sit around and discuss them and kind of agree that they all look pretty much the same.

OK, so I just watched the video - the guys didn't say the results all looked the same, they said the results all looked good.

 

--

Jim

[David Mullen ASC]

It wasn't a particularly well-designed test -- the RED and the 35mm footage could have been shot at similar f-stops, but instead, the RED footage was shot wide-open on a Super Speed at T/1.4 -- no wonder it looks soft.

[Chris Keth]

It wasn't a particularly well-designed test -- the RED and the 35mm footage could have been shot at similar f-stops, but instead, the RED footage was shot wide-open on a Super Speed at T/1.4 -- no wonder it looks soft.

 

Seriously. They could have at least compared the formats with the sharpest lenses available for the format to at least make an attempt to level the playing field. I also would have appreciated shots at a few different stops, something like a T2, a T4, and a T11.

[Paul Bruening]

Any test that makes film look good and RED look bad is automatically an unfair test isn't it?

[David Mullen ASC]

Any test that makes film look good and RED look bad is automatically an unfair test isn't it?

 

Depends on the test, I'm sure you could design a "fair" one where film would shine in comparison (put blowing white window sheers in hot sunlight behind the violinist, for example), but this test wasn't it. You look at this test and your first impression is "why is the RED version so soft?" and then you notice that it is much shallower in focus than the 35mm test.

 

If you are trying to compare two cameras, you try to eliminate other variables as much as possible, like the lenses, the f-stop, etc.

[Tim Carroll]

David,

 

You have shot with the RED camera, is the "ASA" on that camera at something like 100 or 80? I am just curious as to why, if the light stayed the same, and the 500T film was shot at 24fps and f4, why they had to shoot the RED footage at 24P and f1.4? Some of the other cameras shot the scene at f2.8 (the EX3, XHA1, 5DMII, D90) and f3.4 (the HPX-170). Does the RED just need lots more light?

 

Best,

-Tim

[Freya Black]

I don't think it's just a trained eye thing. The first thing I noticed was the difference in depth of field and then I started thinking that the Red footage looked soft and kind of flat immediately after. I think anyone could notice this.

 

What shocked me was how good the HPX170 looked!

I actually really like the increased DOF in this shot and think it really adds something to the composition to be able too see all the details of the bricks behined. I like the composition in these shots and it made me think about how deep DOF can suit certain compositions and really add something.

 

The HPX170 is also the closest to the 35mm in having "more magical" (sorry to the uninitiated to use these technical terms here). The magical is very apparent on the violinists top in the 35mm version and the HPX170 has a lot more magical than the other digital cameras here. It also does very well in magical on the base of the candelabra where it seems to outshine even the film, but I think this is because the lighting has changed so much between the two shots.

 

The letus and red shots are lacking in magical but the red does have a sort of interesting quality of its own when you get used to it. However something seems to have gone badly wrong with the focus or something in those 2 pictures, so I may be judging the magical unfairly.

 

Overall I love the composition and choice of subject matter etc. It's very beautiful and fairly well thought out. It's a shame the testing methodology seems slightly out the window. They needed Phil there he would have sorted them out. ;)

 

love

 

Freya

[David Rakoczy]

This test (biased towards video as it is) reaffirms the obvious... Film is Film and there is nothing like it.

 

Thanks for sharing Tim.

[Tim Carroll]

the RED footage was shot wide-open on a Super Speed at T/1.4 -- no wonder it looks soft.

 

Actually David, it's worse than that. The RED footage was shot wide open on a Zeiss ZF lens. I have used the Zeiss ZF lenses on a Nikon still camera and really don't like the look of them. Not sure why they were chosen for this test. Maybe their RED camera had a Nikon mount instead of a PL mount.

 

Best,

-Tim

[Serge Teulon]

What shocked me was how good the HPX170 looked!

 

The HPX170 is also the closest to the 35mm in having "more magical" (sorry to the uninitiated to use these technical terms here). The magical is very apparent on the violinists top in the 35mm version and the HPX170 has a lot more magical than the other digital cameras here.

 

I agree with you Freya.

I think that the HPX170 is the best looking digital image on there. It is a a bit toooo sharp around the edges for my like....but by far on this test, the best one. (Taking into consideration the whole softness issue on the other images)

[K Borowski]

Seriously. They could have at least compared the formats with the sharpest lenses available for the format to at least make an attempt to level the playing field. I also would have appreciated shots at a few different stops, something like a T2, a T4, and a T11.

 

Well, the Zeiss SS is a very sharp lens, it's an issue of DOF in this case.

 

They should have tried to shoot them all at optimal stops (sharpness wide) and with an optimal allowance for rendering everything relatively sharp.

[David Mullen ASC]

I don't understand why the RED had to be shot wide-open, but apparently even the lighting set-ups weren't all the same, and they lit to some low level for one of the other cameras and then found it was too low for the RED, so they just opened up the stop all the way.

 

You can shoot the RED at 320 ASA and get decent results, and you could have rated a 500T stock at 320 ASA to shoot at the same f-stop & shutter speed if you planned it out. Or you could have rated the RED at 250 ASA in HMI lighting and used a 250D stock to compare it against, lots of scenarios.

 

Yes, perhaps the RED camera had a Nikon mount on it or something that prevented it from using the Super Speed that the film camera used.

[Chris Durham]

I saw this a while back and mostly dismissed it. I thought the Red footage looked like junk, and, like most here I was pleasantly surprised by the HPX. Really it wasn't a surprise that I liked it that much - everything I've seen out of that camera has surprised me. I would love to have seen some Sony footage compared because I've been using the Z5U a lot and am pretty impressed with it. Also conspicuously absent is S16 footage - I mean, theoretically we're looking at the stock, but the reduced size does change things a bit and 16 compares more evenly to HD (though still better I think). All in all an unscientific comparison, but still interesting. I'd like to see a more scientific comparison done independently, say by the ASC or similar organization, comparing top cameras at various tiers from different manufacturers with different capture methods and sizes. Make the playing field as even as possible and see how it looks.

[K Borowski]

I don't understand why the RED had to be shot wide-open, but apparently even the lighting set-ups weren't all the same, and they lit to some low level for one of the other cameras and then found it was too low for the RED, so they just opened up the stop all the way.

 

You can shoot the RED at 320 ASA and get decent results, and you could have rated a 500T stock at 320 ASA to shoot at the same f-stop & shutter speed if you planned it out. Or you could have rated the RED at 250 ASA in HMI lighting and used a 250D stock to compare it against, lots of scenarios.

 

I think, to be fair, you would have to make everything identical, same lens, same lights, same stop, same angle, etc.

 

And each video camera should be shot at their *optimal* ISOs & color temperatures not whatever is most convenient.

 

Didn't ASC Mag do their testing with changed lighting intensities, leaving everything else the same?

 

Another pet-peeve of mine: Why is '18/'19 the gold standard against which everything else is compared? Doesn't anyone shoot '01 outdoors, or 200T indoors anymore?

 

This reminds me of a "test" I saw in 2003 that proclaimed "Digital beats film!" They compared Kodak's (then) new 13.9MP DSLR to Kodak Gold 400, perhaps the grainiest, dirtiest, contrastiest film then on the market.

 

I just don't understand all of the gripes about film grain, with yet everyone always seeming to resort to the fastest stocks that they can get away with.

 

Hard to believe they were making do with 3 1/3 (or more) stops less than that in color until the late 60s.

 

Also, where's the 16mm camera?

[David Rakoczy]

I agree.. S16 was conspicuously absent... maybe they didn't want to show how it blows away the RED :lol: ... and yes, why are these always with 500t?... what about 200t & 100t?

[K Borowski]

I agree.. S16 was conspicuously absent... maybe they didn't want to show how it blows away the RED :lol: ... and yes, why are these always with 500t?... what about 200t & 100t?

 

Well, obviously, it is going to loose resolution-wise, but I bet it would still provide much better color space, and better flesh-rendition.

 

Ultimately bit depth is what everyone should be talking about.

[Gus Sacks]

I personally don't think any of these guys are really saying anything interesting or insightful. And what the hell is a guy like Joe Stunzi doing on a "panel" like this?