Doing some tests with RED Scarlet X, Sony F3 and Canon C300.

[Chris Luker]

I have been invited to do some testing with my Scarlet X, a Sony F3 and a Canon C300 this week.
Should be interesting...
Are there any tests any of you would like to see done?
______________________________________________________________________________
______________________________________________________________________________

My pathetic try at a 'blog' about the shoot is posted at http://www.utahcamerarentals.com.
Working on getting the original files uploaded somewhere permanent... stay tuned!

 

[Ryan De Franco]

Thanks very much Chris, here's my wishlist...

1) Highlight / shadow detail on a human face. Light with a high contrast ratio, see how skin tones hold on both ends.

2) Compression stress-test. Shoot the C-300 with a bunch of moving leaves, water, field of wheat, fireworks, anything with a good deal of motion.
** Fast camerawork would also work, shots wouldn't be identical, but steadicam or fast handheld running with two people in frame would do the trick

3) Night exterior street. I've found my Scarlet to do incredibly well under a single streetlight, and I love the amount ofnoise, but everyone's dying to see this test. Please blackshade first!

4) Shoot Redlog, S-Log f3, and whatever Log the Canon has, put them all through four different extreme grades, see where each falls apart.

Can't thank you enough, most interested in #1 and #2...

 

[Ben Pluimer]

I'd love to see what happens when you drop a Scarlet-X directly onto a C300 from about 6 feet up.

EDIT: And the skintone highlight test too, I guess.

 

[jimhare]

Just make sure you are involved in the post as this is where most tests fall over. Someone with vast experience in processing R3Ds properly needs to be there to protect all of us.

I remember one test that they evaluated the HD-SDI signal instead of processing the R3Ds!

 

[Lliam Worthington]

Hey Chris,

Cool. Look forward to seeing what you guys come up with.

LOW LIGHT!
Just joking. I'm over low light. I do have three basic requests if your asking for suggestions.... some movement. Not many camera comparison tests seem to show any movement. Maybe it's just me, but I also like to see cameras in motion, (objects people too) even just some simple pans. Second... faces, watching tests without any people ever featuring in them at all, I never feel satisfied.

Lit (and varied lighting) scenes. But lit. Even if it's basic. I completely understand why low light, natural light/practicals make up the bulk of tests, but most of us light don't we? Seeing how these three cameras resolve and handle a few different lighting setups is much more ideal for me.

Does the F3 have S log. Like to see these three cams, including Scarlets HDRx go head to head in dealing with something challenging.

Oh, and perhaps you could shoot a 90 minute feature with each one so I can really...

:leaving:

Be Well mate

Lliam

 

[Chris Luker]

Just make sure you are involved in the post as this is where most tests fall over. Someone with vast experience in processing R3Ds properly needs to be there to protect all of us.

I remember one test that they evaluated the HD-SDI signal instead of processing the R3Ds!

Kinda what I'm afraid of going in. But, I will try and get files shot from each camera so that I can do my own post with them.
We will be using HD-SDI, but just to monitor... I think. How clean is the HD-SDI output?

Oh, and perhaps you could shoot a 90 minute feature with each one so I can really...

Lliam

No problem! A 90 minute feature should take about 100 minutes to make right!? (I get that one all the time with bands that want to record an album in the studio - "But it's only ten 5 minute songs... we should be done in an hour!")

 

[Sergio Perez]

1- Color gradation, Highlight rolloff. How all handle overexposure and underexposure
2- Banding. Want to see how all handle situations when Banding can be an issue
3- Chroma Key- How do they handle a Key
4- Low Light. We know the Scarlet is not the best in this particular sector, but would like to see a comparison in REAl world, same time, same place test on this. Look for a situation with a variation in light intensity and color temperature- a typical city street shot. Make sure to try and do a noise reduction pass on the red footage since there's no internal noise reduction in the Reds compared to the other cameras.

5- I want to see a weight comparison in their most portable configuation: Scarlet with side handle, redvolt, sssd and a lens vs the canon with the same configuration and the F3 with the same configuration.

6- Want to see 60FPS 2K Scarlet compared to 60fps 720P in both cameras. Same lens. If Field of view needs to be compensated on the Scarlet so be it. Add an extra 48fps 3K shot just to compare if possible!

7- If possible shoot a face with good lighting and shoot it with the Scarlet at 4K with redcolor 2 and redgamma2 (add 3 if its released this week, that is! ) , the C300 with its cine (non log) setting at 1080p internal codec and external recorder, and F3 with external recorder and internal codec without Slog- this to get what each manufacturer accepts as their optimal "baked in" look.

8- Same as above but all in their respective Log spaces- redlog film, C Log for Canon, S-Log for Sony. Without Color correction and after color correction. A demanding Dynamic range scenario is a must for this!

EDIT- 9- Would like to see a Panoramic shot of a landscape with lots of detail which could present aliasing issues. Would like to see also this scene in Scarlet's native 4K and would like to see both the C300 and the F3 UPREZED to 4K. It will be very difficult for us to see in the internet but you could report on this!

Well, hope this is not too much This will obviously not be a scientific test but will be fun to watch. No comparison between Scarlet, a 4K camera, vs the others, but its still fun to put the racing car vs the road cars on certain "common" roads...

 

[Gary Paul Walker]

I'd love to see what happens when you drop a Scarlet-X directly onto a C300 from about 6 feet up.

EDIT: And the skintone highlight test too, I guess.

Not only is this hilarious, but its also a practical test. Plastic vs Metal - where are you getting your value for money? enquiring minds want to know....

 

[Chase Thompson]

No problem! A 90 minute feature should take about 100 minutes to make right!? (I get that one all the time with bands that want to record an album in the studio - "But it's only ten 5 minute songs... we should be done in an hour!")

Definitely been there Chris, you COULD record an album in an hour for them. Pop open your laptop press record and let them go at it with the internal microphone, all channels set to unity. Lol. Music video? One take with the iPhone. Bam. I always keep a reference on hand to show bands what an "album in an hour" sounds like versus one done RIGHT.

But on topic with the rest of the thread, I second the previous requests. It's nice to see these kind of tests done in real applicable situations. Looking forward to it.

 

[Bruce Allen]

This would be awesome! I would love to see / hear about:

MOST IMPORTANT:
1. Skintones & detail in human skin & hair. That's what I go to the movies for - to watch people... I want to see how the camera reacts to the human face under both tungsten, daylight and fluorescent, traditional 3-point lighting and maybe exteriors too. Put some Macbeth charts in the back too, please

2. How the cams react to people with and without makeup. Try to find some problematic makeup (IR reflective) and also some nice makeup and see what happens.

3. Maybe renting a film camera is too much, but please also take stills on actual 35mm film with a 35mm film still camera. And with a Canon and Nikon DSLR.
Personally, I like film skintones the most. So I would love to see which camera we can grade most closely to match the richness of the film skintones.
My second favorite skintones are on Nikon DSLRs... again that seems like a fun "brand neutral" thing to match. See if we can grade Scarlet / C300 / F-3 footage so that it looks like a nice Nikon DSLR still we took under the same conditions.

4. Or find a scene from a favorite movie (again, maybe shot on film?) that looks really great and natural... and try to recreate that scene lighting-wise... then grade all 3 cameras to see if you can match it to the film print.

After all - that's the real question to me: how close can these cameras get aesthetically to the richness of capturing humans on film...

MUCH LESS IMPORTANT:
1. Not much need to shoot ultra-low light... but imagine you were shooting wide shots for an indie night scene and couldn't run a generator? Maybe try that...

2. Color detail - point the camera at some very detailed colorful stuff, eg patterns that have similar luminance but lots of chroma detail.

3. Latitude on moving shots - (probably HDRx won't work but give it a go?

4. Shoot Scarlet 3K and 2K high speed vs Canon and Sony... try shooting 1080 60i on Canon C300 instead of 720 60p... if you de-interlace intelligently in post, you should get better results from the 60i slowed down than the 720p, maybe.

5. Weight and overall handling

6. Ease of use and stability of cameras

9. For the Canon, I would try one with C Log... and one with C Log tweaked to be a little bit more contrasty (basic idea is that because of the 8-bit recording, you'd want to dial in the contrast a little even when recording Log, so that you are still spreading your image detail across a good range of pixel values. If everything is recorded to mid-grey, we'd get banding when we expand it out).

If you're running out of time... screw all of the non-log or non-RAW recording formats.

I would be vaguely interested in seeing Sony non-S-log. Because that gives you several price points to compare: eg F3, then C300, then F3 with S-log... but don't care that much because I'd never shoot in non-log mode.

Bruce Allen
www.boacinema.com

 

[Sanjin Jukic]

Such a great idea!!!

I'm getting c300 in about a week or so....

Also have a Scarlet on order from December 2, but this will take months to get camera...

Then all the time trying to evaluate Scarlet-X.

Just to tell you that I had already MX upgrade with R1 and was testing it a bit...all it was here about two years ago...

Because RED doesn't make its own sensor then take them several years to get a new one, unfortunately too long....

That's why I also ordered c300 to get hands on the latest and best high ISO performance...

4K is not curtial thing for me because here in Europe especially in Austria they just freshly upgraded to HD (720p) for broadcast.

But I was already evaluating and comparing c300 upscaled MXF to 4K with Scarlet-X 4K RAW examples from RED users.

Looks pretty good and at 2K you don't see any difference between those two cameras just because I think that 2K is optimal resolution from both cameras to get best result...

C300 has much better skin tones than Alexa also Epic/Scarlet as I saw a lot of examples and movies from all those cameras.

Alexa is the worst example especially in the movies like a Melancholia and Hugo that were both a sort color of skin tones torture from my point of view!!!

Also let you know that to get super clean high ISO cameras like an Alexa or Epic/Scarlet can just dream about it...

When you would test c300 try to adjust and follow setting and recommendation from

BBC guy Alan Roberts who made it at his c300 test recently>>>

 

[Sergio Perez]

Sanjin, Red doesn't make their own sensor? That's news to me!

 

[Patrick Tresch]

Hello Sanjin,

Just curious what you think about the compression artefact on the C300. I like the spec. but I was biaised and reluctant about the codec they use.
Are you happy with it? Your input is greatly appreciated.

Thanks

Pat




PS:I hope this thread won't be closed again... ;-)

 

[Sanjin Jukic]

Hello Sanjin,

Just curious what you think about the compression artefact on the C300. I like the spec. but I was biaised and reluctant about the codec they use.
Are you happy with it? Your input is greatly appreciated.

Thanks

Pat

PS:I hope this thread won't be closed again... ;-)

Pat,

I do not have any intention to advertise Canon here but just in the case of comparison

according to legendary Larry Thorpe from Canon C300 use the 50Mbit, MPEG2 422 straight off the camera 10bit footage packed in 8 bit MXF file.

Source>>>

also C300 has got an excellent reply from Academy Award (Oscar) cinematographer Anthony Dod Mantle (BSC, DFF, ASC) >>>> .

Alan Roberts from BBC wrote:

"This camera performs well...If, as Canon claim, the camera’s sweet spot is at ISO850/+2.5 where there is headroom of 800% (3 stops),
this would stretch the available exposure range to 12.5 stops, but this is available only by using the Canon
Log gamma-correction curve."

Those are all the quotes and thoughts from the experts and probably I can't add more.

 

[Niko Pueringer]

Shoot something in the shadow with sunlight in the background, and vice-versa.

 

[Alan Carpenter]

Such a great idea!!!

I'm getting c300 in about a week or so....

Also have a Scarlet on order from December 2, but this will take months to get camera...

Then all the time trying to evaluate Scarlet-X.

Just to tell you that I had already MX upgrade with R1 and was testing it a bit...all it was here about two years ago...

Because RED doesn't make its own sensor then take them several years to get a new one, unfortunately too long....

That's why I also ordered c300 to get hands on the latest and best high ISO performance...

4K is not curtial thing for me because here in Europe especially in Austria they just freshly upgraded to HD (720p) for broadcast.

But I was already evaluating and comparing c300 upscaled MXF to 4K with Scarlet-X 4K RAW examples from RED users.

Looks pretty good and at 2K you don't see any difference between those two cameras just because I think that 2K is optimal resolution from both cameras to get best result...

C300 has much better skin tones than Alexa also Epic/Scarlet as I saw a lot of examples and movies from all those cameras.

Alexa is the worst example especially in the movies like a Melancholia and Hugo that were both a sort color of skin tones torture from my point of view!!!

Also let you know that to get super clean high ISO cameras like an Alexa or Epic/Scarlet can just dream about it...

When you would test c300 try to adjust and follow setting and recommendation from

BBC guy Alan Roberts who made it at his c300 test recently>>>

From this PDF, "HDSDI and HDMI outputs are also both 8-bit depth, although the data-stream is 10-bit. This does not appear
to have any detrimental effect on the camera performance."
What does that mean? Is 10-bit available? Why would there be this qualifier about performance? It seems like a reassuring non sequitur.

 

[Uli Plank]

Try them all on fine patterns to check for aliasing – the F3 is probably the worst (I didn't test C300 yet).

Have some colored neon in the frame at night. The F3 looks deeper into the shadows than RED, but ruins your highlight colors by then.

Check RS, a simple test is a backlit fan at constant speed shot with short exposure times.

 

[Mark Toia]

I have been invited to do some testing with my Scarlet X, a Sony F3 and a Canon C300 this week.
Should be interesting...
Are there any tests any of you would like to see done?

Try This,

Don't down res the 5k EPIC footage to 1/5 of its true resolution (HD), try doing a full 4k output and view on a 4k monitor or 4x2k monitors stitched... Then put that up against the HD Sony and canon footage.

Thats only if you want to do a real side by side test...

 

[Mark Toia]

I have been invited to do some testing with my Scarlet X, a Sony F3 and a Canon C300 this week.
Should be interesting...
Are there any tests any of you would like to see done?

Try This,

Don't down res the 5k EPIC footage to 1/5 of its true resolution (HD), try doing a full 4k output and view on a 4k monitor or 4x2k monitors stitched... or play it over 2x 30 inche monitors.
Then put that up against the likes of the HD Sony and canon footage.

Thats only if you want to do a real side by side test...

 

[Chris Luker]

My pathetic try at a 'blog' about the shoot is posted at http://www.utahcamerarentals.com.
Working on getting the original files uploaded somewhere permanent... stay tuned!
:wink5: