New Epic-W vs. Alexa Mini test

[Phil Holland]

Arri, assuming 4000k 800iso like the others. You can't tell but the blownout lights in Log-C and Log3G10 are almost identical, i.e. the following two images have matching blow out points. Check how the shadows fall.

Helium, lowcon rolloff 4, no other grade.

Correcting tint seems OK to me, lifting shadows should only harm an image so I'd rather they match. Easy to crunch the arri down, harder to lift helium up. I wonder if resolve's SDK is as up to date as RCX?

By the way, all my images are 4k if you open them in another tab.

I'm doing my best to stay out of this chaos, but thank you Anton. You are showing what "needs" to be seen in this comparison.


On a subtle Morpheus-like riddle, I wonder if folks have realized that when you have more resolution you are capturing not only more detail, but also more color detail.


I'll go back to my cave now.

 

[Michael J Brennan]

Key lights 12 inches from a wobbly subject, where moving an inch is not insignificant.. to test skin highlight response?
ISO of two different camera sensors with radically different pixel sizes, matched for "convenience" (presumably the convenience of setting a common stop rather than mess with filters?)


This particular setup is not a sound basis for an empirical test.

Were all Arri shots done first then RED second? What was the time difference between setups? Blood pressure varies and can make a subtle difference if there is no makeup. (ha!..only half joking with that one)
Also perspiration can be a factor, how hot was it 12 inches from those two incandescent bulbs. Was the black jacket for effect or was it actually cold?

Arri has dual processors to handle their dual exposure/readout so in theory more noise but greater DR, possibly one explanation/guess of why it is noisier even though it has larger pixels?


Mike Brennan

 

[Anton Shavlik]

It does look like for that test they had bulbs on dimmers and used that to adjust exposure. I might be wrong though.

Here's another test, Helium at 4000iso with slightly less blown highlights than arri.

Arri, presumably 800iso

Check her hair as well as the window for a sense of dynamic range.

 

[Stacey Spears]

I wonder if resolve's SDK is as up to date as RCX?

Which build of RCX? The preview build with the new IPP2 pipeline is not available in other apps yet.

 

[Marc Wielage]

Then i guess you haven't graded much of the material, where "cinematographer" completely blows his/her exposure and then blaming DIT as in "he said it was fine". Unfortunately, badly exposed material happens much more often than you think and that is when you're happy to have some extra maneuvering room...

This is sadly true. I've often told DPs, "when in doubt, underexpose by 1/2 stop." Or in video terms, knock it down by at least 5-10 units. We can generally save the faces in post that way.

What is pretty subjective is the rest of the camera and workflow. I'd like to see the Alexa doing HDRx for example. Or framing a 6K crop of an 8K source. Or doing internal wireless focus via an app on an iPhone. Or timelapse. And so on.

I've done plenty of Alexa projects where we were able to compress the dynamic range and preserve all the details of blacks and whites in post without having to resort to HDRx. There are always specific operational advantages of one camera over another; if I were you guys, I wouldn't take this so personally. It's just a camera -- it's not a way of life, and it's not an indictment of anything you believe in or choose.

 

[Ben Scott]

Absolutely.

It's just a chisel, not the sculpture. I'm sure yourself and Jake and the other colorists on here have dealt with a ton of brilliant Arri and Red footage and a ton of awful Arri and Red footage.

At the same time though, if I was watching a shootout of chisels I'd like to know there was a continuity of methodology behind the testing, and that the person conducting the test doesn't have a ton of experience with chisel A and zero experience with chisel b.

 

[luigivaltulini]

Helium is certainly less than Alev in DR-term,
as it is compared to DRagon.
How Helium is the king of the resolution.
Nothing new.
Alexa is good for those who like it, like RED, Sony and Panasonic etc. etc ..
Pick your right tool ... and then your colorist will give the final touch to your movie, a nice grading that takes the feel through the colors

however, I allow myself to say that I can not wait the RED to start producing quickly VIsta Vision of DRagon sensor.:devil::devil:

 

[Kemalettin Sert]

Helium Eyes

Arri eyes

Clips are A001C0001_161229_R and B001_C018_1229R4

How come upscaled shitty 10 year old 3.4K Arri sensor seems almost same sharpness with 8K Helium? even left eye sharper little bit sharper.

 

[paulcurtis]

If you want to rate the camera differently you can just download their raw files. They have a file for every stop or so, compare what you think is fair. Arri without a doubt has a stop or two more range, not sure it wins on any other battleground.

No. If you want to rate the camera differently you shoot differently. They lit to a different exposure, that's the whole point. That daylight out of the window is going to clip on a camera that is natively more sensitive unless you stop the light down to match. You wouldn't take 500ASA film and rate it differently unless you were intentionally pushing it.

cheers
Paul

 

[paulcurtis]

This is sadly true. I've often told DPs, "when in doubt, underexpose by 1/2 stop." Or in video terms, knock it down by at least 5-10 units. We can generally save the faces in post that way.


I've done plenty of Alexa projects where we were able to compress the dynamic range and preserve all the details of blacks and whites in post without having to resort to HDRx. There are always specific operational advantages of one camera over another; if I were you guys, I wouldn't take this so personally. It's just a camera -- it's not a way of life, and it's not an indictment of anything you believe in or choose.

Oh i wasn't expounding the use of HDRx in day to day, i just meant that as a whole package if you needed to you have more 'features' to use on Red to get out of sticky situations. If you needed to shoot int and ext and couldn't light the int then you're out of luck on an alexa but you *could* rescue that on the Red.

That is a pro and a con. All this OLPF stuff makes the system look complicated. The truth is you can just stick in the standard and just shoot and forget that there are others, because in the laws of diminishing returns the extra isn't that much. But the flexibility is there if you really wanted to eek out that last few % points.

I've had really tricky Alexa footage and noisy shadows and god knows what. It's how it's shot most of the time as well all know.

I just had an issue with this being a shoot-out and it being totally subjective and shot wrong for that purpose. If it was 'i had to choose between a mini and helium and this is how i choose...' that's fine!

cheers
Paul

 

[paulcurtis]

How come upscaled shitty 10 year old 3.4K Arri sensor seems almost same sharpness with 8K Helium? even left eye sharper little bit sharper.

You're seeing the Lens more than anything else. For there to be a difference in focus on her eyes shows that the DOF is pretty shallow.

Also motion blur, the shutter is only open 50th second (or 48 or whatever), in photographic still terms that's pretty slow. Works great in motion but not so much in stills.

cheers
Paul

 

[David Litchfield]

The resolution, or lack of it seems to be the biggest problem here. It's not like they were using old soft glass.

I guess Arri can wait another year before bringing out the Alev 4.

 

[Hrvoje Simic]

I see dpx and r3d files for download.

Are there raw files for Alexa somewhere ?

Right, as you say, "usable dynamic range." Once you reach the boundary of what is useful (blown plastic face), then what possible use is it to show results well past that boundary and certainly well past what a competent cinematographer would shoot?

The use is knowledge.

Which makes a cinematographer become competent by understanding what happens with the image throughout the range of scenarios per particular camera.

That way he knows what to expect and how to light and expose.
Without that you get to unrealistic expectations, "fix it in post", some folks having a bad day during that process and some when the unrealistic expectations founded on pure guesswork finally meet reality.

 

[Aaron Lochert]

No. If you want to rate the camera differently you shoot differently. They lit to a different exposure, that's the whole point. That daylight out of the window is going to clip on a camera that is natively more sensitive unless you stop the light down to match. You wouldn't take 500ASA film and rate it differently unless you were intentionally pushing it.

cheers
Paul

They ramped through a bunch of different apertures. So just compare a T4 Alexa shot with a T5.6 Helium shot. Remember to ignore sharpness/color/vignetting though, because of inconsistent T stops.

 

[Ben Scott]

At which point what is it precisely you're testing?

 

[Graeme Nattress]

How come upscaled shitty 10 year old 3.4K Arri sensor seems almost same sharpness with 8K Helium? even left eye sharper little bit sharper.

From a different shot, a quick comparison of resolution / detail.

Graeme

 

[Kemalettin Sert]

Thanks Graeme !!!

 

[paulcurtis]

They ramped through a bunch of different apertures. So just compare a T4 Alexa shot with a T5.6 Helium shot. Remember to ignore sharpness/color/vignetting though, because of inconsistent T stops.

But did they in their tests and conclusion? I don't think so...

I will take a look at some point, bit busy at the moment!

cheers
Paul

 

[Anton Shavlik]

They ramped through a bunch of different apertures. So just compare a T4 Alexa shot with a T5.6 Helium shot. Remember to ignore sharpness/color/vignetting though, because of inconsistent T stops.

That's what my posts are doing. The helium shots have deeper depth of field, I looked for shots with similar blown highlights.

Helium is absolutely seeing more detail, scaled to 4K and exported as a jpg is just not the best way to see it. Load them into resolve and it's quite obvious.

 

[Alan Gordon]

Absolutely. The helium is the better option if a 4k finish is required. These particular tests, however, are not designed for observing sharpness. Red has had the edge at resolution since at least the Dragon if not Mysterium-X, so it wasn't really worth testing.

I'm excited to test the full IPP2 when it is released to the full SDK and into DaVinci and will likely start by comparing the Alexa shots to Red shots that are 1-2 stops down.

Hopefully some of the issues on the extreme edges of exposure (and white-balance) will be mitigated.