RED Komodo to start limited shipping soon for $6k

ahalpert said:
Oh believe me, I like what I see too. Both the body and the images. Probably not the right camera for me to buy next, though.
Click to expand...

Yeah, I won't be getting one either. At least not in the next year. Just going to be too many around, and easy to rent eventually, and I already have a couple friends with the Stormtroopers, and technically this thing is still in beta. But getting with that kind of image and global shutter in such a tiny low power draw camera, it really is addicting. It doesn't replace the Alexa look, no way, but it is still very much a true cine camera by today's standards.

They really tried to meet that 2lbs body too. The thing is so light, that it almost feels... like a toy, but it is the beta version, still. Yet being so lightweight is part of what makes it so nice.
 
James0b57 said:
I must admit, it is a different thing once it is in your hands. That's when it hits home, that this is a tiny power sipping RED.
Click to expand...

It doesn’t quite make sense. I’ve never wanted a RED and the Komodo is kind of the opposite of what I want/like in a camera body, but I really dig it.
 
Zyla chart DR test @ 800 ISO log G310 gamma.

50207742346_e98ed04caf_b.jpg


Zyla chart DR test @ 800 ISO REC709 gamma

50208028597_c33d742559_b.jpg
 
Razz16mm said:
Some casual comparison tests coming from Tim Daust. Komodo v Gemini v Helium.

Click to expand...

That a very interesting test. Komodo clearly outresolves Gemini (by more than I'd expect from the jump from 5k-6k). It also seems to be the cleanest of the three noise-wise. DR seems to be essentially the same between them based on those crop-ins.

Very impressive.
 
Grug said:
That a very interesting test. Komodo clearly outresolves Gemini (by more than I'd expect from the jump from 5k-6k). It also seems to be the cleanest of the three noise-wise. DR seems to be essentially the same between them based on those crop-ins.

Very impressive.
Click to expand...

The Gemini may also be using a larger sensor, so that would exaggerate the resolution differences. With a larger sensor, foliage in the distance is "farther" away.

The Komodo would technically have the best downsample to 4K of the three cameras. The Gemini would have the better anamorphic crop and full sensor frame rates. After 60p, the Helium and Komodo begin to lose their resolution, and this causes some inconsistency in the IQ.

The Skintone Highlight Priority OLPF does look better than standard, in particular in tungsten, and it fixes the orbs issue, however, the standard OLPF and Low Light OLPF offer an extra stop of sensitivity, and on the Gemini that goes a long way.

Having only played around briefly with the Komodo, I realized how spoiled I was with the sensitivity of the Gemini sensor. Makes ambient night shots possible, even with anamorphic T2.4 lenses. The Komodo, is a stellar camera though, and it is just uncanny at times, for how well the battery life lasts and the size and weight.... then still getting some of the best IQ and global shutter of any camera, that is pretty rad. the first "GoPro" that doesn't feel like a compromise.
 
Last edited:
James0b57 said:
The Gemini may also be using a larger sensor, so that would exaggerate the resolution differences. With a larger sensor, foliage in the distance is "farther" away.

The Komodo would technically have the best downsample to 4K of the three cameras. The Gemini would have the better anamorphic crop and full sensor frame rates. After 60p, the Helium and Komodo begin to lose their resolution, and this causes some inconsistency in the IQ.

The Skintone Highlight Priority OLPF does look better than standard, in particular in tungsten, and it fixes the orbs issue, however, the standard OLPF and Low Light OLPF offer an extra stop of sensitivity, and on the Gemini that goes a long way.

Having only played around briefly with the Komodo, I realized how spoiled I was with the sensitivity of the Gemini sensor. Makes ambient night shots possible, even with anamorphic T2.4 lenses. The Komodo, is a stellar camera though, and it is just uncanny at times, for how well the battery life lasts and the size and weight.... then still getting some of the best IQ and global shutter of any camera, that is pretty rad. the first "GoPro" that doesn't feel like a compromise.
Click to expand...

Do we know the native ISO of the Komodo sensor yet? Is it dual ISO, or a high sensitivity sensor like the Gemini? I see 800 in the tests but is that it's best ISO? What does it look like at 2500 ISO?
 
puredrifting said:
Do we know the native ISO of the Komodo sensor yet? Is it dual ISO, or a high sensitivity sensor like the Gemini? I see 800 in the tests but is that it's best ISO? What does it look like at 2500 ISO?
Click to expand...

iso800 is a good rating for the Komodo.

anything over iso1600 is looking pretty noisy. No doubt someone will say that iso3200 with denoise is usable, but I think that can be a little misleading at times. For example, if someone ETTR a stop to get clean iso3200, then isn't that the same thing as just exposing iso1600 correctly?
 
400-800 ISO is pretty much the optimum light sensitivity range of silicon based sensors. Pushing beyond that is starving the sensor of light no matter what games you play with gain and noise reduction after the fact. IQ and DR suffer.
Except for Gemini, all Reds shift exposure index for ISO changes, the first stage gain structure is fixed for optimum IQ at native ISO. Changing ISO on camera just gains up the monitor feed and writes different exposure metadata to the raw file header. It doesn't affect how the sensor captures light.

There is a lot of headroom for highlight recovery with Komodo at 800 ISO. Graded this sample with IPP2 color science in RCX.

50226285742_8f396bafbd_k.jpg
 
Well, Pure, there you go. No need to take my word for it. Looks like my impressions were a little conservative.

Seems the Komodo is definitely improved in the shadow colour shift at higher iso's.

This test may have been done with the STHP OLPF, in which case that means that the Gemini and Helium could be a stop more or less sensitive with a OLPF swap. Meaning, they could be worse if the test was done with the Standard OLPF's.

My impression when shooting the Komodo along side the Gemini, was that the Gemini in LL mode was cleaner.

I also find, for obvious reasons, that the downsampled ProRes 2K on all REDs is a bout a stop cleaner

The Gemini can be less forgiving to underexposure than the Helium, since the pixels are fewer and bigger, you'll see noise a hint faster than the Helium, yet, the Gemini will overall perform better in low light. Personally, I am ok with some noise, despite rating very conservatively on public forums, because there are many variables. Like tungsten light can effect noise in the blue channel faster, so a camera that can handle iso1600 in sunlight, maybe only get to iso1000 in tungsten before the blue channel starts acting up. just depends. I find Tungsten light more challenging to shoot, as there is typically a "source" in the shot, requiring additional top end DR, and light levels tend to be lower, almost asking for iso1600 or 2000, which further challenges a CMOS sensor.
 
Last edited:
In a lot of tests, I have seen the Gemini skew purple in the shadows as the iso goes up, but I find that the noise above 1600 in standard mode is enough that one wouldn't be going to iso 3200 in standard mode. So that colour drift is not really ever experienced, as no one seems to shoot that far up.

Now the Gemini in Low Light mode at is3200 is fine. Mostly clean, but no room for a lot of under exposure. And I think up to iso 5000, things are still tasteful.


But in any case, the Komodo in Daust's gets looks good up to is3200, in part due to the better colour shift performance.
 
Last edited:
When Canon released the C300, they said at ISO 12800 it had the same noise as 35mm film, which is fine by me. I'm sure the C300 was applying ample noise reduction at that point, though.

In the tests by Tim Daust, the cameras looked fine to me up to 12800, probably cleaner than 35mm film. It's hard to tell through Youtube compression, though.

DISCLAIMER: I don't do this anymore for a living and am not worried about pleasing clients, some of whom may be allergic to the slightest speck.
 
combatentropy said:
When Canon released the C300, they said at ISO 12800 it had the same noise as 35mm film, which is fine by me. I'm sure the C300 was applying ample noise reduction at that point, though.

In the tests by Tim Daust, the cameras looked fine to me up to 12800, probably cleaner than 35mm film. It's hard to tell through Youtube compression, though.

DISCLAIMER: I don't do this anymore for a living and am not worried about pleasing clients, some of whom may be allergic to the slightest speck.
Click to expand...

Interesting. When me and my friend did iso12800, it looked bad, even on the little top screen. But I never got it in post with the Chroma denoise.

One thing I have noticed, and you can see it in the Daust comparison, is that black gets a lot of those swimmy marbled chroma denoise patterns. It can wreck havoc on natural skin tone. And make black objects and dark shadows look bad. Like using a cheap sensor (with insane highlight retention! haha)

But I am loving how much less colour shift there is going on in the shadows.
 
Last edited:
I'll be honest though, I didn't expect the Komodo to be this good, in general.

Man, Combat, you can just run this thing in ProRes 2K mode, buy yourself an extra stop of sensitivity.
 
You must log in or register to reply here.
Back
Top