Epic-W Test - low light, rolloff and lens comparison (Cooke, super baltar, leica)

[Jason Han]

Had a back and forth with Jason via Vimeo about it.

You'll notice the Cooke anamorphic and Super Baltar primes exhibit this the most while the footage shot with the Leica Summilux-C will barely if at all exhibit the chromatic aberration. This is more to do with the optical design and contrast ratio, which is common DP related stuff really.

He also experienced the CA with the Alexa, which is common for the Cookes, which I think are lovely lenses btw.

Solutions.

DRAGONcolor2 enhances blues through violets a good amount. Switching to DRAGONcolor will desaturate the CA to less noticeable levels. There's additional color developments on the horizon in a few weeks as well for RED shooters that may lend a hand here.

Real solution. Stopping the lens down by 2-3 stops is usually the key for spherical and vintage lenses to avoid CA at maximum and wider apertures.

Thanks for filling that in for me Phil, I think it's easy to avoid by stopping down on the wides, it's really the wide Cooke's that cause it I found.

From my experience - this is a combination of both the color science (which enhances the color palette) and the sheer resolution of the sensor - the later being the key contributor. IMHO it really comes down to the fact the these sensors resolve so much more in the same scenarios. At lower resolution (and I do not mean cropping the Helium - I mean using sensor with less resolution for the same area of exposure) the CA is simply blended with both sides of the high contrast areas and therefore not so well defined. At 8K we resolve so much more that we clearly pickup these lens imperfections in dedicated pixels without any blending. And this is then enhanced by the color science. Bottom line - there is absolutely nothing wrong with the Helium - the fault lies with the specific lenses. This simply comes down to the fact that the sensor is now so good that it picks up all the little imperfections in great detail and therefore won't let the lenses of the hook so easily. We really need new generation lenses specifically designed for these high resolutions. That's my take...r

Some great insight Peter thanks for sharing

 

[Stacey Spears]

To replicate the CA/fringing issue, I've been shooting low across a black, high detail texture which is edge lit from behind. Think speaker grilles, for example. The back-lit highlights on the fine texture tend to quickly show that purple CA/fringing. It's visible on the DSMC2 7" touch.

Please send some of your worst examples to Graeme in the form of an R3D so he can investigate. Always good to have test footage. Thank you for sharing.

 

[Nick Morrison]

Truth is, all Helium footage is currently being viewed via DRAGON color...it's like wearing the wrong prescription. We really don't KNOW what we're seeing. I'm pretty confident a lot of these issues will resolve themselves once Helium Color comes out...very soon.

 

[Jacob Ryan]

Hmm.... I haven't tried shooting a black object across a light source. This might be why I haven't seen the CA on either the 7" DSMC Touch or my Mac. I think I need to run another set of tests. I'll get back to you.

Sure. I'll pull some R3Ds tomorrow.

Generally speaking, I'm seeing the issue pop up with more severity on our "sharper" lenses. Let me admit that we have not tested with any high-end cine lenses yet. No need to beat around the bush.

I've been using our EF Canon L and Sigma Art lenses. Both of which frankly, have been solid performers for budget glass. The Sigma Arts, which have been sharp and solid on other bodies, show this issue heavily -- At least under DragonColor2.

Using DragonColor or RWG (in RedCineX) reduces the CA/fringing by a significant margin.

To replicate the CA/fringing issue, I've been shooting low across a black, high detail texture which is edge lit from behind. Think speaker grilles, for example. The back-lit highlights on the fine texture tend to quickly show that purple CA/fringing. It's visible on the DSMC2 7" touch.

Given how much the different color space settings are affecting the fringing, I think this is just a temporary issue. For the most part. To echo Nick's sentiments, I'm confident this will be pretty much tamed in upcoming firmware, once the sensor is a little more tuned.

And one final disclaimer: I'm totally willing to accept that our glass has a little CA, if you pixel peep. But, there is definitely something else going on here.

 

[Jacob Ryan]

Jason, did you already got your Epic-W from your S-W trade in? If so, boy that was fast!! Red must've produced a ton of E-W then.

 

[Jason Han]

Truth is, all Helium footage is currently being viewed via DRAGON color...it's like wearing the wrong prescription. We really don't KNOW what we're seeing. I'm pretty confident a lot of these issues will resolve themselves once Helium Color comes out...very soon.

Yeah I definitely see what you mean after comparing the dragon and helium, that's what I was thinking as well. Really hope that comes out soon, but overall color is great just minor quirks that need to be worked

 

[Jason Han]

Jason, did you already got your Epic-W from your S-W trade in? If so, boy that was fast!! Red must've produced a ton of E-W then.

Yeah! I contacted my dealer and they traded it in for me, dealers seem to be the way to go!

 

[Peter Majtan]

Truth is, all Helium footage is currently being viewed via DRAGON color...it's like wearing the wrong prescription. We really don't KNOW what we're seeing. I'm pretty confident a lot of these issues will resolve themselves once Helium Color comes out...very soon.

Nick - worth remembering that new color science is not going to fix the lenses. I am sure Graeme will do his woodoo magic and help patch up these lens imperfections the best he can - but that's what it always going to be - a patch up. The only way to eliminate this completely is to avoid using the offending lenses...

:sifone: Peter

 

[Nick Morrison]

Nick - worth remembering that new color science is not going to fix the lenses. I am sure Graeme will do his woodoo magic and help patch up these lens imperfections the best he can - but that's what it always going to be - a patch up. The only way to eliminate this completely is to avoid using the offending lenses...

:sifone: Peter

True, the CA will always be there. But I have to imagine Helium will respond to CA much like other sensors do, unless the smaller pixels are exacerbating it...

 

[Asif Limbada]

Jason, I would recommend testing it with RED Mg or Ti PL MOUNT. However, after reading some comments here that suggest not to use vintage or Cooke Anamorphics lenses with HELIUM is not only counter productive to objectively find the solution, but was a stupid suggestion IMHO.

However, we will soon recreate the same light setup and test it again, with different OLPFs to factor out the various variables scientifically along with Mini, SXT, Dragon, BMD, F55.

 

[Jon MIchael Puntervold]

Nice test, however I miss good outdoor shots in daylight and "harsh" sun. The Dragon sensor with skin-tone OLPF was perfect in this regard, really curious to see how the Helium works in sun and shadows. Nightime and indoor is covered nicely by most modern cameras already?

 

[Björn Benckert]

Nice test, however I miss good outdoor shots in daylight and "harsh" sun. The Dragon sensor with skin-tone OLPF was perfect in this regard, really curious to see how the Helium works in sun and shadows. Nightime and indoor is covered nicely by most modern cameras already?

The nice highlights of skin tone olpf is just that it works like a Natural density filler by about a stop and a half. So if if you close down that amount or ND you get pretty much the same result in the highlights with any olpf. So simply it´s depending on how low / high you expose. More to the left and you got better highlight protection, more to the right and you cover the darks better. Simple as that.

 

[Grzegorz Gill]

Nice,
Thanks for sharing!

 

[Tommaso Alvisi]

The nice highlights of skin tone olpf is just that it works like a Natural density filler by about a stop and a half. So if if you close down that amount or ND you get pretty much the same result in the highlights with any olpf. So simply it´s depending on how low / high you expose. More to the left and you got better highlight protection, more to the right and you cover the darks better. Simple as that.

Yeah, sort of, but I find the rolloff of the STH more gentle by default, you can tune it somewhat of course but the starting point is better with the STH...imho...