MX vs. DRAGON w/ Low-Light OLPF

[Alexander Manning]

Got a couple of questions for my fellow REDUsers!

The first thing you should know is that over the last year I have absolutely fallen in love the MX sensor. Fantastic in the darks, pretty good in the highlights. There's no denying it's still an incredible camera. Now 99.9% of the time shoot in 5K HD and am working toward a 1080p Master (pretty standard in the commercial world). I very rarely stray away from the 16x9 aspect ratio. I DO crop quite a bit and I DO try to make full use of the extra resolution. I understand the need for 6K BUT it's not that big a deal to me - 5K is more than enough.

Recently my Studio took the DRAGON upgrade and purchased both OLPF's (LL & STH). Besides the initial disappointment in the ST performance in the mids and darks, overall I've been very pleased with the camera. I do however have a couple of things I wanted to get opinions/the facts on.

While my real world tests of the DRAGON are few (we haven't had it very long) I am a little confused as to the real difference (besides $11 Grand) are from the MX.

I understand the new OLPF system and the differences between the LLF and STH. Shoot LLF when you don't have tons of available light or you're working with a lot of shadows. Shoot STH when you have light at your disposal and you are working with the highlights.


Here are my questions:

1) Is the low-light performance (stops into darkness) of the DRAGON w/ the LOW-LIGHT OLPF equal to or better than the Mysterium-X sensor?

2) Is the low-light performance (stops into darkness) of the DRAGON w/ the SKIN-TONE HIGHLIGHT OLPF equal to or better than the Mysterium-X sensor?

3) Is the general dynamic range of the DRAGON w/ the LOW-LIGHT OLPF equal to or better than the MX?

4) Is the noise performance of the DRAGON w/ the LOW-LIGHT OLPF equal to or better than the Mysterium-X sensor?

5) The color is obviously better on the STH than the LLF. Does using the DRAGON with the LOW-LIGHT OLPF negate the color advantages of the DRAGON sensor over the MX?

6) Are there any differences in compression quality on the DRAGON vs. the MX? (i.e...is 12:1 the new 8:1?)

7) Are there ANY disadvantages to shooting 5KHD on the DRAGON vs. shooting 5KHD on the MX? Noise? Available framerates? Compression?

8) What is the general consensus on D.E.B? Are there any advantages/disadvantages to checking it?


I really appreciate this forum and all the time and energy you guys put into it - I not only can say I'm a better shooter for being a part of it, but a more efficient one!

Thanks in advance!

Best,
Alex

 

[Brian Merlen]

1/2 is debated by some but i think so..

3/4 yes more range

5 its close but STH is slightly better, very close

6 compression looks better but everyone still shoots low as possible

7 slightly different FOV, smaller pixels

8 gets rid of LLO speckles, does not work with rocket=long transcodes

ps phil posts good tests on this stuff...

 

[Phil Holland]

ps phil posts good tests on this stuff...

Cheers Brian.


Just to add. The general color difference between the Low Light Optimized OLPF and the Skin Tone - Highlight OLPF is fairly small. Meaning you can work with both easily on the same project without concern. The Color and Highlight Roll-Off are the "extra drop" nicer on the STH OLPF and that tends to be what's on my camera a lot because of that.

REDCODE has indeed been re-tooled or re-written or optimized for Dragon. It holds detail better at higher compression rations across the tonal and color range. I wouldn't say 12:1 is the new 8:1 exactly as it's not really a good way of defining. Over the last year most have still be shooting with standard 5:1 to 8:1 REDCODE RAW for 24fps projects. You'll need to test that for yourself to see what's optimal for general shooting for your needs. I'm mostly at 5:1 myself, 4:1 for anamorphic, and overcranking is where I tend to dip into the higher compression ratios.

 

[Robert Ruffo New]

Got a couple of questions for my fellow REDUsers!

The first thing you should know is that over the last year I have absolutely fallen in love the MX sensor. Fantastic in the darks, pretty good in the highlights. There's no denying it's still an incredible camera. Now 99.9% of the time shoot in 5K HD and am working toward a 1080p Master (pretty standard in the commercial world). I very rarely stray away from the 16x9 aspect ratio. I DO crop quite a bit and I DO try to make full use of the extra resolution. I understand the need for 6K BUT it's not that big a deal to me - 5K is more than enough.

Recently my Studio took the DRAGON upgrade and purchased both OLPF's (LL & STH). Besides the initial disappointment in the ST performance in the mids and darks, overall I've been very pleased with the camera. I do however have a couple of things I wanted to get opinions/the facts on.

While my real world tests of the DRAGON are few (we haven't had it very long) I am a little confused as to the real difference (besides $11 Grand) are from the MX.

I understand the new OLPF system and the differences between the LLF and STH. Shoot LLF when you don't have tons of available light or you're working with a lot of shadows. Shoot STH when you have light at your disposal and you are working with the highlights.


Here are my questions:

1) Is the low-light performance (stops into darkness) of the DRAGON w/ the LOW-LIGHT OLPF equal to or better than the Mysterium-X sensor?

2) Is the low-light performance (stops into darkness) of the DRAGON w/ the SKIN-TONE HIGHLIGHT OLPF equal to or better than the Mysterium-X sensor?

3) Is the general dynamic range of the DRAGON w/ the LOW-LIGHT OLPF equal to or better than the MX?

4) Is the noise performance of the DRAGON w/ the LOW-LIGHT OLPF equal to or better than the Mysterium-X sensor?

5) The color is obviously better on the STH than the LLF. Does using the DRAGON with the LOW-LIGHT OLPF negate the color advantages of the DRAGON sensor over the MX?

6) Are there any differences in compression quality on the DRAGON vs. the MX? (i.e...is 12:1 the new 8:1?)

7) Are there ANY disadvantages to shooting 5KHD on the DRAGON vs. shooting 5KHD on the MX? Noise? Available framerates? Compression?

8) What is the general consensus on D.E.B? Are there any advantages/disadvantages to checking it?


I really appreciate this forum and all the time and energy you guys put into it - I not only can say I'm a better shooter for being a part of it, but a more efficient one!

Thanks in advance!

Best,
Alex

My extensive tests showed that

1 - About 3/4 of a stop better than MX

2 - Much less good than MX - and blacks clip hard, they don;t fall off into noise gently like MX

3 - Better than MX, about 1.5 - 2 stops - hard to say exactly because they behave a bit differently. Some say more than that but really 2 stops is a lot more

4 - A bit better than MX

5 - I saw little difference in color performance between the OLPFs. Low Light OLPF has pink orb issue around bright objects, and "SKin Tone" does not. Low light somewhat better than MX, but MX, barring IR issues, is very good since Redcolor 3, and now even better with Redcolor 4. In my experience grading can make them all match exactly in terms of color precision and look. MX Skin Freckles issue, better on Epic MX than Red One MX, seems not really present on Dragon. This is rarely mentioned, and a big thing. Freckles are very hard to fix in post, as they are the same color as shadowed skin.

6 - Compression on Dragon a bit more efficient, but not by much.

7 - 5K on Epic is a wider field of view of your lens, it's like a bigger sensor (it is a bigger crop size, as Dragon pixels are smaller)

8 - DEB works great - work really, really well removing noise and has few artifacts that are readily visible. Often none at all. However, clipped blacks are still just black, so no miracles. Low Light OLPF still gives you much better reach into darks.

9 - Don't forget that Dragon has a nearly global shutter. This is also a big thing often overlooked, which can be a real plus in some cases over MX, depending on what you are shooting and how.

 

[Carl Dieker]

My extensive tests showed that

9 - Don't forget that Dragon has a nearly global shutter. This is also a big thing often overlooked, which can be a real plus in some cases over MX, depending on what you are shooting and how.

Hi Robert,

I understand the concept of "nearly global shutter", but it is the first time I heard about it being attributed to the Dragon sensor. It would be great to see some stills comparing the MX-vs the Dragon sensor regarding this claim. Is there maybe a thread about this? If so, please point me in the right direction.

/Calle

 

[Nick Morrison]

Hi Robert,

I understand the concept of "nearly global shutter", but it is the first time I heard about it being attributed to the Dragon sensor. It would be great to see some stills comparing the MX-vs the Dragon sensor regarding this claim. Is there maybe a thread about this? If so, please point me in the right direction.

/Calle

Dragon has twice the "read" time of MX sensor, meaning it cuts down on rolling shutter twice as well. Toia noticed results of this in his early tests last year (he shot a bunch of racing cars and noticed the wheels spun almost perfectly).

Hope this helps...