Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Dragon??? So after nearly 10months what do we think? NATIVE 2000ASA no 18stops no ..

Did not compare F65 to Dragon (but i shoot some F65 last year and yes it´s great) - but i am very happy with Dragon and everytime i shoot with i am looking forward to have the stuff in the color suite. Regarding low light, there are cameras that see more in the dark, but i am fine with the performance of Dragon. With DEB the "noise" is improved when you are forced to push
i will post some night stuff 2morrow.
 

Attachments

  • Palais_Mak_of.jpg
    Palais_Mak_of.jpg
    84.6 KB · Views: 0
  • PALAIS DE L´EUROPE.jpg
    PALAIS DE L´EUROPE.jpg
    97.8 KB · Views: 0
With all the respect i have for Red, i don't think that the resolution is a valid argument, who can really tell the difference between a 6k image and a 4k one? As 90% of theaters (at least here in Paris) diffuse in 1080p.. People will talk about crop possibilities and downsampling quality over a native 1080p, let's be serious, and let's talk about "what really matters": color, dynamic range, low light capabilities, noise..

Red Dragon is at her early age, i'm sure we will see some crazy improvements in the next few months, but to be honest, and i've already said it in other threads, no Dragon footage made me go crazy yet, in fact, all i've seen until now could have been made by the Mx sensor.. In term of color, creaminess and texture, Alexa is still above for me..


For people who tried, what are the dragon low light capabilities vs Alexa?



Let me get in my small town with 4 cinemas are equipped to 4k and you're in paris 1080?!?!?


That's why maybe you do not see the difference ....
Talk about foley color creaminess?! and then. if you ask someone to test some Dragon vs. Alexa low light!
But then you've never used a Dragon!
Excuse me, how do you say be better Alexa?
I advise you to do as a first experiment Dragon vs. Alexa, take 2 people with brown hair in a dark room you do a good backlight, then go into a beautiful film equipped with a 4k projector, looks like your Alexa make a single block where you do not distinguished even a hair, while Dragon be all there .. this for the resolution since you say you not be important ...


Please let me love Alexa, I 2 I had until the end of last year, but now I'm sorry but Dragon and especially on the higher resolution and color.


I did a quick comparison only and only for the noise, between dragon and alexa new OLPF 800 iso.
http://www.reduser.net/forum/showthread.php?119034-250-ISO-Looks-great/page8


just my thought
 
@ Mendes

Dragon ... it´s just a small jpg (i will upload this later on my flickr), but this image was shot on friday in strasbourg in 6K WS, and i scaled it to 4K and saw it on a 4K screen. maybe Alexa is above for you, i see Dragon in front.
 

Attachments

  • RHIN_SHIP.jpg
    RHIN_SHIP.jpg
    101.4 KB · Views: 0
Yes, most theater in Paris are 2k. Sadly, you have to pay extra to get 4k in selected theaters only . There are like 5 in the Paris area.

Despite that, the difference can be seen. Especially in the front row to the middle of the screen for average viewer. And for experimented viewer, even in the last rows.
I can spot 4k downsampled versus 2k HD, so I guess everyone here professional should be able to.
 
@ Mendes

Dragon ... it´s just a small jpg (i will upload this later on my flickr), but this image was shot on friday in strasbourg in 6K WS, and i scaled it to 4K and saw it on a 4K screen. maybe Alexa is above for you, i see Dragon in front.

That's a sick shot Jan!

I used to shoot a lot of action and for a year now, I'm shooting more and more fashion and commercial.
I upgraded my Epic-M to Dragon for many reasons.
First I wanted to try the new sensor extensively and not only on a rental period.
Second, some of my clients were asking for it quite a lot... Even if most of them won't see any difference between Dragon and M-X.
The day after I received my upgraded camera, I shot this: https://vimeo.com/97373933
Laetitia, our model, has a really pale skin, our location has a lot of variation in terms of contrast and lightning, we shot the pool scene at mid-day and, for all that, the camera performed better than I hoped.
Skin tones are definitely improved, latitude in the highlights helped me a lot to expose most of these shots then deal with them in post.
2 months ago, I exposed at 800 iso when shooting outside, I now rate the camera and my light meter between 250 and 500 iso max.
I shoot a lot of high-speed and I was limited by the compression allowed by my RedMag which is a bummer, I now have a 512GB Turbo SSD and it's definitely a plus.
After my test shoot, I was thinking of uploading on Youtube a 4K version as the 1080p looks great but the original edit is super clean.
I'm now waiting for Redcinex with DEB to be available to do that just because I want to check if there's a big difference with what we already released.
I watched al these tests and I read a lot of posts on Reduser, I agree on some points but if you ask me, Dragon is better than MX, I can see these improvements every time I use the camera.
 
That's a sick shot Jan!

it is ;-)

i can´t watch it enough on a 4K Screen, of course it is graded in Da Vinci, but the Raw of dragon delivers luxury data ... you want to touch the water - it´s ultra real ;-) will post a bigger one later the week ... nice screensaver heh heh
 
F65 with 8k sensor is the only camera capable of producing a true 6k images. Just because Dragon produces bayered 6k image doesn't make it into a final 6k image. It's just not possible...

F65 is not 8K. They count two staggered sensor rows at a time to arrive at 8k - the spec is 8192 x 2160 in a 17:9 aspect

That's 8.8 MP green, 4.4 MP red, 4.4 MP blue
Not true 4K per colour channel in the way they'd have you believe, and not capable of delivering 6K debayered either.

Dragon 6K has 9.7 MP green, 4.85 MP red, 4.85 MP blue - 5K debayered, or 4K debayered in the S35 zone.

Of course my monochrome Dragon is 6K no questions asked ;) hehe.
 
Each of these cameras has its strengths and weeknesses. Our job is to figure that out and deliver the indended results to our clients. Is the F65 an 8K camera. No fucking way. And Sony's deliberate marketing strategy to obfuscate the truth about how their sensor functions just pisses me off. Does the F65 take fine pictures. Absolutely. Stand behind that and not blatently make up new math to bolster your resolution claims.

Sergio, as for your original topic. I've been shooting on Dragons for over a year now since I first tested one of the prototypes in my junkyard spec shoot. Every iteration of the camera has its own particular nuances. I've had to continually re-evaluate my stance on how I treat the camera for each of these iterations. I don't mind doing that. Because EACH offered some unique properties. Dragon is an evolving platform. Have we seen the last of its features and performance characteristics, I don't think so.

MOST people will agree that the first OLPF demonstrated that the ability to have clean high ISO exists in the sensor. The OLPF 2 has also shown how the highlight performance and color rendition can be maximized. I believe for anyone not content with the current offering, then having interchangeable OLPFs, or going back to the original OLPF will be a solution. I'm also interested in what this poses for future developments with the sensor. Maybe a middle-ground OLPF will be introduced, maybe pre-ND'd OLPFs will come out in each of these technologies... For me, that's appealing. If I'm going to be shooting days and days of bright exteriors, perhaps I want a OLPF2/ND.9 combo on my camera... If I'm doing tons of murky low-light city nighttime shots, perhaps OLPF 1 or a future version of that will be on my camera.

I bought into Epics pretty early on in their release and saw that camera evolve and advance for 3 years or so into a robust workhouse with an image I constantly marveled at what I could do with. Dragon is no different. With the form factor, resolution, modularity, price, ect ect... I cannot complain one bit. Sure I'm adapting to each change and alteration of that platform, but then again, no one whined at Kodak for changing the characterists on a new film stock offering. AT LEAST these changes are somewhat OPTIONAL... Not many other cameras can offer that.
 
Each of these cameras has its strengths and weeknesses. Our job is to figure that out and deliver the indended results to our clients. Is the F65 an 8K camera. No fucking way. And Sony's deliberate marketing strategy to obfuscate the truth about how their sensor functions just pisses me off. Does the F65 take fine pictures. Absolutely. Stand behind that and not blatently make up new math to bolster your resolution claims.

But then doesn't it equally bother you that RED still claims Dragon is rated at 16.5+ stops and 2000ISO in the same way that Sony is saying their sensor is 8k? Sure, Sony's math may be *ahem* 'generous', but then again so is RED's (if 2.5 stops of DR is unseeable and 2-3 stops of gain is unusable due to noise).
 
That's a sick shot Jan!

I used to shoot a lot of action and for a year now, I'm shooting more and more fashion and commercial.
I upgraded my Epic-M to Dragon for many reasons.
First I wanted to try the new sensor extensively and not only on a rental period.
Second, some of my clients were asking for it quite a lot... Even if most of them won't see any difference between Dragon and M-X.
The day after I received my upgraded camera, I shot this: https://vimeo.com/97373933
Laetitia, our model, has a really pale skin, our location has a lot of variation in terms of contrast and lightning, we shot the pool scene at mid-day and, for all that, the camera performed better than I hoped.
Skin tones are definitely improved, latitude in the highlights helped me a lot to expose most of these shots then deal with them in post.
2 months ago, I exposed at 800 iso when shooting outside, I now rate the camera and my light meter between 250 and 500 iso max.
I shoot a lot of high-speed and I was limited by the compression allowed by my RedMag which is a bummer, I now have a 512GB Turbo SSD and it's definitely a plus.
After my test shoot, I was thinking of uploading on Youtube a 4K version as the 1080p looks great but the original edit is super clean.
I'm now waiting for Redcinex with DEB to be available to do that just because I want to check if there's a big difference with what we already released.
I watched al these tests and I read a lot of posts on Reduser, I agree on some points but if you ask me, Dragon is better than MX, I can see these improvements every time I use the camera.

Dragon loved the Model and brought its best out of box i am sure ;-)
 
I'm not sure what the currently claims are since the OLPF change. Certainly the original OLPF was much more inline with those claims. Personally, I'm reserving judgement until the fallout of this whole situation settles down.
 
I've got to take issue here-

The only good way to measure DR is with a Xyla-21 - and Dragon show 16+ stops when you use the chart properly and view the material properly - ie not as a still frame JPEG online, but projected on a good screen.

On ISO- if you expose appropriately ISO 2000 makes perfectly good images. One of the most important concepts we teach at Reducation is proper use of ISO, and it's one of the most common misconceptions. Push ISO in low light and all you'll see is noise, pull ISO in bright light and you get an unpleasant roll off to clipping. I know working pros aren't keen on being told they're doing things wrong, but you must approach the camera the right way-

fact is, Dragon like all Red cameras has only iris, filtration and shutter controlling exposure, so ISO should be a curve shaping tool, not a gain/exposure control to make light out of dark.
 
Yeah, Sony's own technical white papers have all the true specs on the F65. It's actually not even a full 6K sensor. The problem is their marketing department counts each of the staggered corners as two pixels in between the primary center-lined row of pixels they are actually counting. So, in a nutshell 5.7K becomes 8.6K horizontally. It's still a 17.7 megapixel Bayer CFA sensor with square pixels and a 17:9 aspect ratio. The catch is that the pixel raster is rotated 45° off kilter, but that has no bearing on anything other than marketing. The real reason for the rotation has to do with manufacturing quantities of their CMOS wafer yield. What's funny is their marketing department only counts the horizontal pixels this way and not the vertical, but I guess that's just so they could claim it's 8K. And their marketing likes to claim that by rotating the pixel raster by 45°, it suddenly gains a 30% increase in resolution.

I have yet to see any tests, charts, real-world production footage, etc.. that shows an F65 with more nominally measured resolution than Dragon. And it can't without distorting the physical laws of the universe because it has fewer pixels.

All that said, the F65 is a fine camera and I've shot on it a bunch over the past year. Personally I think it's ergonomically clumsy and just awkward all around. Then again, I hear the same thing about the EPIC all the time from other AC's and operators. To each their own.



I just hope people will take a look at the fine print and finite details and call it for what is. And what it is, is not an 8K camera...
 
Touche. Truth is, for all their continual tweaking/fixing (and potential headaches said tweaking causes), RED is still the only ones offering affordable digital cinema. More over, I hate how cumbersome and heavy other digital cinema platforms are (requiring so much more support, whilst ruling out gimbals and things that offer creative shots).

For me, the debate isn't so much about how much better Dragon is than MX/Alexa/F55/etc, it's more about how much value it has for $9500 (which could literally purchase an entirely shootable Scarlet package or F5 package, or three A7s packages with 4k recorder, etc. for b-cam.) I'm inclined to believe I could make more money and/or save more time on set, having two Scarlet-X Packages, than one Scarlet-Dragon package, particularly when Dragon isn't blowing the doors off of Alexa in terms of performance (which at least would have given the camera clout come negotiation time.)

@John, that might be true, but that same methodology shows MX at ~15+stops (and presumably Alexa at 16+ stops) and that's just a bogus/inflated number. Can anyone point to an example where Dragon has - hands-down - surpassed Alexa's 4 year old sensor in terms of real-world DR? I haven't seen it, and even if it exists, I bet it isn't by 2.5 stops over Alexa's 14. That's a haggard situation to be in, 'yeah, Dragon is 16.5 stops, but it's no more useable than 14... and you have to light it differently and/or push/pull in a proprietary way, that isn't intuitive in the same way Sony, Canon, and Arri (or even MX) are, or how film has been for the last 100 years...' Explaining that to clients is not gonna win any favours. And if that's the state Dragon is in, staying MX seems a little more digestible, no?... At least until the dust settles anyway.
 
Last edited:
Mike

you make an interesting point. 2 Scarlets on set with two redrockets in a sonnet box. Taking care of churning out dailies is a very solid high end workflow and it's very manageable in terms of data rate and render power. Heck. You could put four redrockets in a box now and have a super hype system for supporting those cameras while for dragon it's a boat anchor.

The r3d the dragon creates rendered existing tech useless, so when it comes to hard costs what you suggest with two Scarlets is a very viable, cost effective and efficient idea and production strategy.

Battistella
 
Haha, yeah I know David...

...Still kinda want a Dragon though.

I think Scarlet-to-Dragon is a harder sell than Epic-to-Dragon, as $9500 is less than half the cost of an Epic (originally and after price drop), whereas it's the full cost of a shootable Scarlet-X package. It's a tough call made tougher by the growing pains Dragon is going through.
 
Last edited:
I hope the solution is a little bit more elegant than telling customers to use other cameras if they don't like what they see just now or alternatively that the problems are due to them practicing their trade wrong.

Both PR disasters.
 
I don't believe Sony ever claimed F65 to be an 8K camera. It would be disingenuous to do so, as it is not even technically possible to create 8k image out of 8k bayered sensor. My point is what is what's good for the goose is good for the gander- Red can't claim true 6K images out of Dragon as well. Let's not accuse Sony of " deliberate marketing strategy to obfuscate the truth about how their sensor functions" while dismissing the same of Red. I'm sorry to see Sergio's post degenerate into Sony vs Red marketing.
The question was asked:
What camera shoots 6K?
And I answered- Sony F65.
Let's move on...
 
... and you have to light it differently and/or push/pull in a proprietary way, that isn't intuitive in the same way Sony, Canon, and Arri (or even MX) are, or how film has been for the last 100 years...' Explaining that to clients is not gonna win any favours. And if that's the state Dragon is in, staying MX seems a little more digestible, no?... At least until the dust settles anyway.

Please go into more detail about what's proprietary about lighting for Dragon, especially versus lighting for film. I'm asking because I've read differently.

Timur posted the following in another thread that directly contradicts that notion.

I think the new dragon OLPF light response is awesome. It lets me shoot the way i used to on 35mm. Expose for the shadows, with fill, or open up half a stop, then print down. Dont worry about the highlights, they will be there and look awesome.

RED shifted the balance of DR upwards, much more like 35mm. Dont forget how effing grainy 35 is, and the way to combat that is to add some fill, over expose a bit, the nprint down. Shoot at 800, add a bit of fill, then develop at 400/500ISO and you will be golden on New OLPF. Just like film. This is not a Documentary camera.

I was worried at first, till i started working to its strengths, and now its just unbelievable. Dont ever wanna go back.
 
Back
Top