Welcome to our community

Be a part of something great, join today!

Log in
  • 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 :)

I'll be honest - Dragon disappoints me

Status
Not open for further replies.
I hope they blow the competition away, combining the best stuff from different cameras this would be the areas to focus on for me:

- lower noise levels and better low light sensitivity (in the c300 range)
- better color reproduction (alexa)
- higher dynamic range (35mm film range)
 
Max M. said:
I hope they blow the competition away, combining the best stuff from different cameras this would be the areas to focus on for me:

- lower noise levels and better low light sensitivity (in the c300 range)
- better color reproduction (alexa)
- higher dynamic range (35mm film range)
Click to expand...

Sure...

To replace 35mmm DR, you need a lot more than any single 35mm stock can handle, as you cannot just swap your 800 for a 50. The other way to get the same flexibility would be a 15 stop chip with built in IRND's to compensate for the different exposure levels.

So in practical work a chip needs to surpass any given stock to get the same result as multiple stocks.

As to colors...

Have you ab'ed much lately?

Cheers!

G
 
Daniel Soderberg said:
I know it's a little outdated, but director Rian Johnson's "Red Facts" outlines the ways Red can improve to create a better image very well: http://www.rcjohnso.com/REDFACTS.html
"Red samples at 12-bits and sends 12-bit linear files to the recorder [now 16 with Epic]; F23/F35/Genesis sample at 42-bit and send 30-bit logarithmic files to the recorder."
Click to expand...

Calling these cameras 42 bit makes for a very off-kilter comparison. Summing together bits per channel is not equivalent to representing bit depth in the way we'd normally understand it. The Epic has 16 bits per pixel when delivered at native resolution. Downsampled to match HD delivery size using the method above the Epic has over 100 bits of data per pixel. Practically speaking it'll get quantized back to 10 bits anyway.

Upscale the Sony HD to 5k similarly and you have just about 6 or 7 bits per pixel. Realistically you're not going to have a low bitdepth image though, just one that has less data spread among more pixels.

As always the proof of the pudding is in the eating.
 
Daniel Soderberg said:
As people have said, I don't know if the Dragon sensor will address more than the only 3 specs they've let us in on... but I agree that instead of focusing on more resolution (moving to 6k), more time should be spent on increasing color depth (more bits per channel), less compression, and less noise.
I know it's a little outdated, but director Rian Johnson's "Red Facts" outlines the ways Red can improve to create a better image very well: http://www.rcjohnso.com/REDFACTS.html
"Red samples at 12-bits and sends 12-bit linear files to the recorder [now 16 with Epic]; F23/F35/Genesis sample at 42-bit and send 30-bit logarithmic files to the recorder."
This being said, Red One and Epic are amazing cameras and I am so thankful for what Red has made affordable for us!
Click to expand...

That article is, of course utter bollocks. It's not just outdated, it's wrong, factually incorrect and written in an absurdly biassed manner ignoring even contemporaneous evidence to the contrary.

Graeme
 
Gunleik Groven said:
Wooot????

LOL, guys.

Noone but very few hacve seen the Dragon, and Epic is not a bad cam as it is...
Click to expand...

Epic is a GREAT cam as it is, which is why I don't even Red's mission of finding meaningful ways to make it way better. This thread is only an attempt to help by eliciting feedback.suggestions, and provocative title to get many people to look. (13 pages later - they did...)
 
Paul Ellington said:
Constructive critiques are fine, but when you are always complaining it turns into white noise.
Click to expand...

I almost never complain about Red products and I often defend them. I am not complaining. I am describing my future needs as I estimate them.

to me noise si "awesome! great!" over and over again. What could anyone learn from that?
 
Rob Ruffo said:
What we really wanted was very good performance at extreme high ISOs. It's not every shoot that needs this, but it does open up possibilities that would have been expensive or impossible to do without it. (I'm NOT advocating doing away with lights, just using lower amp heads in tough locations where power and the ability to disturb local residents are at a premium.)

We also wanted a bigger field of view. Give us full frame. That's cool, and useful.

Give us even more precise color fidelity. There is no end to how good color rendition can be. More of that is always better.

Instead the main touted feature is even higher pixel density, and "6K".

6K is a problem. Its' not a good thing. 5K already nearly chokes most fo our workstations here, and we never, EVER say "Gee I wish this was higher resolution" Computer tech is advancing slowly these days. It's not just our computers, but the computers of all the people we hand in footage to that would have to be upgraded. My head hurst just thinking about it. If anything we would prefer slightly FEWER pixels per square inch of field of view, not more. Or just keep it the same.

Most camera makers have understood that the "megapixel race" has become silly. In my personal opinion, at the current 5K, Red should too.

I think proof is that there is very little excitement over Dragon. Epic solved a real problems ("Gee, I love my Red, but I wish it were lighter and started up faster, and that 3K and 2K slomo is cool, I wish it were more and full-frame)") and got/gets everybody excited. Mx upgrade also solved many very real problems with the old M sensor, as every Red one D.P. had a time when they wished they had more useable high ISOs with nicer-looking and less noise.

Dragon seems to mainly aim to solve an imaginary problem of lack of resolution that no one is really having, not even if you use an Epic for high-end stills work.

WIth MX and Epic all in the community were clamoring with excitement and anticipation. Dragon? Let's face it, not really.

I say this with the best of intentions. It's not too late for Dragon to be "subject to change".
Click to expand...


If you don't like it and think it's not what you need don't buy it. Simple as that...why complain about it....if you want to stay at 4k or 5k, stay there. The beauty of RED..
 
Rob, raising a discussion about what it is camera owners want out of the sensor must be one of the most useful things to do on this forum.
To be sure, I'll be upgrading ASAP, for the added DR and low light capabilities.
Thanks, Red, for the opportunity.
Rob, I agree with the priorities of what you'd like to see in a sensor.
I think they reflect the concerns of the vast majority of owners and production needs.

More than one DIT has landed in the "more pixels is great" side of this thread.
They'll be earning more overtime money, backing up and transcoding,

I don't need an ounce more pixel resolution, if the lens resolution is already maxed. Telling yourself that you'll now be able to blow up a shot 50% in post, but ignoring the fact the you are quartering the optical performance of the lens is stupid.
I don't need an ounce more pixel resolution if it comes at any expense of DR or low light performance.
After all, pixel resolution is the MX's strong point. It's the sensor's other performance levels that the designers should be striving for, to make a significant leap ahead of the other cameras already on the market for years. Surely.
To the "well you can still shoot 4k or 5k" argument:
I don't need smaller pixels if it means I'm digging into the optical resolution of the lens to get the same pixel resolution as an MX.
And smaller pixels does mean that. If you want to maintain the same lens resolution with smaller pixels, you'll have no choice but to shoot 6k.
This isn't complaining or FUD. It's just common sense.
 
Hey Graeme,

If the article is that terribly incorrect, then I apologize for posting it... But I'd love to hear a 'bit' more about what exactly is incorrect about it.

Graeme Nattress said:
That article is, of course utter bollocks. It's not just outdated, it's wrong, factually incorrect and written in an absurdly biassed manner ignoring even contemporaneous evidence to the contrary.

Graeme
Click to expand...
 
Maybe I don't fully understand...but the sensor, to me, seems like a great upgrade. 2 stops more native dynamic range is awesome, and Jim mentioned the improvements being at both the high and low end - so better lowlight and better highlight retention. A lot of lenses I use vignette/porthole at 5K on the Epic right now, so 5K on the Dragon fitting in a Super-35mm space makes them more usable. 6K allows for higher resolution when I need it (plates, stills, etc), and allows for more look-around for reframing/stabilization/tracking/etc (if I use lenses that cover it) if I shoot 6K with a 5K frame guide (right now we do a lot of 5K with a 4K frame guide). And the added height (both physical and pixel) will allow better use of anamorphics for me.

For everyone wishing for a larger sensor - what lenses would you use with it? What cine glass, besides the CP.2s and other converted stills glass, would fully cover something larger than the MX's 30mm-wide sensor? I ask because I honestly don't know.

EDIT:
Daniel Soderberg said:
Hey Graeme,

If the article is that terribly incorrect, then I apologize for posting it... But I'd love to hear a 'bit' more about what exactly is incorrect about it.
Click to expand...
Not Graeme, but I can point out one obvious one. For bit-depth, he's taking Epic's bit-depth for one channel vs. the others' for three. 30 bits is 10 bits per channel, and 42 bits is 14 per channel. The Epic is 16-bit, so four times the information of the 14-bit and 64 that of the 10-bit.
 
Daniel Soderberg said:
Hey Graeme,

If the article is that terribly incorrect, then I apologize for posting it... But I'd love to hear a 'bit' more about what exactly is incorrect about it.
Click to expand...

You'll have to search back through posts on reduser for that, as there's so much wrong, so much based on inadequate understanding and false premises.

Graeme
 
Kevin Marshall said:
For everyone wishing for a larger sensor - what lenses would you use with it? What cine glass, besides the CP.2s and other converted stills glass, would fully cover something larger than the MX's 30mm-wide sensor? I ask because I honestly don't know.
Click to expand...

You just mentioned them yourself: Anamorphics.
 
This discussion of pixel dimensions versus sensitivity is happening in the still camera forums too. I mean, if pixels were all that mattered, did everyone junk their Canons for the Nikon D800? (Tom -- did you get a Nikon D800?) On the flip side, as a cinematographer, do I really need 6400+ ASA except in very rare situations? Or do I need 8K? Sort of depends on the project but I think if you are going to go for the broadest shooting circumstances, then obviously you want to pick a sweet spot between pixel density and sensitivity, and find a way to increase DR and reduce noise. Trouble is that everyone is going to pick a different sweet spot.
 
Mark Pugh said:
You just mentioned them yourself: Anamorphics.
Click to expand...
Aha. I've never used anamorphics (have always wanted to, though), so I guess I was unfortunately a little ignorant on that front. What about on the spherical front, though?

Mark Pugh said:
They still won't cover the sensor, will they?
Click to expand...
They'll at least cover the 5K portion of the Dragon, which was said to be roughly the same area as 4K on the MX.
 
Frazier Nivens said:
If you don't like it and think it's not what you need don't buy it. Simple as that...why complain about it....if you want to stay at 4k or 5k, stay there. The beauty of RED..
Click to expand...

I'm not complaining I'm trying to give - and illicit from others - constructive feedback on what is important me and to the Red owning community in general from a new sensor who's specs and features are perhaps not yet 100% cemented. I am a bit surprised that was not totally obvious to everyone.
 
Rob Ruffo said:
I'm not complaining I'm trying to give - and illicit from others - constructive feedback on what is important me and to the Red owning community in general from a new sensor who's specs and features are perhaps not yet 100% cemented. I am a bit surprised that was not totally obvious to everyone.
Click to expand...

It's the thread title, and some of the phrasing. It came off a little on the "whining" side. Otherwise I think this thread has been great about voicing real world concerns about the (insert number here)K arms race that is about to happen in the camera world.

Although this isn't the Red of old. The info they release now is probably at least 4-6 months behind where RED actually is in development. It may be too late.

RED can loose focus about what's important in day to day shooting (*cough* power management *cough*) because they're off pushing the limits of future cinema.
 
Status
Not open for further replies.
Back
Top