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 :)

Dragon - Initial Reaction After Taken Delivery

H

Hans von Sonntag

Well-known member
Joined
Jan 15, 2007
Messages
1,968
Reaction score
0
Points
0
Website
www.hubbertvonsonntag.com
Hi,

I've been on the Red ship since 2007 and took all the upgrade paths from RedOne to Dragon. Red-Team, thanks for this opportunity, second to none in our business!

Today my dragonized Epic came back home after a 14 day journey to California (pretty quick!). I was excited to realise that not only the sensor was replaced but also the cooling system. The camera is now much, much, much - said I much? - quiter than before. Very welcome!

Now to the sensor:

It has a smaller FOV when comparing equal resolutions of Dragon and MX Sensor. On MX I shot most of my stuff on 4KHD. 4KHD was a tad smaller than Super-35 which has been my favourite format before Red came in my life. On Dragon 4KHD is even more smaller than on the MX, so I checked 5KHD. 5KHD seems finally the new S-35 format and works well with my PL-glass, no vignetting so far. 6K does vignette considerably depending on the focal length.

Dragon exhibits a similar amount of noise compared to MX. Telling from experience with the MX sensor I would rate Dragon the same, something around 400 ASA. But with more DR. Shooting 800 ASA seems to be a fair bet in most cases. For green screen in the studio 400 ASA would be more desirable, keeping noise completely out of sight.

DR is better than MX, I would say at least 1½ stop, perhaps 2 stops, surely Alexa territory.

Colour ist the best part of Dragon. Here, Red got a lot right. The biggest complain I always had with digital is the missing colour richness in skin-tones. The M-sensor, especially in the pre-16 builds was unusable for commercial work when cosmetics were the subject. Later on, the colour science got better but none of the RedColor iterations could convince me and I sticked to CameraRGB / RedLogFilm.

Dragon has it's own colour science. It is well more saturated than CameraRGB but - and this is the vital part - it carries significantly more colour information, especially in the darker parts. It might show a slight green tint but this is not my problem, film always had some sort of cast. But to my delight with Dragon and DragonColour skin-tones exhibit delicate, subtle colour nuances and keep them deep into in the shadows.

Thanks again for this journey! It was bumpy and took over 6 years but finally I arrived. With Dragon I won't miss the film days anymore.

Hans
 
Thanks for that info...its funny, because I put my Optimo 24-290 on when my Dragon arrived and noticed zero vignette at 5kFF...and I swear the Optimo used to vignette on the non-dragon...I thought my short term memory was in play...Im doing more tests.
 
The 24-290 does indeed vignette (at least) heavily at 5k FF MX.

I am not sure how it plays on Dragon, but I would guess 5k WS at least should be ok.
 
So, I know this has been beaten to death, but the comparable noise btwn Dragon and MX bothers me. Don't get me wrong, I'm SO excited for the better color and dynamic range. The higher resolution is cool, and I'm happy to have it, but not needed for my work. Higher light sensitivity, however, had me really excited... and we WERE lied to. Jannard said, in no vague terms, that this camera would be as clean at ISO 2000 as the MX is at ISO 800.
That being said, I can't wait to get a dragon, but I hope the bugs are worked out. I just upgraded from Scarlet to Epic and my Epic stopped working correctly almost immediately (it's currently back at Red for evaluation).
Anyway, guess I'll have to wait for the Dragon-X to get better sensitivity
 
Greg M said:
Thanks for that info...its funny, because I put my Optimo 24-290 on when my Dragon arrived and noticed zero vignette at 5kFF...and I swear the Optimo used to vignette on the non-dragon...I thought my short term memory was in play...Im doing more tests.
Click to expand...


Greg, I'd be very interested in any 12x tests on the dragon. Thanks.
 
This is a simple test...lens at f22/Infinity

just put the Super Speeds up and at 5K they all cover fine. At 6kHD still good. At 6K the 18mm vignettes in extreme corners.

24-290 Optimo
Epic MX Sensor- 5k vignettes thru 50mm
Epic Dragon Sensor- 5K is clean at all fl. 6kHD vignettes from 24mm-55mm. 6k vignettes from 24mm- 80mm.
 
Nathaniel Kramer said:
We are still rating at 320
Click to expand...

In a controlled environment not a bad idea if maximal clean pictures are desired.

For DR and exposure test I usually cross the street and visits the park in my neighbourhood. On a sunny day shadows and highlights can easily excel 16 anticipated Dragon stops. I found in thiose conditions ISO 800 better than 400 or 320 because the additional stop helps in the highlights.

Required is a log gamma.

Hans
 
Greg M said:
This is a simple test...lens at f22/Infinity

just put the Super Speeds up and at 5K they all cover fine. At 6kHD still good. At 6K the 18mm vignettes in extreme corners.

24-290 Optimo
Epic MX Sensor- 5k vignettes thru 50mm
Epic Dragon Sensor- 5K is clean at all fl. 6kHD vignettes from 24mm-55mm. 6k vignettes from 24mm- 80mm.
Click to expand...

Greg, to be true I only tried the 16-42 Optimo. It works as expected with 5KHD but shows slight corner vigentting. Haven't tried my super speeds yet, but I'm sure your findings are spot on.

Hans
 
Stephen Franchek said:
So, I know this has been beaten to death, but the comparable noise btwn Dragon and MX bothers me. Don't get me wrong, I'm SO excited for the better color and dynamic range. The higher resolution is cool, and I'm happy to have it, but not needed for my work. Higher light sensitivity, however, had me really excited... and we WERE lied to. Jannard said, in no vague terms, that this camera would be as clean at ISO 2000 as the MX is at ISO 800.
That being said, I can't wait to get a dragon, but I hope the bugs are worked out. I just upgraded from Scarlet to Epic and my Epic stopped working correctly almost immediately (it's currently back at Red for evaluation).
Anyway, guess I'll have to wait for the Dragon-X to get better sensitivity
Click to expand...

Stephen,

If you don`t mind me asking, what is the problem with your EPIC? I ask because I have also traded in my Scarlet for an EPIC and the very first time I hit record, it showed a warning message about an audio buffer error and then froze. It happened almost every time i tried to record...
 
I think the confusion with the noise between MX and Dragon comes from the fact the most people do not account for the different sensors where Dragon has much smaller photosites. So yes - you MIGHT be seeing the same noise level if you look at 5K frame from both sensors, but you need to realize that the Dragon ones were taken with much smaller photosites. So that on its own is a major achievement. In order to properly compare the noise level you need to shoot the same image - and for the same FoV you need to shoot higher resolution on the Dragon. Then take the two images (they should look the same, just the Dragon will have a higher resolution) - then scale them to 4K and only then look at the noise level and DR - since both improve when scaling down. Unless you are shooting stills - nobody is delivering in 5K or 6K. So look at your 4K master and compare that...

:sifone: Peter
 
I looked at my Super Speeds at 6K HD, all covered fine except the 14mm Zeiss Standard Speed had just the slightest vignette (almost imperceptible). Also, the 15-40 Optimo coved fine as well.
 
Peter Majtan said:
scale them to 4K and only then look at the noise level and DR - since both improve when scaling down. Unless you are shooting stills - nobody is delivering in 5K or 6K. So look at your 4K master and compare that...
Click to expand...

and when you need to denoise, the algorithm has much more spatial information to work on!

So the slight amount of sharpness lost is not noticeable when scaled down 6K -> 4K Master.
 
steve green said:
I looked at my Super Speeds at 6K HD, all covered fine except the 14mm Zeiss Standard Speed had just the slightest vignette (almost imperceptible). Also, the 15-40 Optimo coved fine as well.
Click to expand...

If you test wide open it's hard to see what is happening...look at your lenses stopped down and focused at infinity
 
Greg M said:
Thanks for that info...its funny, because I put my Optimo 24-290 on when my Dragon arrived and noticed zero vignette at 5kFF...and I swear the Optimo used to vignette on the non-dragon...I thought my short term memory was in play...Im doing more tests.
Click to expand...

Photosite size on the Dragon is 5.0um vs. 5.4um on the M/ MX sensors. So 5K on Dragon is 25.60mm wide and it was 27.65mm wide on MX. In a nutshell, 5K on Dragon now fits within the common S35 image circle. And at 6K the sensor is still a bit larger than MX. Other than the cropped height, we're in Canon APS-H territory for sensor sizing.
 
You must log in or register to reply here.
Back
Top