Welcome to our community

Be a part of something great, join today!

Register 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 :)

Scarlet Dragon native iso

S

Shermen L

Well-known member
Joined
Oct 26, 2011
Messages
51
Reaction score
0
Points
0
Since I can't find in spec, what's the Scarlet Dragon native iso supposed to be ? Is it 2000 ?

Thanks
 
I'd say the Dragon sensor has a working range rather than a fixed base ISO.

If you treat dragon like a film stock you can safely rate the camera between 250-1280iso.

i have rated and tested up to 3200 ISO but I feel that there is too much texture/grain/noise in those images for me.

Rating 1280 and push processing to 2000 can yield some favorable results compared to rating at 2000. But everyone should test for themselves with metered light.

The factory preset is 800 ISO which will give you the and even number of steps above and below middle grey when you are exposing. It's best to test and come up with where the sweet spot is for you with your own tolerance for grain and exposure. I and others (especially Phil holland through his extensive testing) have found that 250-1280 is a really solid working range of ISO's for dragon.

Battistella
 
ISO-wise, SCARLET DRAGON and EPIC DRAGON are identical. When SCARLET DRAGON can match the resolution, frame rate, and compression of EPIC DRAGON (i.e., 5K 24fps REDCODE 8:1) it is identical to EPIC DRAGON.

That said, 5K using REDCODE 8:1 is not going to deliver the same image quality as 6K using RECODE 5:1 (no surprise!) but you may well find that rating the SCARLET DRAGON at ISO 250-320 may well catch what EPIC DRAGON can do at ISO 1000-1280 despite its higher resolution and lower compression ratio. But that's also not a hard-and-fast rule. It's only a rule for the many, many test cases people have submitted to REDUSER (including myself) where the first thing people say about an image shot at ISO 800 is "that image is under-exposed by a good 2 stops!". In that case, a properly exposed SCARLET image (at ISO 250-320) will catch up to a more highly tweaked EPIC image (exposed at ISO 1000-1280).

Bottom line: when everything else is equal, SCARLET DRAGON == EPIC DRAGON. EPIC DRAGON can deal with more inequalities than SCARLET DRAGON (because of larger resolutions and/or lower compression ratios).
 
Julian Holzwarth said:
where can i find info about the compression rates of the scarlet dragon?
on the german red website it says 48fps 5k (FF) at a rate of 5:1, but that doesn't sound realistic to me.
Click to expand...

The fun starts on page 78 (of 80) of this document if you are running the latest software.

At 5K / 48fps, you are looking at REDCODE 16:1.
 
Thanks, just wonder if I record Red Scarlet for 25fps (5:1) compression, how long can I record the video in a 64GB SSD ?
 
Shermen L said:
Thanks, just wonder if I record Red Scarlet for 25fps (5:1) compression, how long can I record the video in a 64GB SSD ?
Click to expand...

If you're talking Scarlet Dragon at 5K then your looking at under 10 minutes of record time...

With Scarlet MX at 4K then around 20 minutes.
 
Shermen L said:
Thanks, just wonder if I record Red Scarlet for 25fps (5:1) compression, how long can I record the video in a 64GB SSD ?
Click to expand...

That's a trick question. If you record at 25fps, you must use 9:1 compression at 5K (per the previous document I referenced). You can calculate recording times by entering the correct resolution, compression, and frame rates into this tool. Unfortunately SCARLET DRAGON is not one of the options you can select, so you have to select EPIC DRAGON and enter parameters from there. The tool will tell you 19 minutes.
 
Thank you for info, indeed, I've been a red one user and use redcode 36 mainly, so how's the quality affected for a 5K 9:1 compression on dragon sensor compared with a 5K 5:1 compression on a Epic MX sensor for example ? Or I should not judge by the numbers ?
 
Shermen L said:
Thank you for info, indeed, I've been a red one user and use redcode 36 mainly, so how's the quality affected for a 5K 9:1 compression on dragon sensor compared with a 5K 5:1 compression on a Epic MX sensor for example ? Or I should not judge by the numbers ?
Click to expand...

You have to see it for yourself. I think that the colors of the DRAGON more than compensate for the compression of SCARLET, but that's a tradeoff, not an absolute.
 
Shermen L said:
Thank you for info, indeed, I've been a red one user and use redcode 36 mainly, so how's the quality affected for a 5K 9:1 compression on dragon sensor compared with a 5K 5:1 compression on a Epic MX sensor for example ? Or I should not judge by the numbers ?
Click to expand...

I would be interested in the answer too.
 
Christopher-lee dos Santo said:
From what I have heard, dragon compression is much better than MX compression. Apparently Dragon 16:1 looks like MX 10:1. So I'd imagine Dragon 9:1 would be comparable to MX 5:1? Or am I just being hopeful?
Click to expand...

Unfortunately it doesn't seem to be the case. In his tests (http://nofilmschool.com/2014/07/red-epic-dragon-vs-mx-sensor-raw-camera-shootout) Ryan Walters explains that the compression is the same on both mx and dragon. And here, http://www.reduser.net/forum/showthread.php?122407-Scarlet-Dragon-or-Epic-Dragon, the compression example is really awful.
 
sam karr said:
Unfortunately it doesn't seem to be the case. In his tests (http://nofilmschool.com/2014/07/red-epic-dragon-vs-mx-sensor-raw-camera-shootout) Ryan Walters explains that the compression is the same on both mx and dragon. And here, http://www.reduser.net/forum/showthread.php?122407-Scarlet-Dragon-or-Epic-Dragon, the compression example is really awful.
Click to expand...

If one were to read just that on this thread and not actually read what was said in the other thread, one might have a very wrong impression. The "awful" compression example is at 16:1. I've never shot MX, so I don't know whether MX at 10:1 was wonderful or awful, but general focus of the questions has not been about 16:1, but about 9:1. The example in the above link shows 8:1 compression, which has far smaller artifacts than the terrible purple fringing of the tester's lens. Which means that compression at 8:1 (and likely 9:1) is not a deal-breaker. Compression at 16:1 may be a deal-breaker, especially in highly detailed foliage. It may also not be such a deal-breaker, depending on whether the macro-blocking is visible in the actual motion footage and/or whether or not one needs to really push the image around in post.

I shoot SCARLET DRAGON at 8:1 compression, and I have yet to see any adverse effects of compression in my images (which tend to have small bits of high detail and larger bits of less detail). I am lucky to live in a part of the world where 24fps (and hence 8:1 compression) is acceptable. In PAL-land, compression bumps up to 9:1, and in NTSC-land it bumps up to 10:1 (for 30fps).

Only one's own testing can answer for sure where the line is that things work or fall apart. Just looking at the numbers, if I lived in PAL-land I'd probably shoot 5K 2:1 or 5K-UHD, both of which work at 8:1. If I had to shoot at 30 fps, I'd see whether 5K WS at 8:1 was better or worse for me than the 10:1 compression of other 5K formats. Unfortunately, SCARLET DRAGON doesn't give half-K resolution increments, so if you don't like 5K, you can't just go down to 4.5K, you have to drop all the way to 4K. But at 4K, 30fps is a happy 7:1 compression ratio.

The really good news is that many of these questions can be perfectly answered by renting a EPIC DRAGON for a day. An EPIC DRAGON can be dialed in to SCARLET image and compression settings, and then can be compared to the less compressed reference of EPIC.
 
You must log in or register to reply here.
Back
Top