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 :)

MX vs Alexa

Gunleik Groven said:
That's not the reason... -:)

But if it's practical, I'll post some arriRAW too...
Click to expand...

Alexa ARRI RAW to download, such great news!

My GlueTools ARRI RAW plug-in for FCP is waiting for it!!!!!
 
Hi guys...
I was offline for a few days, enjoying the Labor Day holiday in the States by installing a kitchen backsplash. It was harder than I thought... lots of cuts and many, many 1"x4" tumbled marble pieces. Came out pretty good though. A nice way to recharge the batteries.
Sanjin.
I think I'm calmed down now. You probably sensed the excitement in my post. Yes, screening film and Alexa side by side and seeing very little difference was pretty exciting.
I was also excited when I went to see footage from my first Red filmed out to IMAX.
I was also excited when I fired up my first Red after the MX upgrade and saw the amazing difference.
I imagine I'll be excited when I go out with my first EPIC.
Turns out I'm a camera geek and easily excited by new imagemaking tools or techniques. And we live in exciting times.
However, maybe I'm too format-agnostic for this forum. I guess I'll just calm down some more.
OK, Sanjin?
Cheers,
Harry
 
Sanjin Jukic said:
Rob Ruffo said:
5D commercials on Vimeo>>>

5D commercials on YouTube>>>

7D commercials on Vimeo>>>

7D commercials on YouTube>>>
Click to expand...

They are still low-end tools, for mostly low-end jobs, and their popularity is waning because they are really no fun to work with in a professional context. I was all excited when 5D came out and eager to shoot with it, now I flatly refuse to do so. Too much stress over the long list of things that can go wrong, ugly-ish low-res low dynamic range clippy image, no benefits at all other than small size. Besides, most of those "commercials' you list are not real, they are demo-spots, done without the real stress of a real commercial shoot running at $5000 an hour. Those that are real are local spots, the kind that used to shoot on actual Betacams.
Click to expand...
 
If you compare production stills (shot on 5D by Adam J. Wilt and he doesn't allow to post them here) and put them side by side on your desktop with Alexa shots,

you can see and also probably conclude yourself that 5D image looks nicer, more natural, better colors, etc,... then Alexa images at the night shots on the same location.

Also Alexa can go ISO 1600 at the maximum.

But MX can go ISO 6400 max and with a good de-noise combo treatment in post you can get easily a very nice footage out of it.

At the end let us have a look what Gunleik will bring us here with his comparison test between MX & Alexa

and with promised RAW files to download we can see hopefully all what we would like to know about "The Good, the Bad and the Ugly" from those two digital cine camera systems.
 
I realy wanna see the noise comparison at 1600 ASA between RED1 MX 4K RAW and ARRI ALEXA 2.8K RAW. I am pretty sure that ALEXA won't have the same results like it has in 1080p quicktime pro res files.
 
Philip Orlandic said:
I realy wanna see noise comparison at 1600 ASA between RED1 MX 4K RAW and ARRI ALEXA 2.8K RAW. I am pretty sure that ALEXA won't have the same results like it has in 1080p quicktime pro res files.
Click to expand...

Keep in mind that MX @ ISO 800 doesn't equal Alexa @ EI 800.
 
arriRAW?

arriRAW?

Gunleik Groven said:
That's not the reason... -:)

But if it's practical, I'll post some arriRAW too...
Click to expand...

I was told by the Arri rep that arriRAW in Alexa is not enabled yet, has that changed?

Are the arriRAW files having uncompressed true RAW Bayer data in them?

If they do I could extract the Bayer data and ignore the header and process them in the de-Bayer program I am working on, that was why I was asking Arri about sample files, I wanted to add a menu option for Alexa ArriRAW files to my program since I have options for Acam dII, Kinor-2K, SI-2K etc. now. I talked to Drama cam about samples and he did not seem to have any yet for download, he is working on is own de-Bayer for his camera that will output DPX etc.

If you get your Alexa enabled and can post some sample arriRAW frames for people to look at that would be a help to confirm various workflows.

About the noise issues, Arri seems to be using the exact 1/1.5x ratio of pixel sizes since it gives fewer de-Bayer artifacts than odd down-size ratios, and can be done in hardware quickly. Since 5/9 of the pixels in the 1920x1080 image are interpolated you get about a 6db advantage in S/N ratio over using a 1:1 pixel size de-bayer for the interpolated pixels, so that oversample gives Alexa somewhat better dark shooting (maybe 3db total?). That is, there are TWO interpolations, one converts 9 pixels area to 4 pixels area, and the other is the normal de-Bayer interpolation. When you resize a RED image you also average the data, so there should also be some gain from that going from 4k or 5k to 1920x1080?

Because Alexa more or less needs the 1/1.5x resample ratio, the 3392x2200 maximum image area does not compute to a standard image size, the closest seems to be 2048x1152,

2048x1.5=3072
1152x1.5=1728

For Epic to use the same 1/1.5x trick, the sensor would need to be 6k to get 4k results, not 5k, so with 5k you are using a ratio of 1/1.25x to get 4k, the de-Bayer artifacts do not null with the 1/1.25x ratio like they do with the 1/1.5x ratio, but that may not matter since the way REDCODE is processed now does not take advantage of oversample ratios in the RED ONE...

Is the arriRAW Bayer data stored 16bit, or 12bit in their file format? (that can be found maybe from the file sizes)

2880x1620x2 = 9.4MB frame for 16bit data unencoded
2880x1620x1.5 = 7MB frame for 12bit data unencoded

If delta encoding was used the files would be from 50% to 100% that size, with about 80% being typical, maybe.

At 7MB*24*60 = 10GB per minute for uncompressed data. With a 1TB HD going for less than $90 now, you get 100 minutes for less than maybe 4 minutes cost of ECN, so how is Kodak going to compete?

Does anyone know if the "mathematically lossless" recording mode in Epic will allow for the output of Bayer DNG files (not Linear DNG which are like a non white balance TIF already being interpolated so that the de-Bayer cannot be used)?

Bayer type DNG based workflows would the be open to Epic users, similar to converting ArriRAW to Bayer DNG maybe, so you could mix True RAW data in the same workflow from both cameras then.
 
As I said, my timeframe is a week or two, if we publish in two batches, or 3-4 weeks if we publish all at once...

I don't think I am saying anything wrong by saying I have seen RAW recording units on Alexa. The unit and workflow will be tested before we shoot day two, sometime after IBC.

But the first day I recorded only prores to the Alexa.

As to the proceccing, debayering, RAW signalchain et al, I am afraid I am not the one to ask. Though - as a slight comment to your CML post - I cannot really see evidence that there is applied heavy sharpening in the Alexa. So far.

I shot a small pickup today, to try to document what the Alexa ISO/EI does. It's pretty sweet and unlike anything I have ever seen, but then again: I haven't seen them all and they're all changing daily now, as they're all officially computers with a sensor on.

The fun part with the initial shooting, is that it generated more (to me) unexpected questions than expected answers, questions I am now following up on, thanks to the ones helping me.
 
Gunleik Groven said:
I shot a small pickup today, to try to document what the Alexa ISO/EI does. It's pretty sweet and unlike anything I have ever seen,...
Click to expand...

Gunleik,

would you be so kind to specify what does it mean???!!!
 
Dan Hudgins said:
Does anyone know if the "mathematically lossless" recording mode in Epic will allow for the output of Bayer DNG files (not Linear DNG which are like a non white balance TIF already being interpolated so that the de-Bayer cannot be used)?
Click to expand...

I'm also interested in this (and would like the answer to be "yes").
 
I haven't read all the thread but i remember that Jim said there is an upcoming firmware update that improves RED MX's image quality/dynamic range further. Will you wait for it before doing the test?
 
When I shoot is not really only up to me, but I have offered to re-shoot anything if there are issues that might command that.

I am definitely eager to slap on any alfa/beta/gamma firmware for the purpose of a test, if it is made available to me... -:)
 
Sanjin Jukic said:
If you compare production stills (shot on 5D by Adam J. Wilt and he doesn't allow to post them here) and put them side by side on your desktop with Alexa shots,

you can see and also probably conclude yourself that 5D image looks nicer, more natural, better colors, etc,... then Alexa images at the night shots on the same location.
Click to expand...


You can't be serious.
 
Sanjin Jukic said:
If you compare production stills (shot on 5D by Adam J. Wilt and he doesn't allow to post them here) and put them side by side on your desktop with Alexa shots,

you can see and also probably conclude yourself that 5D image looks nicer, more natural, better colors, etc,... then Alexa images at the night shots on the same location.
Click to expand...

So... you want people to compare production stills from a 5D with ProRes from an ALEXA.

I just compared production stills I pulled off the web from David Fincher's "Seven" (shot with a Panasonic Lumix) with 35mm stills from Lawrence of Arabia. From that, I have determined that the 5D still looks better. :laugh:

Sanjin - you *really* don't like the ALEXA. Yet you have never shot with, never done post with it, and don't have any actual real-world experience with it. That, combined with ridiculous comments like this (really) make it hard to take your input seriously on this particular subject!

And you do have serious input, so... go out and do some of your own real-world tests and I bet you'd have some great results to post that would be very interesting for the community here!

With a smile,

Lucas

Lucas Wilson
------------
ASSIMILATE, inc.
LA, CA, USA
 
You must log in or register to reply here.
Back
Top