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 :)

rodney charters mx and alexa

Status
Not open for further replies.
I just took a couple of framegrabs from the video - the same area of the scene. One is from MX lookaround area, the other is from Alexa's main image area.
I can hardly see the difference between the two, in terms of highlight detail.
Arguably, the Alexa actually shows more clipping in the left of the picture, but that may be an illusion because of the different angles of the two cameras.
I do think that I can see a greater difference in the top and second top two panes of glass on the right, in the MX than in the Alexa. (!)

The Alexa lookaround area shows no more detail than it's recorded area, that I can see.
The MX show a crapload more info in the lookaround. And that least that amount of detail is being recorded to the R3D. (I don't know if the lookaround here repesents Redcolor monitoring, or RAW. If it's Redcolor, then there could well be even more highlight detail being recorded.

Why R.C. would say he's seeing "staggeringly more detail out the window" with Alexa, without
A): pointing out what's clearly there in the MX lookaround
B): flicking the Video on the MX to RAW to check it out, rather than waiting to have a look "in post"....
..I don't get.

It would be great if R.C. could upload the R3D.
 

Attachments

  • MX lookaround.jpg
    MX lookaround.jpg
    4.9 KB · Views: 0
  • Alexa.jpg
    Alexa.jpg
    7.3 KB · Views: 0
Video "explanations"

Video "explanations"

I certainly didn't know this, but Rodney states in another video that Alexa actually records in a semi-HDR mode by taking two frames for every single frame of recorded media. One reflects the settings you chose in the system, then one that automatically adjusts and captures into the highlights, then combining them before the final image is recorded- hence the added DR he is saying the Alexa has.
Click to expand...

Well the only problem with that statement is that Arri's US rep has already publicly stated that it does not do that.

i.e. the comments in that video about taking two frames to make one are factually wrong..

http://www.reduser.net/forum/showpost.php?p=575490&postcount=513
 
MX capability

MX capability

(I don't know if the lookaround here repesents Redcolor monitoring, or RAW.

If it's Redcolor, then there could well be even more highlight detail being recorded.
Click to expand...

The LOOK AROUND area is a 50% opacity black rectangle mixed over the color corrected and gamma corrected output of the sensor. So normally that means it's REDcolor, but it always tracks the selection of the camera viewing mode - RAW / REDcolor / User. i.e. it would never be REDcolor inside the rectangle and RAW within it.

What the stills suggest to me is simply that the RED ONE HD-SDI output has been set up inappropriately.
 
Last edited:
Stuart English said:
The LOOK AROUND is 50% opacity black mixed onto the color corrected and gamma corrected output of the sensor. (i.e. REDcolor)
Click to expand...

So chances are there was even more detail in the MX highlights (in the R3D) than can be seen in that lookaround area, which already matches what the Alexa can offer, at least to my eye.
 
While I have to take the actual appearance of the monitor with a gigantic grain of salt (it is a monitor as recorded by what seems to be a cellphone and then uploaded for the web and compressed once more...) it does seem interesting that there is significantly more detail in the look-around area. Clearly, the data is there. I'd love to learn more about the settings. Given the response here, however...I'm not sure that will happen.
 
The images on the ARRI site are interesting:

3_SensitivityCompare2.jpg


What is really interesting about this image is the exposure. If You notice the ARRI is at 800 ASA with 360˚ shutter, while the "competitor A" is at 450 ASA with 180˚ shutter. So not only the ARRI is using double exposure, but they are also at almost double the sensitivity. Yet the images appear identical when it comes to overall ambiance.

So either the "competitor A" was intentionally underexposed and then "brightened" in post, or the "competitor A" has a astonishingly more sensitive sensor... Hmmmmmmm....

2_SensitivityCompareBlowUp.img_assist_custom-784x240.jpg


If this is indeed RED ONE, it seams to be the "normal" Mysterium sensor (not M-X) and camera build with older color science. The latest images from M-X sensor do not exhibit this blue noise. Apples to Oranges...
 
Michael Olsen said:
Clearly, the data is there. I'd love to learn more about the settings.
Click to expand...

Just show me the R3D.
Maybe Red should consider having some sort of LOOK flag on the HD-SDI display when in non-menu mode, to illustrate what's going on... and if there's a custom LUT enabled, even have a small graphic representation of the curve on screen.
It could be an asset on set, certainly it would be a reminder that what's being seen may be a massive variation of what's being recorded in the R3D, and in the case of this Vimeo posting, would show it up for what it is.
 
Peter Majtan said:
...

So either the "competitor A" was intentionally underexposed and then "brightened" in post, or the "competitor A" has a astonishingly more sensitive sensor... Hmmmmmmm....
Click to expand...

Good points, Peter.
 
Mark Pugh said:
It's Apples to Apples, if you consider that the off-the shelf Red-one is not MX.
It's why I think that Red should stop selling non-MX Red Ones.
Click to expand...

Alexa is still "BETA" with many features not enabled/activated. RED ONE with non-MX is 3-year old model. The "current" (read 2010) model is MX sensor. Sure You can buy the 3-year old model for 17500, but You can also buy the current MX model for 1/2 the price of the ProRes Alexa, or 1/4 the price of RAW Alexa. In reality they should really be comparing Epic MX with Alexa, as both are BETA pending final release...

Apples to Oranges...
 
Pawel Achtel said:
Good points, Peter.
Click to expand...

360/180 degree shutter difference is mindboglingly, intentionally deceptive. Definitely Apples for Oranges.
The 800/450 ASA difference shouldn't matter if it's being processed RAW, though...

can a Red One even be set to 450ASA? I'm sure it's 400/500/640/800...

Once again, though, Red should stop selling Red One non-MXed. It would mean a test like this stadium one, is between Alexa and a no-longer available model.

MP
 
I used to be a huge fan of ARRI. I still adore some of their products. But when it comes to digital cinema cameras I honestly feel disappointed. Not because Alexa is bad. Not at all. Alexa is a fantastic product and true achievement from ARRI. It is the way their PR is trying so hard to compare it to RED at the cost of spreading FUD (and borderline misleading information). This is what hurts me the most. If they purely focused on what Alexa is and its features (without trying to trash the "competition" with FUD) and the key purpose, people will see it for what it is. I can see Alexa being preferred camera for TV series and even broadcast and other HDTV content. When it comes to theatrical features - sure You can shoot it on Alexa, but ARRI has posted themselves study (by their own R&D department) highlighting why 4K+ is the future. You can be sure there will be 4K+ camera coming from ARRI as soon as they can make it work (maybe even next year). I have always had nothing but respect for ARRI and I know most of the key people in Munich by first name, having met them personally. But the current PR surrounding Alexa is destroying their reputation and diluting my trust in the truthful and honest info they always used to provide...

Just my sincere opinion...

Peter
 
Peter Majtan said:
While ISO is pure meta-data and can be changed in post - it does affect the way You expose Your images.
Click to expand...

It makes no difference to the discussion - both images where shot at 1.3.
Having "Competitor A" set to a lower ISO would only make you give it more light, anyway...
 
Mark, but do You agree that the image was intentionally underexposed and then blowed up to exaggerate the noise level? That is relevant to the discussion... And so it the fact that they are comparing 3-year old technology with their latest state-of-the-art one...

Anyway, I have made my point and I leave it there. I am only so passionate about this because I have truly "loved" and admired ARRI. I'm so disappointed with their PR strategy...

Howgh...

image3_1235756918.jpg
 
Mark Pugh said:
It makes no difference to the discussion - both images where shot at 1.3.
Having "Competitor A" set to a lower ISO would only make you give it more light, anyway...
Click to expand...

Yeah, both were shot at 1.3 but one at 360 degrees and one at 180. One camera intentionally under exposed.

I agree with Peter, I can't understand why a company like Arri is stooping to this kind of crap and I'm extremely disappointed. For all Jim and Red's famed hype, their attitude to their competitors is, generally, respectful. Blunt perhaps but respectful and, as far as I've been able to tell, always truthful.
 
Wow. I was at SIM the day Rodney was doing this test and met him for the first time.

I have to say I am truly disappointed at the "method" of testing.

There is no method!?

You can judge nothing from this test.

Really disappointing.

Although I am excited by Alexa - this "test" provides no useful information other than - "what camera holds more DR in a video tap".

Meaningless.
 
Back to the Rodney Charters' Vimeo posting:
This offers ammunition against anyone wanting to shoot on a Red to a format with a baked-in look, instead of R3Ds.
If a DP of R.C's caliber, or his ACs, can make a mistake about look-ups on camera (I'm making a presumption there), then heaven help the average user.
To create images with a proxy module that look like this, then have the editors misuse those proxys... it's scary.
 
Peter Majtan said:
Mark, but do You agree that the image was intentionally underexposed and then blowed up to exaggerate the noise level?
Click to expand...

Exposure was twice as long on the Alexa at 360 degrees, for no decent reason, and only admitted in small font.
Shame on them. Like selling snake oil.
 
Status
Not open for further replies.
Back
Top