ARRI ALEXA LF - 4k, FullFrame announced

[Björn Benckert]

You are mistaken.



Yes that was in the infancy of digital.
And today Alev3 holds most highlights and has the largest buckets by a strange coincidence.



Light sensitivity is not the only aspect of DR.



That is not a DR advantage.

What do you mean by Alev3 holds the most highlights? Do you mean it´s the least light sensitive sensor on the planet? If so I can beat it with my pocket cam, I just leave the lens cap on and it holds the highlights better it actually then holds the highlights until the lens cap melts of from the heat of those highlights.

You can not only look at the highlights of a camera if you not also, at the same time look at the other end of the spectrum. If you show me a picture where this fantatstic alexa Alev3 sensor holds the highlights and also sees further into the dark then monstro. Then yes, you are right. But not before then.


And don't give me this bull about so and so many stops above rated iso. Iso is just a rating and it´s just a way of distributing the DR. you have to set both sensors at their clipping point or noise floor and then compare who got the longest stick in the other end. Comparing them from some sort of fabricated midpoint that is not validated by anyone is not the way to do it.

And sure its more difficult to make small pixels that are better than larger pixels. But things have moved forward. There are today smaller pixels that are more light sensitive and house a wider range than the pixels of alexa, helium is prof of that and also monstro.

 

[Hrvoje Simic]

What do you mean by Alev3 holds the most highlights?

That it has the best highlight range.
Sorry for the clumsy formulation.

You can not only look at the highlights of a camera if you not also, at the same time look at the other end of the spectrum.

I am not looking only into the highlights, you can check my previous posts.

Your rhetoric is based around shadow range DR perception which ignores the highlight lattittude, hence the comment.

And don't give me this bull about so and so many stops above rated iso.
Iso is just a rating and it´s just a way of distributing the DR. you have to set both sensors at their clipping point of or noise floor and then compare who got the longest stick in the other end. Comparing them from some sort of fabricated midpoint that is not validated by anyone is not the way to do it.

.....okay....perhaps someone here has the will to explain why this matters, I currently don't.
I'm still tired after "it doesn't matter if you shoot on S16 or IMAX, it's just DoF " type of fun.

 

[Brian Boyer]

If a LARGE pixel on a sensor could 'count' 1000 photons each frame before clipping, was compared to another sensor where 4 SMALL pixels each with the same 1000 photo counter (and where the smaller pixels were 1/4 the area of the large pixels):

4 x SMALL pixels would be able to count 0-4000 photons per frame in the same area that the 1 x LARGE pixel could count 0-1000 pixels.

However as wonderful as it is to have 4 x the Dynamic Range (in this example) - there is huge drawback : You had to deal with 4 x as many pixels; 4 x as much accumulation, 4 x the bandwidth, and a lot more HEAT that you would have to dispense with, in addition to noise and power.

My speculation is that Arri would have reduced the size of their pixels (which might even IMPROVE their dynamic range capabilities), but have not yet resolved the Wattage requirements in the pixels processing.

AJ

My quote...

This is what I've been trying to stress, albeit a bit clumsily at times. Arri is NOT ready to shrink pixels because they know how important highlight latitude is to the look and success of the Alexa. They're not going to give that up.

was based on this...

3) Spatial resolution oversampling and minimizing noise visibility that way affects the shadow part of DR. Not the highlight part. And as you decrease the pixel size full DR can be similar due to noise size reduction but the DR distribution cannot and mid point moves to the right, trading highlight range for shadow range.

If you disagree with this assertion, feel free to take it up with Hrvoje, as Björn has already done. You all can hash that out. I'm not a sensor designer so I'm not in a position to say yay or nay.

All I know is, whatever the case, what you say here...

Edit : More specifically : If Arri could create a 6k Sensor with dimensions as their 4K LF sensor for a UHD output, and in the same power budget -> I believe they would.

ain't gonna happen unless they can match or best the Alev III's current highlight latitude capability. They're not going to give up one of their two main advantages and join the ranks of every other sensor just so they can achieve true UHD resolutions. The last 8 years of a 2.8/3.2K sensor is proof of that.

 

[Björn Benckert]

That it has the best highlight range.
Sorry for the clumsy formulation.





I am not looking only into the highlights, you can check my previous posts.

Your rhetoric is based around shadow range DR perception which ignores the highlight lattittude, hence the comment.



.....okay....perhaps someone here has the will to explain why this matters, I currently don't.
I'm still tired after "it doesn't matter if you shoot on S16 or IMAX, it's just DoF " type of fun.

Highlight range on a cmos sensor... to me thats defined how close to the center of your color cube you choose to expose your highlights. To me its not really possible to say that alexa has the best of such a thing. You need to compare the sizes of both cubes and then you need to expose them equally, which means if the two sensors have differences in light sensitivity (they do) then you need to take that into account before you start to compare.

S16, IMAX and other sizes of formats matter as long as you use the same medium. A given film stock can be seen as set medium. If you increase the area of it, then you get better resolution and also a bigger DR as your noise floor becomes less prominent / The extra resolution then gives you additional range. But the worst of film stocks shot in IMAX does for sure not need to look better or have greater range then the best of film stocks shot in S16. So again its all depending what kind of medium you use.

Same for digital. Size of the sensor can easily be compensated by a better "medium" such as a more dense array or photo sites that has the same or greater DR. Such exists and its an ongoing progression.

 

[Aaron Lochert]

I'm still tired after "it doesn't matter if you shoot on S16 or IMAX, it's just DoF " type of fun.

If that's what you took from the other thread, it's no wonder you're tired. No one was making the argument that DoF is the only differing factor between all the formats.

 

[David J. Buchanan]

I'm pretty sure the cons and pros to a lot these cameras is the programming/processing. How do we really know either of which is better than the other? It's probably just easier for the software engineers to figure out the curve in a more organic way through programming on an Alexa than a RED-- probably. That's some of the reasons why phone cameras look better than other phone cameras with litterally the same specs.

I know there are other specifics, but if you look at a real digital chart through software on a computer, you can see REDs highlight clip and don't fall off at all (not tested with IPP2 yet) and the Alexa doesn't... It has jagged lines, yes, but it still curves until the very end of the spectrum.

 

[Björn Benckert]

I'm pretty sure the cons and pros to a lot these cameras is the programming/processing. How do we really know either of which is better than the other? It's probably just easier for the software engineers to figure out the curve in a more organic way through programming on an Alexa than a RED-- probably. That's some of the reasons why phone cameras look better than other phone cameras with litterally the same specs.

I know there are other specifics, but if you look at a real digital chart through software on a computer, you can see REDs highlight clip and don't fall off at all (not tested with IPP2 yet) and the Alexa doesn't... It has jagged lines, yes, but it still curves until the very end of the spectrum.

Thats wrong. Then you examen this stuff with curves applied in the process. As both goes very much linear to clip. The rest is to taste. If you use Arri seasoning on arri meat you might love it. If you use red seasoning on red meat you might find that it taste horrible. But you have to get rid of the seasoning to really know who got the best meat.

Understood?


Dont take the dynamic rounding and the rest of the post processing into account when comparing raw pictures.

 

[Aris_Gavriilidis]

We can all sit here and talk - or to some degree make guesses - about curves, pixels, science behind light gathering, internal processing, but without any images from these cameras to compare, from skintones to overexposure to the "boring" charts, all the pages on this thread basically amount to nothing, other than stating the obvious fact that "ARRI is more popular than RED". Personally, as an indie filmmaker, I stopped caring about all this stuff the day I received my Epic Dragon, which is a camera that can create FANTASTIC images and can do everything I want, resolution's there and will be enough for years to come, DR is there, I've NEVER had an issue with latitude, there's enough color data I can manipulate in post to my heart's desire, the list goes on. Is the Alexa LF spec-wise better? I definitely think so. Do I care? No, not really, I can get EXACTLY the image I want from my Epic, albeit after some work in the grade, but I don't mind at all, it's a camera with idiosyncrasies and I've learned to love it. Only reason I thought I should post on this thread is because I don't believe the introduction of this camera will just make RED completely irrelevant.

 

[Steve Sherrick]

Arri needs RED and RED needs Arri. And cinematographers need choices. It really is a great win-win for everyone. The only surprise for me is that this is not the sensor I thought it was going to be. So, perhaps I’ve had some incorrect information or Arri is still working on a different sensor that is still not ready for prime time. They needed this camera at this point of time so that DPs were not being locked out of the choice to shoot Arri for Netflix and Amazon. I think it’s a smart interim solution and buys them some time for the other stuff they are working on.

I love seeing these companies push each other. It really is great for the industry. And I respect each artist’s choice of format. And of course, sometimes it’s not an exclusive choice as many films shoot with multiple cameras/formats. But having these various palettes to work with is incredible. Even the low end is very exciting and opens up the opportunities for so many talented people.

Enjoy the ride, we are in a glorious time of innovation.

 

[Patrick Tresch]

So, perhaps I’ve had some incorrect information or Arri is still working on a different sensor that is still not ready for prime time.

You are probably right. The FPS that this camera can do shows that there are some more horsepower under the hood that could be used for a 6k+ FF sensor @ some interesting frame rates.
Strange thing they don't communicate about this speculated sensor upgrade. It surely could be a good way of future proofing their camera. (copying RED again ;-) )

Pat

 

[Björn Benckert]

Arri needs RED and RED needs Arri. And cinematographers need choices. It really is a great win-win for everyone. The only surprise for me is that this is not the sensor I thought it was going to be. So, perhaps I’ve had some incorrect information or Arri is still working on a different sensor that is still not ready for prime time. They needed this camera at this point of time so that DPs were not being locked out of the choice to shoot Arri for Netflix and Amazon. I think it’s a smart interim solution and buys them some time for the other stuff they are working on.

I love seeing these companies push each other. It really is great for the industry. And I respect each artist’s choice of format. And of course, sometimes it’s not an exclusive choice as many films shoot with multiple cameras/formats. But having these various palettes to work with is incredible. Even the low end is very exciting and opens up the opportunities for so many talented people.

Enjoy the ride, we are in a glorious time of innovation.

Yes,
This fits with what I heard as well. I think this is the last squeeze out of the ALEV sensor line. Then they will have something more dense. I think no one here is saying its a bad camera that comming. But I´m a bit surprised by everyone stating ALEV got more DR and than monstro. Monstro has more DR than Helium... and I compared helium with arri quite a bit. But sure everyone can have their own opinion about usable DR and what they like and don't like. And I dont really care what others use or think, even though I find it a bit odd that people toss statements around technical stuff that I know for a fact is not true.

 

[Hrvoje Simic]

Highlight range on a cmos sensor... to me thats defined how close to the center of your color cube you choose to expose your highlights. To me its not really possible to say that alexa has the best of such a thing.

Well to you it's not possible because you insist on your own perception of DR and testing and utilising it so you don't have a reference and as a result all is hanging in the air.

You need to compare the sizes of both cubes and then you need to expose them equally, which means if the two sensors have differences in light sensitivity (they do) then you need to take that into account before you start to compare.

You are still focused only on sensitivity and ignoring sensor highlight capture capacity.
Yes you expose them equally. And yes I compared "sizes of both cubes" to use your terminology. Which is why in one of my previous posts you can see mentioned saturation to signal intensity. Guess which camera rules in that department.

Same for digital. Size of the sensor can easily be compensated by a better "medium" such as a more dense array or photo sites that has the same or greater DR. Such exists and its an ongoing progression.

"More dense arrary" does not necessarily entail every image aspect superiority. As clearly manifested in Alexa.

Size of the sensor cannot be compensated in spatial geometry, defined with corresponding glass.

 

[Hrvoje Simic]

Thats wrong. Then you examen this stuff with curves applied in the process. As both goes very much linear to clip. The rest is to taste.

No, that's wrong.

They differ in sensor response, how the signal is electronically processed and how it is mapped into base color science.
And they both can achieve fantastic imagery.

Bjorn, I am aware you are an expert in your field and really like your work. Please assume there is more stuff existing outside your expertise and current focus.

 

[Hrvoje Simic]

If that's what you took from the other thread, it's no wonder you're tired. No one was making the argument that DoF is the only differing factor between all the formats.

That was a benign simplistic joke.
I'm not the dismissive one in that thread.

 

[Brian Boyer]

...I stopped caring about all this stuff the day I received my Epic Dragon, which is a camera that can create FANTASTIC images and can do everything I want, resolution's there and will be enough for years to come, DR is there, I've NEVER had an issue with latitude, there's enough color data I can manipulate in post to my heart's desire, the list goes on. Is the Alexa LF spec-wise better? I definitely think so. Do I care? No, not really, I can get EXACTLY the image I want from my Epic, albeit after some work in the grade, but I don't mind at all, it's a camera with idiosyncrasies and I've learned to love it.

Let me start out by saying everyone is entitled to their opinions and free to express them in the manner in which you did. Your voice deserves to be heard and what you have to say is as valid as anyone else's comments. Also, I think RED appreciates hearing from users like you.

Having said that, yours is the type of post that I hope RED ultimately chooses to ignore in terms of future development, despite loving what you have to say. RED may not always like what I have to say but I hope they at least think I'm being reasonable, honest and objective when I say it.

Your quote is basically interchangeable with posts from every era of RED cameras. All you have to do is substitute Dragon with R1M, R1M-X and Epic M-X. There's always someone who talks as if it's the last camera they'll ever want or need.

It's fitting the Olympics are going on right now because I view these types of posts like judges score performance. They throw out the highest and lowest score and what's left is averaged.

The highest of high praise is flattering but not all that helpful. The most critical of critiques, where everything is bad, is also useless when it comes to improving a product. The stuff that drives a company and their products forward lies somewhere in the middle.

Only reason I thought I should post on this thread is because I don't believe the introduction of this camera will just make RED completely irrelevant.

I wouldn't say "completely" because 1) RED is popular among many segments for many reasons and 2) the LF has some special requirements.

I do think the LF can conceivably make RED less relevant in yet another area where they had a decided advantage over Alexa. Arri has been slowly and very deliberately chipping away at RED's strong points.

In the early days. Recode and RAW were a lot less easy to deal with. Arri went with ProRes, which most TV shows have standardized on for their workflows.

RED had an advantage in size for the burgeoning gimbal and drone markets. Arri released the Alexa Mini.

RED had 6K. Arri released the 6K Alexa 65. It's large format and debayers to a true 4K output.

RED had the advantage as far as streaming services like Netflix and Amazon requiring 4K capture. Arri introduced the LF, which also simultaneously addresses the large format appeal Monstro and Venice have.

They did all of this with the same sensor. Is worth noting and examining.

 

[Sabovic Adis]

Since no one wants a cheap system, RED should only up the prices in order to beat ARRI in every segment.

 

[Phil Holland]

Since no one wants a cheap system, RED should only up the prices in order to beat ARRI in every segment.

There is another player in the room of "larger format". Sony Venice with their FF35 6K sensor. Which is interestingly priced in relationship to both RED and ARRI. Also, at least for me as a filmmaker, interesting to see the various nuances in feature sets and image quality.

There will be at least one more player coming shortly. Two in a slightly longer timeframe.

 

[David J. Buchanan]

Since no one wants a cheap system, RED should only up the prices in order to beat ARRI in every segment.

Sarcasm?

Did anyone see the post about RED prosumer cameras... Geezus.

 

[Eric Lange]

There is another player in the room of "larger format". Sony Venice with their FF35 6K sensor. Which is interestingly priced in relationship to both RED and ARRI. Also, at least for me as a filmmaker, interesting to see the various nuances in feature sets and image quality.

There will be at least one more player coming shortly. Two in a slightly longer timeframe.

I'm glad you mentioned the Sony Venice (I think the price point is of the order $45K) , so I have been "Sitting out" on most of this discussion as I feel that the Sony Venice would be amore relevant comparison to this "New" Arri Alexa LF ("Large Format 4K ALEVIII Sensor (from the ALEXA family)" " thing " , rather than direct comparisons with VV based Monstro systems?

 

[Eric Lange]

Sarcasm?

Did anyone see the post about RED prosumer cameras... Geezus.

+1

And I think RED also has new configurations that leap frog over other systems by virtue of having a business model that goes hand in glove with their increased modularity and "Obsolescence Obsolete" model.

_____________________________________________________________________________________________________________________________________________________________________________________________

OT/BTW @ David J. Buchanan I really like your avatar... What is that from or about?