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 :)

Desired Changes with Scarlet..

karapetkov said:
Does it look as good as real ramping, I've never tried that.
Click to expand...

For excellent results you need no motion blur in the original footage and you can't push it too far... however Twixtor which is a plug in for AE does a marvelous job... if you pushed clean 120 footage to 180 you would be hard pressed to notice it is cranked in post. The key is to give it as much info as possible... so the 120fps is a god-send.
 
I don't know if the old "shoot the highest frame rate you need then ramp in post" is exactly as good as in-camera ramping. Maybe one of the pros here can discuss that.

I guess exposure time might play into it. But I would think for most ramping you could just shoot at the highest rate you want/need for the slo-mo part of the ramp, and then speeding up the other parts (ramping up) would be fairly straightforward in AE or PP CS4, etc. I've played around with 60p overcranked footage in AE using its ramping, and it works great.
 
2k sensor for 1080p delivery

2k sensor for 1080p delivery

I do recognize that the 2/3" scarlet has blown the tiny-sensor-competition out of the water. I mean, who want's to buy an EX1 or a HVX now? Or an F23 for that matter...
It is just that: I want to get rid of the fuzzy image and light losses of my 35mm adapter. And 1080p target resolution is enough for a whole bunch of years to come. The movie theaters here have projectors that are only 1080p, and it looks totally ok. I even once produced some ad for screening on 35mm film - produced it as 1280x720 and it looked OK on the 25m screen.
I realize that red has skew handled, but why can Nikon make a camera with S35 sized chip for $1k? I would gladly pay $5k for a scarlet that made 2k at 50fps from an S35 sensor. Even the Arri D20 is 2k, and no one is deriding it for being just 2k, or at least not very loudly. RED is about modularity. That modularity is of limited usefulness if resolutions are maxed out even now? Give me a measly 2k now and I'll upgrade later when there's an actual demand for 4k, 6k, 8k. I mean that would have been a great way to differentiate the scarlet from the RED ONE, scarlet: 2k, R1: 4k.

5k on the S35 will make my quad core MacPro seem really old. And $7k will require me to have a solid business plan. But maybe that's a good thing...
You know, all these folks shooting with DoF adapters are now totally used to FF35 DoF and going back to 2/3" is really hard.
Now I don't think that changes in the sensor program are likely to happen still, and that's what would be needed to accommodate my desired changes. And I don't even know if making the S35 sensor only 2k would make its cost decrease significantly.

Well, so much for 1st post. Sorry if this is coming across somewhat negative.
It's not that I am not totally hyped about the redvolution, because I am. Because now, if I figure out a way to make a buck from making films, I know that there is a camera that will totally deliver for a reasonable amount of money.
 
Will the 2/3" Scarlet shoot 1080P as the REDONE is going to do?

It should.

I think this will make it real competition to the new Varicams.
 
karapetkov said:
Tom, do you mean the Time Remapping thing in AE?

Actually it's in PPro too.
Click to expand...

Hehe, do you already have CS4? :) :whistling:

Yeah. That feature works great. You just have to shoot high FPS to acquire an entire sequence, and then you can essentially speed up parts of it - for that ramping effect. Again, I seem to remember that there might be some advantages to ramping in camera, but I'm not an expert on this. I know that when you're shooting film, you obviously don't want to just overcrank everything, because it's expensive. But there might be other reasons, like shutter time, that would cause people to want to ramp in camera.

silly-pictures, keep in mind that when Scarlet was announced, it was 2/3 3K with a fixed lens, period. Many of us here begged Jim to make a 35mm sensor, and we said we were willing to pay more for it. So he has answered our prayers. I'm sorry that he did not make one EXACTLY like you need for the price you can pay, but he is offering a hell of a lot of options.

35mm 2K makes no sense at all, because you will essentially be left with an image that is about 720 quality - which is of little use to most of us. Yes, Genesis and other pro cameras are 2K or 1080p, but most of them hugely oversample the image to achieve that 1080 or 2K. A Bayer-pattern 2K camera will not result in a clean 1080p or 2K.
 
Tom said:
I don't know if the old "shoot the highest frame rate you need then ramp in post" is exactly as good as in-camera ramping. Maybe one of the pros here can discuss that.

I guess exposure time might play into it. But I would think for most ramping you could just shoot at the highest rate you want/need for the slo-mo part of the ramp, and then speeding up the other parts (ramping up) would be fairly straightforward in AE or PP CS4, etc. I've played around with 60p overcranked footage in AE using its ramping, and it works great.
Click to expand...

The whole idea is that the more frames you have the more information Twixtor will have closest to the point in time for which a frame needs to be created. The higher the frame rate the smaller the spaces get in between the frames that you captured. I am not so sure how Twixtor works, but someone told me that two of the things it does is to look at each pixel in the frame before and after closest to the point in time for which it is creating a frame for... and it also looks at several frames before and after to analyze the motion in the frame... it is some real intricate algorithm stuff... so what you want is clean pixels, meaning without motion blur.. real pixels... and to get as much info as possible for Twixtor to be able to refer to.


karapetkov said:
No, I was asking about ramping, not overcranking.
Click to expand...

Twixtor allows you to make up your ramps in post.. you tell it what you want to have and give it what you captured, and it creates the frames. Overcranking is only used to give as much info to Twixtor as possible for it to have the highest probability of creating accurate frames.


I was told that with clean 60fps you can get decent 120fps, so perhaps 80fps out of 60fps would be clean enough for most people.




So as an example you could shoot something at 120fps, then get it into Twixtor and turn it into 24fps at first, then create a ramp that lasts X seconds and goes from 24fps to 200fps, then hold the 200fps for X seconds and then create an other ramp that lasts X seconds and goes back down from 200fps to 24fps... all from the 120fps captured footage.
 
Tom said:
Hehe, do you already have CS4? :) :whistling:

Yeah. That feature works great. You just have to shoot high FPS to acquire an entire sequence, and then you can essentially speed up parts of it - for that ramping effect. Again, I seem to remember that there might be some advantages to ramping in camera, but I'm not an expert on this. I know that when you're shooting film, you obviously don't want to just overcrank everything, because it's expensive. But there might be other reasons, like shutter time, that would cause people to want to ramp in camera
Click to expand...

Exactly... that is what I am saying.


Ramping in-camera is excellent because it is real frames captured at the exact points in time needed for the ramp... so it is sharp as could be. The only "problem" would be that you need to get it perfect on location because tweaking it will be nearly impossible or at least a royal pain.. especially because you would not have a dense timeline filled with frames... you would only have 24fps making the gaps as big as they could be so re-mapping would be done with little information.

However raping in camera is kind cause it is real. Ohhh... the beauty of a perfect ramp... just... magic. :love:
 
USlatin, I think we are kind of mis-communicating. karapetkov was not asking about turning 120fps into 200fps.... that's a separate issue. he was talking about straight-forward ramping, which does require the creation of interpolated frames to create a slo-mo effect -- the slo-mo part was already captured by the camera. all you need is a ramping program to speed up parts of the 120fps video, for example, to make them appear to run closer to real-time speed.

Edit: okay, i just saw your last post and it seems like we are on the same page now. :) Where are you from, USlatin?
 
Yea, it is all the same... I was just trying to explain it using an example that also referred to the limits of the feature which would be on the top end as undercranking is always easier for these programs.

Am I not making any sense at all that you ask me where I am from? lol... I am not feeling that great right now.. tired and hungry, plus I've been breathing in smoke from the fire in the north end of the valley, about 10 miles from here...lol

I was born in Argentina, Buenos Aires, moved to Boca Raton 12+ years ago and here about 3 years ago. So I am half and half, basically. :)
 
USlatin said:
Am I not making any sense at all that you ask me where I am from? lol... I am not feeling that great right now.. tired and hungry, plus I've been breathing in smoke from the fire in the north end of the valley, about 10 miles from here...lol
Click to expand...

LOL, sorry. We have so many people from around the world that I sometimes just assume it's a language thing... hehe. No worries.
 
Why 2/3"?

Why 2/3"?

I don't get the choice of a 2/3 sensor at all. A 3k sensor in a 35mm form factor is technologically EASIER to achieve than trying to increase pixel density. Furthermore since the cell size would be larger it would also yield a much better s/n ratio and better dynamic range. All the other electronics and a non fixed lens mount as an option would make a great entry level pro camera.
 
not true

not true

Bigger sensors at the same pixel density might cost more money.
At lower pixel densities they don't. They cost consumers more money because they are in more expensive cameras. And they don't cost that much more! i.e. a $900 camera vs a $400 camera. Would you pay $500 more for a 35mm size sensor?
Bear in mind that the smaller sensor is going to cost you more because you'll need a better lens!
 
Would it be possible for Jim to make a 3k s35 camera for around $3k? I understand that a bigger physical size of sensor might be more expensive to manufacture, but making it at lower resolution might also reduce the cost.

I see a huge advantage to be able to upgrade the body and use the same lenses. I agree that the 2/3" sensor has it's place, but it would be hugely useful if you could start out with a 3k resolution, which is plenty for todays standard requirements, and have the option to upgrade the body without getting a whole new lens set.
I see this as a flaw in the lineup that was announced. It was stated that you would be able to upgrade by just upgrading the "brains" and not the whole system. This is only possible for the higher end cameras leaving out the pro-sumer level and leaving a big gap between $2,500 and $7000. Sure I know that Red will come out with higher resolutions later and probably offer upgrades for the 2/3" as well as the *35 versions but those who need upgrades or additions before this happens would be up a creek.

For example if someone starting out got a 3k 2/3" camera than added a second S35 camera as primary and the 3k camera became secondary than the lenses would not be usable between both cameras. A whole new set of lenses would need to be purchased.

I the announcement last Thursday totally blew my mind and I am very grateful that Jim made all the changes that he did. However I noticed that he didn't remove the features and specs are subject to change, in fact count on it statement. I take that to mean that he still might be willing to make some changes.
I would like to see options from the ground up on both lens options. I can see someone paying up to $3500 for a 3k S35. I don't see any reason that making a bigger sensor at the same resolution would cost more than $1000 more than a 2/3" sensor.
I would see this option as really making competition RED in the face. That's what you want right Jim?

Any comments Jim?
 
I think it is just to be able to offer 2/3" sensors... some people prefer them because it makes DoF deeper and still allows for decent shallow DoF when needed... the fact that it is 3K is probably because they don't want to go lower than that so that all their cameras offer true debayered 2k output. just my guess.
 
2/3"

2/3"

3K on 2/3" is a sweet spot for controllable DOF and excellent image quality across a very large number of applications.
 
Stuart English said:
3K on 2/3" is a sweet spot for controllable DOF and excellent image quality across a very large number of applications.
Click to expand...


That is why 2/3" is good. This still doesn't address the lens issue. When one goes with this size though they will probably end up with thousands of $$$ in glass that won't be useful with an s35 or FF35 upgrade when they want to add a more powerful camera. All the mini reds will limit the res on a **35 when used together.

This means that when upgrading from 2/3" model to a **35 model a whole new set of lenses will need to be purchased. This is not the "upgraded simply by purchasing a new brain" talked about in the release document. Starting with a 3k 35mm option will be a lot less expensive than having to push out a higher res 2/3" model soon after the release. For me these seem to be the only two options to make that "upgraded simply by purchasing a new brain" marketing phrase come true.

You guys have an awesome operation going here!! I hope you don't let your marketing ploys not match your delivery of product.
 
You must log in or register to reply here.
Back
Top