Red Scarlet vs F3

My budget wouldn't look like others. That's the first thing. Your 600 day cam package will be different than my own.

But even then, I would still go out and get a decent DSLR setup if I cOuld not get better on a 4-7K shooting budget for camera only. The disservice to a client is Not turning out a balanced product.

Two different setups, of course if I had a prod co I would be amortizing my costs over rentals etc. You are right.
 
TimurCivan said:
The epic/scarlet mx chip is more advanced than the r1mx chip.
Click to expand...

Just wanted to jump in here - the folks at RED who know the nitty gritty details have said multiple times that the M-X chips in RED ONE M-X cameras and the M-X chips going into Epic and Scarlet are exactly the same. A Mysterium-X S35 sensor is a Mysterium-X S35 sensor. There's only one version. The processing components and pathways behind the sensor, however, are different and contribute to the better image coming out of Epic (and presumably Scarlet).
 
Exactly. No difference in the sensor at all.

And Duke about me using RED on a nano budget feature: I used two, actually. I'll tell you that I spent less to do so than a Gh2 14-140 kit. The average micro filmmaker cannot do that.

They can do a DSLR.
 
Kholi said:
And Duke about me using RED on a nano budget feature: I used two, actually. I'll tell you that I spent less to do so than a Gh2 14-140 kit. The average micro filmmaker cannot do that.
They can do a DSLR.
Click to expand...
Now that deserves a tutorial!
 
Last edited:
Really?

I was under the impression the mx in the epic was a different chip. My bad yo.
 
Michael Olsen said:
Just wanted to jump in here - the folks at RED who know the nitty gritty details have said multiple times that the M-X chips in RED ONE M-X cameras and the M-X chips going into Epic and Scarlet are exactly the same. A Mysterium-X S35 sensor is a Mysterium-X S35 sensor. There's only one version. The processing components and pathways behind the sensor, however, are different and contribute to the better image coming out of Epic (and presumably Scarlet).
Click to expand...

if it's identical, than how is the Epic's chip physically bigger than the r1mx's?
 
TimurCivan said:
Really?

I was under the impression the mx in the epic was a different chip. My bad yo.
Click to expand...
I thought it was the same no worries. The size of the R1 is a bit of a deterrent for me. Scarlet means I can keep my head and sticks etc. I don't know mane...i have to wait til I see the other players on the same field.
 
tom.wong said:
if it's identical, than how is the Epic's chip physically bigger than the r1mx's?
Click to expand...

An excellent question! First, let's understand that every Mysterium-X chip has approximately 5,000 horizontal lines and 2,700 vertical lines, earning it the name '5K'. Apparently, the RED ONE M-X can only process a maximum of 4.5K resolution of the total 5K chip due to their older hardware. This physical "windowing" means that the maximum effective sensor size is smaller on RED ONE M-X than EPIC, which has the muscle to process the entire 5K image. This is also what happens with the Scarlet, which shoots at 4K (or 3K, or 2K, or 1K). The effective size of the sensor decreases in comparison to 5K on the EPIC, but the physical sensor itself is the same. It's just not recording every pixel as a part of the frame.

This image is from REDUSER.

RedScarletXCropFactors.jpg


One can see that when shooting 5K, the full sensor size is used. Shooting 4K, however, is not shooting 5K then scaling that to 4K. It's actually shooting a 4K section of the sensor - which is smaller than the 5K section. And so on.

I'm guessing that this is one of the cost-cutting mechanisms that allows the Scarlet to exist at it's price-point. RED doesn't have to spend all the money on R&D for three different chips. And it doesn't have to pay for space, labor, equipment, and logistics for three different assembly lines. And it doesn't have to engineer the processing and colorimetry of three different chips. Instead, it just makes one very capable chip and then puts it in front of different hardware which allows it to perform differently at different price points. Really, it's rather smart.

Of course, this is a double-edged sword. Resolution choice will impact resolution (of course), but also the size of the sensor - meaning that the field of view will be altered which could require re-lensing or repositioning to get the shot.
 
Last edited:
Dont forget that the Olpf is probably optimized for 5K as well. Maybe RED made an Olpf specifically for scarlet? Their MX Olpf? But does it fit?
 
starcentral said:
Excellent post. Now try saying that on RED forums and see what happens.

:D
Click to expand...

I'm not sure why anything special should - the benefits of having the same sensor in three cameras is apparent there...as are the issues of windowing instead of scaling and of different fields of view at different resolutions. Those have been around for years, since the RED ONE.

--

Getting away from that topic, I have no idea what they may have done with the OLPFs between RED ONE, RED ONE M-X, EPIC, and now Scarlet.
 
Kholi said:
Dont forget that the Olpf is probably optimized for 5K as well. Maybe RED made an Olpf specifically for scarlet? Their MX Olpf? But does it fit?
Click to expand...
The OLPF would work for any resolution, because they're always sampling at pixel-to-pixel ratios. What works for 5K would also work for 2K, specifically because they're not trying to use the whole sensor size.
 
Barry_Green said:
The OLPF would work for any resolution, because they're always sampling at pixel-to-pixel ratios. What works for 5K would also work for 2K, specifically because they're not trying to use the whole sensor size.
Click to expand...

Okay that makes more sense. I always thought the drop in IQ that goes along with the drop in resolution (5k to 4k or 4k to 3K) had to do with the OLPF, but may simply just be the cause of shooting with glass not designed for the application? Or is it the sensor crop itself?

There's a pretty discernible difference between 5K and 4K, and of course 4K to 3K, then 3K to 2K, and doesn't feel like it's just a resolution drop. Feel being the key and questionable word.
 
Michael Olsen said:
An excellent question! First, let's understand that every Mysterium-X chip has approximately 5,000 horizontal lines and 2,700 vertical lines, earning it the name '5K'. Apparently, the RED ONE M-X can only process a maximum of 4.5K resolution of the total 5K chip due to their older hardware. This physical "windowing" means that the maximum effective sensor size is smaller on RED ONE M-X than EPIC, which has the muscle to process the entire 5K image. This is also what happens with the Scarlet, which shoots at 4K (or 3K, or 2K, or 1K). The effective size of the sensor decreases in comparison to 5K on the EPIC, but the physical sensor itself is the same. It's just not recording every pixel as a part of the frame.

This image is from REDUSER.

RedScarletXCropFactors.jpg


One can see that when shooting 5K, the full sensor size is used. Shooting 4K, however, is not shooting 5K then scaling that to 4K. It's actually shooting a 4K section of the sensor - which is smaller than the 5K section. And so on.

I'm guessing that this is one of the cost-cutting mechanisms that allows the Scarlet to exist at it's price-point. RED doesn't have to spend all the money on R&D for three different chips. And it doesn't have to pay for space, labor, equipment, and logistics for three different assembly lines. And it doesn't have to engineer the processing and colorimetry of three different chips. Instead, it just makes one very capable chip and then puts it in front of different hardware which allows it to perform differently at different price points. Really, it's rather smart.

Of course, this is a double-edged sword. Resolution choice will impact resolution (of course), but also the size of the sensor - meaning that the field of view will be altered which could require re-lensing or repositioning to get the shot.
Click to expand...



well i guess the better question is, why when doing sensor size comparisons, the Epic is said to have a bigger chip, and R1 MX is always said to have a smaller one? when they are both identical in size, and we are just using less of the chip on the R1 MX and thus have reduced field of view. These are conflicting sources of information.
 
I know one thing, the difference in 5K and 4K is astounding all around. I love shooting on a 50mm now, it's become my favorite when using Epic @ 5K.
 
Duke,
Indeed tell the story FAR exceeds the sensor and resolution. However, client expectations and their subscribing to marketing hype isn't accounted for in this. I'll get more jobs when my Scarlet comes just to be able to say, "I shoot with a Red." Show up, shoot with the F3 if need dictates and they'll not even care.
 
The sensor is the same, but only the EPIC can make use of all the photocells.

The OLPF must fit the size of the photocells, not the window. All REDs are different from a Canon 5D or 7D in this respect, since the RED is always using every single photocell, while the Canons are doing this only in photographic mode. Their OLPF is correct for photography, but not for video, hence the massive aliasing.

The lens plays an important role. If you use the full sensor and scale the image down to HDTV or 2K, any decent lens should resolve to the max. If you use a window in HDTV size, the lens must be two times as good. That's why lenses fully resolving HDTV on the much smaller 2/3" sensors are so expensive.
 
Duke M. said:
I'm good with the F3 as it is for the next 3 years. Half the households in the US don't even have HD TV and a majority of the digital projectors are still 2K. There will be a time for 5k finishing at 4k but for the average indie the support systems (file storage, editing, CC, etc) are a lot of hassle when most things are going to 2k anyway.

I still think the most important factor is that resolution doesn't tell a story. Movies like Avatar, Benjamin Button, Star Wars and Slum Dog Millionaire won many awards and were shot at 1080p. Che', despite a lot of hype and attention because a Red was used, was disappointing no matter how sharp the images were.

The fact is, you can cut together F3 footage and Red footage quite well. The audience doesn't notice.
Click to expand...



I recently asked a buddy who works at one of the best digital effects houses in town what he thought about working in 4K. To which, he said, "why would we be working with 4K, that would be a pain." So it would seem that RED is getting well ahead of the curve and are only really benefiting the memory makers and hard drive makers. How many people shooting on Epic and Scarlet are delivering in 4K? Didn't think so.

Given how long it took SMPTE and Congress to agree on a new HD standard (in fact it took so long, that it was really almost obsolete when it came out) I don't think we will see new higher resolution TV standards anytime soon (remember, the Gov needs to find airwaves to broadcast a 4K signal, to keep basic TV "free"). In fact, I think the studios would like to see the theater experience be the only place you can go for the 4K experience. Or the 4K-3D-Smell-O-Vision, experience. They will almost certainly drag their feet when it comes to a new standard for television. As is, the pirates are having a field day sharing 1080p and 720p multi-Gigabyte (with a Big G) versions of films already. Is there any wonder the film industry is fighting back with Pipa and Sopa and the Open Act.

End of rant...
 
You must log in or register to reply here.
Back
Top