Welcome to our community

Be a part of something great, join today!

  • 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 :)

The future of low light performance in prosumer cameras?

Don't forget that a full functioning Scarlet will be considerably more expensive. The 5D even with the issues decribed above, its still a steal. The only Scarlet I'm interested in is the S35, which will probably run about 15k for a bare bones package.
 
Why compare it with the 2/3" though? The 1D Mark IV goes for $5k on B&H, which seems more comparable with the S35, not the 2/3" (especially considering the huge differences in other features).

When the IV can do > 100fps without line skipping you may be entitled to a comparison.

Graeme
 
Don't forget that a full functioning Scarlet will be considerably more expensive.

Depends...like I said there's a pretty huge difference in other features. It's only a steal if you have a pretty narrow set of tool needs. As that expands the value of the Canons plummet.

Edit: Are you including a lens in that estimate? You'd need to add that onto the 1D's price too.

Graeme Nattress said:
When the IV can do > 100fps without line skipping you may be entitled to a comparison.

Indeed. :)
 
Why compare it with the 2/3" though? The 1D Mark IV goes for $5k on B&H, which seems more comparable with the S35, not the 2/3" (especially considering the huge differences in other features).

Why compare it with any Red product as a motion camera when it's not designed for that? All of the current Canon products are still cameras. They are designed that way, and they are marketed that way. Even if you go directly to Canon's site, these cameras are, first and foremost, digital still cameras. The video capabilities are not even mentioned on the first page. Red currently makes a motion camera that can produce acceptable performance for limited still use, but the Canons are far and away superior for high end still use. The next generation of Red products will provide better performance for still use. When Canon produces a motion camera that has good still performance, there might be a basis for direct product comparison. But for now, there isn't one because those using the Canon products solely for motion video are not going to get optimal quality because that's not what they were built to do.
 
Why compare it with any Red product as a motion camera when it's not designed for that? All of the current Canon products are still cameras. They are designed that way, and they are marketed that way. Even if you go directly to Canon's site, these cameras are, first and foremost, digital still cameras. The video capabilities are not even mentioned on the first page. Red currently makes a motion camera that can produce acceptable performance for limited still use. The next generation of Red products will provide better performance for still use. When Canon produces a motion camera that has good still performance, there might be a basis for direct product comparison. But for now, there isn't one because those using the Canon products solely for motion video are not going to get optimal quality because that's not what they were built to do.


I think any argument most poeple have is what to do with their time while they wait for scarlet to come out, low light or not low light, scarlet will blow the 1d out of the park.

oh yeah, and there's the red one while we wait :)
 
I think any argument most poeple have is what to do with their time while they wait for scarlet to come out, low light or not low light, scarlet will blow the 1d out of the park.

For motion video? Sure. For stills? Much more questionable.

As I said, you're comparing the motion performance of a still camera to the motion performance of a motion camera. If the purpose designed motion camera wasn't superior, there would be something wrong.
 
Depends...like I said there's a pretty huge difference in other features. It's only a steal if you have a pretty narrow set of tool needs. As that expands the value of the Canons plummet.

Last time I checked, its the ability of the user and not the tool that produces quality filmmaking. As far as it be narrow is a matter of opinion. I don't know anyone who doesnt think the 5D is steal unless its been common to see full frame interchangable lens cameras for 3k.

Edit: Are you including a lens in that estimate? You'd need to add that onto the 1D's price too.

I was speaking of the 5D and I wouldn't upgrade to the 1D because of the considerable price increase and its not full frame. There are a ton of quality glass to use with the 5D at reasonable prices. No matter which way you cut it, Scarlet will significantly be more expensive and dont forget the processing power needed to handle raw.
 
Last time I checked, its the ability of the user and not the tool that produces quality filmmaking.

Last time I checked skilled filmmakers can make better films if they have better tools.

The 5D can certainly be a good deal if it fits your needs. But you can't ignore that the gap between the feature set of the 5D and Scarlet is a lot wider than the gap in price.

And seriously, if working with RAW is a problem for you, then a camera like Scarlet shouldn't even be on your radar. Why bother using a camera designed for a proper post workflow if you can't/don't know how?
 
Last time I checked skilled filmmakers can make better films if they have better tools.

I dont discount that but there is no reason to think that you can't produce quality films on the 5D. In fact, in some applications, the 5D would be my choice over the Scarlet which currently is not available and who knows when it will come out since everything is subject to change.

The 5D can certainly be a good deal if it fits your needs. But you can't ignore that the gap between the feature set of the 5D and Scarlet is a lot wider than the gap in price.

Again that is matter of opinion. I don't need all the bells and whistles in a camera; however, the 5D is far from perfect with its aliasing issues and having to use double system audio. Having said that, its another tool in my arsenal and I'm quite happy with it and in low light, it kicks any camera's arse.

And seriously, if working with RAW is a problem for you, then a camera like Scarlet shouldn't even be on your radar. Why bother using a camera designed for a proper post workflow if you can't/don't know how?

I never said its a problem or that I dont know how to handle the workflow. What I said was that you will need powerful machines to handle it which will increase cost.
 
For motion video? Sure. For stills? Much more questionable.

As I said, you're comparing the motion performance of a still camera to the motion performance of a motion camera. If the purpose designed motion camera wasn't superior, there would be something wrong.

EDIT: I did not know about the 1D Mk. IV when I posted. Two more stops low light performance over the 5D Mk. II, and usable frame rates -- absolutely incredible. I should have said, "1D Mk. IV" rather than "5D Mk. II," as this appears to be the new benchmark for low light performance.



Talking about the low light performance of the 5D Mk. II seems helpful when trying to predict the future of low light performance in video/digital movie cameras. But, to this end at least, I'm not sure how useful it is to compare the video (Redcode RAW) of the RED One to the video of the Canon 5D Mk. II.

If we want to understand what present day sensors are capable of, it would seem we should compare the low light performance of present and future video/digital movie cameras to RAW stills of the Canon, rather than its video. Otherwise, we are largely talking about the Canon's interpolation chip and codec, rather than the sensor.

Of course, the video of the Canon does have enough rolling shutter artifacts to make it unacceptable for action shots and some other applications. Makes me wonder if being able to quickly refresh the sensor is somehow at odds with low light performance? If this is the case, we may not have a tool that can do both action and low light for some time.

Just throwing around some ideas. The people here know a thousand times more than I about this stuff.
 
Last edited:
And seriously, if working with RAW is a problem for you, then a camera like Scarlet shouldn't even be on your radar. Why bother using a camera designed for a proper post workflow if you can't/don't know how?
In all fairness, a RAW motion-image workflow -- particularly a RedCode RAW workflow -- would be unfamiliar to most people, since most people working in HD are used to HD video workflows. Also, unless Red has changed its plans, in interviews conducted at NAB 2008 Ted Schilowitz mentioned a simplified workflow for 2/3-inch Scarlet. By that I assume that presets would yield good results, but that all the configurability available to Red One users would be there, too. And as for any hardware-related issues dealing with Scarlet footage, granted, it will take a fairly hefty setup. But, remember, the original Scarlet, though considered a professional camera, was expected to appeal to well-heeled serious enthusiasts -- and that was when it was said to support a data rate of up to 100 MBps.
 
Have you heard Canon's announcement of the 1D Mark IV?
With epic lowlight capabilities, basically better than the human eye, as described by Vincent Laforet.




And the other part is you'll be able to shoot stuff that's in focus without either:
a) nasty aliasing artifacts
b) h264 compression mush

There's very little point in "HD" if everything is soft or out of focus....

Graeme

hmmmmm. that sounds quite interesting. well for the time being, i'll keep shooting with my 5d mk2 and you keep shooting with your scarlet, and we'll come back and compare footage.....oh wait...too soon?


I'm sure canon will have "caught up" by the time scarlet is actually released.
 
I'm sure canon will have "caught up" by the time scarlet is actually released.

Their current SLR sensor chips and processing hardware can't handle more than 10 or 12 frames per second without skipping lines to bring it down to 1080p. How do you expect that they can make one in the next few months that will get anywhere near full 5K at 30 frames per second, or 3K at 120 frames per second?

Also, I very much doubt that they will move to raw workflow because of the data rates required, which needs much faster and more expensive CF cards, and the fact that there isn't yet any RAW codecs that are widely supported, which would make the workflow too difficult for the market segment they're targeting.
 
can anyone post some video links of red footage in low light? I haven't seen much myself, I know the RED's sensor is bigger, but just interested regardless.
 
When the IV can do > 100fps without line skipping you may be entitled to a comparison.

Graeme

does the speed of the sensor affect it's sensitivity? Is it very difficult (or currently impossible) for a sensor to be as fast as Epic's with the sensitivity of the 1DMk4?

Also I wonder how much of the 1D's low light performance is due to processing in camera before it's encoded H.264. Is that something that RED could emulate in camera or in Redcine? I guess you probably already are.
 
Well, h264 at that kind of data rate has to be throwing a lot of the noise away as it just can't encode the noise. Similarly, if you blur the RED's output to the extent that the "1080p" looks blurry from a Canon, that'd kill a bunch of noise too.
 
It's an interesting comparison for this reason: The Canon 5D MKII, while being designed and marketed for stills, is being chosen OVER R1 for low light motion photography. This is a bold statement considering the many drawbacks of shooting motion on the 5DMKII. I shot both formats on a recent project in downtown L.A. with very little supplemental lighting. The 5D blew the R1 away in it's low light performance, and while compressed and 8 bit and H264, etc etc, it still looked better than what I was able to achieve on the R1 with 1.3 lenses and a Full Debayer post process. (The Canon lenses were 1.8 or 1.4) Sure the 5DMKII has A LOT of drawbacks but I think it's important to note that sensor sensitivity is at the top of many DP's lists for desired features.

There's a lot of talk about modules, Wi-Fi, GPS, Remotes, and all these things but to me one of the most important things is achieving clean images with high DR in low light. It opens up whole new possibilities for approaching projects in new ways and creating unique images. I am really hoping the MX is significantly more sensitive than the current RED Mysterium sensor.
 
hmmmmm. that sounds quite interesting. well for the time being, i'll keep shooting with my 5d mk2 and you keep shooting with your scarlet, and we'll come back and compare footage.....oh wait...too soon?


I'm sure canon will have "caught up" by the time scarlet is actually released.

thats hilarious... and it makes me cry... Nov 30th is going to make me jump off a bridge like oct 30th tried to...

sorry mario, but you princess is in another castle.

I really anticipate new cameras from Sony and Canon before scarlet is actually available in a tangible, adult who makes a living sense.

here is to hoping my well founded cynicism is way wrong!

till then, i will keep using my XHA1

(which has really decent low light performance!)

real name Jeff Morgan
 
Back
Top