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 :)

ISO 500-4000 comparison...

Status
Not open for further replies.
Well whatever image you post, someone is going to point out even the tiniest flaws in them. Please keep them coming... the big picture is that it looks awesome so far. If other cameras are held to the same scrutiny, it is very likely that Red comes out ahead (most cameras will produce terrible images at 32,000 ISO).
 
Thanks Glenn - I think you're right about the kind of scrutiny our images get. Looking at the big picture though, I'm happy where we are today. I had no idea 32k ISO could look even like an image that you can recognise!

Graeme
 
It is simply INCREDIBLE. period.

Nobody has ever dreamed about such high resolution AND quality AND ISO capabilities...At THE SAME TIME !!!!!

Cheers !
 
Jannard said:
Correcting for anomalies at ISO 32000 is not the highest priority on our list... but yes, it can be corrected for.

Jim
Click to expand...

Well if it's just a near invisible lens characteristic then the only correction necessary would be to use a higher gain instead of boosting it in post. Did all of the tests (4,8,32k iso) use the same gain setting?
 
Jannard said:
Correcting for anomalies at ISO 32000 is not the highest priority on our list... but yes, it can be corrected for. All the posts were made to show the elasticity of our image and to have a bit of fun. None are perfect, but we haven't come near to finishing the DSP.

We heard that one of our competitors recently said that RED was a noisy ISO 250 camera. We kinda wanted to poke a bit back at them. :-)

Jim
Click to expand...

And Poke you did! I would call it more like a deathly blow or a killing jab though.
 
I still don't understand the differene between gain and ISO as it relates to digital. Isn't a higher ISO just gain? Is there an adjustment for both? I always thought putting an ISO function on a digital camera was just to be a "familiar" feature vs calling it gain.

Cheers,
Chris
 
KnockOutFilms said:
I still don't understand the differene between gain and ISO as it relates to digital. Isn't a higher ISO just gain? Is there an adjustment for both? I always thought putting an ISO function on a digital camera was just to be a "familiar" feature vs calling it gain.

Cheers,
Chris
Click to expand...
I admit, I've been a bit confused with this as well. I'm equally confused as to why - with such great sensitivity levels - they'd even consider using gain in the first place (especially when these are images they know will be scrutinized to death). Jim said something relating to it in another post:

Jannard said:
Right now, gain up takes an upload. So we just shot everything with gain today. Tomorrow we shoot normal.

Jim
Click to expand...
...but what "gain takes an upload" means or how it's different from the ISO rating is beyond me. Perhaps someone could shed some more light on this subject? (har, har...) :)
 
Taken from Bobdiaz on dvxuser

"The Gain Switch can be thought of as a volume control for the video signal. When you play music on the stereo and the music comes to a soft quiet spot, you turn up the volume to make the music louder. In the same way, Gain increases the voltage level in the video signal making the image look brighter. Voltage Gain is calculated as follows:

Gain (in dB) = 20 x LOG(Voltage Out / Voltage In)


To save everyone a lot of trouble, just remember, a +6dB Gain is equal to a 2x increase in the signal. A +12dB Gain is equal to 2 x 2 or 4x increase in the signal. A +18dB is a 2 x 2 x 2 or 8x increase in the signal. When we express Gain as dB, every +6dB increase represents another doubling the signal. Another way to think of it as F Stops. If I open the lens from F8 to F5.6 (1 F Stop), I double the light. If I open the lens from F8 to F4 (2 F Stops), it's 2 x 2 or 4x increase in light. You could say it's like this:

+6dB = Adds 1 F Stop of light
+12dB = Adds 2 F Stops of light
+18dB = Adds 3 F Stops of light

If +6dB is like adding an F Stop, +3dB is like adding 1/2 an F Stop of light and +2dB is like adding 1/3 an F Stop."

I presume Jim was refering to "gain takes an upload" as meaning he has no switch on the camera, rather he has to upload it via CF or USB and recalibrate the sensor with extra gain.

I also presume that Jim and Co. created the normal shots without the use of gain, using ND's, f22, etc to create the ISO ratings rather than modifying the settings of the sensor.

My presumptions have been known to be wrong btw lol :)
 
There's two things - analogue sensor gain ahead of the A-to-D, and digital gain, which occurs post digitisation. Both are useful.

What we're doing in these tests is to show that if you under expose, and gain up in post (ie digital gain) what results you get.

Graeme
 
Graeme Nattress said:
There's two things - analogue sensor gain ahead of the A-to-D, and digital gain, which occurs post digitisation. Both are useful.

What we're doing in these tests is to show that if you under expose, and gain up in post (ie digital gain) what results you get.

Graeme
Click to expand...

So when you mention ISO, you are talking about gain in post? Does the Red have the ability to gain between the sensor and wrting the file? Plus Gain/ISO as it converts from Raw to whatever? Maybe I'm asking the wrong question because of my lack of understanding?

Thanks,
Chris
 
We have ability to do both types of gain. What these tests are are "equivalent ISO".

Graeme
 
semi off topic but:

i just got a canon digital rebel xti, and im really happy with the way everything looks. if red looked like this, i would be more than happy to shoot my movie with it. but i know that red probably blows the hell out of the canon digital rebel xti.. can anyone shine a light on this for me? like how much canon xti ass does red kick, exactly...
 
I own a Canon DR XTi (D400) and I've shot a ton of 10 megapixel RAW pics with it. It is an excellent camera for the price, and the results are great...but I assure you IMO frames from RED One should blow the XTi's frames away. They're reasonably close in technology...but they'll be far apart in results IMO.
 
Does this mean that the shots posted at high ISOs "with full gain" were using both kinds of gain - the "full gain" that the camera applied, as well as the digital gain applied to the image in post to bring it out of the dark?

All this talk of gain is making my head spin. I think I'm just going to expose correctly the first time. :)
 
zak forrest said:
semi off topic but:

i just got a canon digital rebel xti, and im really happy with the way everything looks. if red looked like this, i would be more than happy to shoot my movie with it. but i know that red probably blows the hell out of the canon digital rebel xti.. can anyone shine a light on this for me? like how much canon xti ass does red kick, exactly...
Click to expand...

The Canon digital rebel xti (400D) is about the best comparitive digital SLR sensor available, size and pixel wise at least.

The following shows the sensor figures for both the RedOne and the Canon:

RedOne (@35mm 4k RedCode)
Sensor Physical size: 22.2mm x 12.49mm
Pixel Size: 5.5µm
Pixels: 4096 x 2304
Megapixels: 9.5


Canon EOS 400D Digital Rebel XTi:
Sensor Physical size: 22.2 x 14.8mm
Pixel Size: 5.7µm
Pixels: 3888 x 2592
Megapixels: 10.1

If you take a picture on the Canon then crop it by 405 pixels in height (Although it'll be slightly lower resolution) it ought to give you the same field of view as that lens would when attached to the RedOne.

As for colour and light only Graeme and co with hands on could inform in any way shape or form at this point.
 
Gibby said:
but I assure you IMO frames from RED One should blow the XTi's frames away. Reasonable close in technology...but far apart in results IMO.
Click to expand...
That's actually an interesting theory, Steve. Since both cameras are ~12MP it would be very interesting to do a side-by-side and see how the results differ. After all, many people are calling RED a "dSLR on steroids." Perhaps we can add some XTi test shots to the LART; I'll add that suggestion to the list.
 
Out of interest here's the figures for the Nikon D40x:

Nikon D40x:
Sensor Physical size: 23.6 x 15.8 mm
Pixel Size: 5.89µm
Pixels: 3872 x 2592
Megapixels: 10.1

If you took pictures on the D40x you'd need to crop them by -230 px width wise and -543 px height wise in order to see the same field of view as the attached lens would on the RedOne.
 
I was studying sensor sizes on comparitive DSLR's the other day by the way which is why I worked these out... hope they're correct now!

The formula to check how much you need to crop an image on your camera is as follows (You will need to know the sensor's physical size and amount of pixels horizontally and vertically, your sensor must also be bigger than 22.2mm x 12.49mm, but DSLR's typically are):

Height
Ph = Pixel Height
Sh = Sensor Height
(Ph/Sh)x(Sh-22.2)=X


Width
Pw = Pixel Width
Sw = Sensor Width
(Pw/Sw)x(Sw-12.49)=X

X being the amount to crop either horizontally or vertically.

Canon sensor info can be found halfway down this page.
 
Would be nice to see what 10k$ worth of glass would do to a rebel's image... -;)

Or did I misintertpret where you plan to put the PL lenses? -;)

I Certainly hope the glass only will be a big advantage for the Red.

Cheers!
Gunleik
 
Status
Not open for further replies.
Back
Top