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 :)

Off the chart...

sounds ridiculously amazing...
where does this leave the red one?

As the cool cam it's always been?
As REDs currently only available camera?
As the one you can pay your Epic with?
If you have an MX RED1... -:)
xxxx ?

You know, if someone to all of a sudden give me a D-20 or F35, I am sure I could find uses for them. -:)
 
Jarred. I have an idea of what HDRx might be. Instead of taking multiple frames at a different exposure setting, you twisted the whole idea around on us...

Hear me out tho, because if you didn't use the technique I am about to mention, you should consider it for R&D and credit the name of the method to me "The O'Connor effect" or something like that sounds good enough to me. :)

In all seriousness though. The technique involves giving EACH PIXEL the ability to read and adjust to different amounts of light that come in contact with it through the lens. I.E. Giving each pixel a constantly changing dynamic ISO or sensitivity range. This is (I believe) something that has never been done before and could yield some amazing results.
For example: Say our are shooting a subject outside in bright daylight. You have to, in some degree, expose more for the subject or the sky. If you expose properly for the subject, you blow out the sky. If you expose for the sky, you crush your subject in blacks. This we all know.
We also know that DR can go a long way in bringing out the subject/sky from an improperly exposed image, but if you had a smart enough processor / chip / software, it could be doing all the thinking for you and manipulate each pixel for the correct sensitivity to keep BOTH the background AND the subject in a correct exposure range.

I hope this hasn't confused anyone. It makes sense to me and I would be more than happy to try and re-explain it.

Regardless of how you pulled HDRx off, I am sure it is amazing and I can't wait to see where you are all going. I love RED and tell everyone I know about you guys even if they have no idea what your achievements mean to the industry.
 
Once HDRx is implemented (I assume via a firmware upgrade) will it become the default way that the Epic shoots? Meaning there are effectively no compromises?
 
We still have some work to do but the entire team was very quiet, eyes and mouths wide open when the 1st images came up on Graeme's screen.Jim

Whew..what a scene..like something out of edgar allan poe..alot of suspense for this test chart...
 
Jarred. I have an idea of what HDRx might be. Instead of taking multiple frames at a different exposure setting, you twisted the whole idea around on us...

Hear me out tho, because if you didn't use the technique I am about to mention, you should consider it for R&D and credit the name of the method to me "The O'Connor effect" or something like that sounds good enough to me. :)

In all seriousness though. The technique involves giving EACH PIXEL the ability to read and adjust to different amounts of light that come in contact with it through the lens. I.E. Giving each pixel a constantly changing dynamic ISO or sensitivity range. This is (I believe) something that has never been done before and could yield some amazing results.
For example: Say our are shooting a subject outside in bright daylight. You have to, in some degree, expose more for the subject or the sky. If you expose properly for the subject, you blow out the sky. If you expose for the sky, you crush your subject in blacks. This we all know.
We also know that DR can go a long way in bringing out the subject/sky from an improperly exposed image, but if you had a smart enough processor / chip / software, it could be doing all the thinking for you and manipulate each pixel for the correct sensitivity to keep BOTH the background AND the subject in a correct exposure range.

I hope this hasn't confused anyone. It makes sense to me and I would be more than happy to try and re-explain it.

Regardless of how you pulled HDRx off, I am sure it is amazing and I can't wait to see where you are all going. I love RED and tell everyone I know about you guys even if they have no idea what your achievements mean to the industry.

The problem is that you have to know about what you are shooting before you are shooting. For anything in motion, every exposure of the pixel could be different, so having a dynamic system that could both detect what it thinks the right exposure is would have to know about the future, since in a typical exposure, like 1/48, could change.

If you were to add this to the silicon, the biggest problem is that extra circuitry that is inactive introduces more problems. It decreases active area and adds noise and heat. Both of these of course degrading the image. This is one of the reasons I have been told global electronic shutters are not used because what they take away from the sensor is not worth what they add. The ability to send control signals to every pixel to control for sensitivity would take time, including time for the circuit to stabilize.

One method could be using multiple photosites instead of one large one that are tuned differently for sensitivity, and the results brought together (under the same color in the filter pattern). In this case, I would make one set as sensitive as possible and the other globally controllable to two different sensitivities. That way, during low light shooting, they'd be the same and offer noise rejection (having multiple sites can do this under a single lens) or, set the second set to be WAY less sensitive so all the blown out parts will not be, and just use a larger A2D converter once the values are summed.
 
The 18+ Stops is at 5k, right?

Did Jim say that? I wouldn't presume that to be the case. The definitive target has been 4k with an eye toward more. If there is a sacrifice to be made to achieve their HDR, I might expect to spend that 1k as overhead... Oversample for noise reduction perhaps. I am eager to learn more about what they are doing.
 
We just gave birth to RED HDRx last night and the results are jaw-dropping. There is still a ton of work to do to tune up the program. After talking to the engineers, it seems highly unlikely we can tie all this stuff up before we begin shipping EPIC... which is what I said in an earlier post. This is a feature that will be added to the camera when fully ready. It is complicated stuff. But very cool.

Jim

A LOT OF PEOPLE
(whiny, child-like)
But Jiiiiim! I want it noooow!
 
5K 120fps 13.5 stops DR/3K 150fps burst 12+ stops DR is more than good enough to ship cameras as long as they're bomb proof. We really appreciate all the extra icing on the cake, we really do. What's more exciting to me (right now) is your devotion to getting the cameras out of the gate as soon as they're ready.

Thanks RED.
 
If you were to add this to the silicon, the biggest problem is that extra circuitry that is inactive introduces more problems. It decreases active area and adds noise and heat. Both of these of course degrading the image. This is one of the reasons I have been told global electronic shutters are not used because what they take away from the sensor is not worth what they add. The ability to send control signals to every pixel to control for sensitivity would take time, including time for the circuit to stabilize.

That is what I was thinking as I tried to explain the Technique. Noise could be an issue as well as heat from the (literally) millions of processes required by the processor to dynamically control each pixel. I guess this would have to start at a small level of resolution, like VGA, and once you perfect that level move up the Pixel count over time as the technology catches up with the methods.

It was just an idea I came up with about a month ago when I was trying to figure out how HDR could work with motion... I guess I still don't know how it could work on a technical level. Can anyone explain to me how HDR is captured on the RED? Do you capture multiple frames with different exposures? What accounts for the motion that happens between frames? I am confused now...:001_unsure:
 
We just gave birth to RED HDRx last night and the results are jaw-dropping. There is still a ton of work to do to tune up the program. After talking to the engineers, it seems highly unlikely we can tie all this stuff up before we begin shipping EPIC... which is what I said in an earlier post. This is a feature that will be added to the camera when fully ready. It is complicated stuff. But very cool.

Jim

Understood. :)

Were you able to record the HDRx mode to CF cards?
 
Will the HDRx be a option in Post if shot in Non-hdr? and Vi-ca versa shot in hdr and in post turned off? Not that the later is appealing to me but it does weigh in!
 
We just gave birth to RED HDRx last night and the results are jaw-dropping. There is still a ton of work to do to tune up the program. After talking to the engineers, it seems highly unlikely we can tie all this stuff up before we begin shipping EPIC... which is what I said in an earlier post. This is a feature that will be added to the camera when fully ready. It is complicated stuff. But very cool.

Jim

Jim, since I never thought that I would see the day that a motion imaging device would surpass the human eye in DR, I can wait a bit longer. Thank you for again making the impossible REAL! And very cool...

Mike
 
Oh, no more waiting please! Revoke all weekend passes from the Bomb Squad, and let them do what they do best!
 
This is really amazing guys.

But I think after all this waiting what most of us want is EPIC and Scarlet to come out as fast as possible.

Give us the bells and whistles later on. :smiley:
 
Back
Top