More HDRx...

[Subhadip Sen]

My apologies. I don't mean to denigrate the tech. HDRx™ is amazing. It's groundbreaking. I really want it. But it's not exactly the same as the real thing. Almost everything I work on uses the Foundry's motion blur. It's not always perfect.

To deny that HDRx™ isn't the same as actually capturing the true motion blur of native 18 stop sensor is to misrepresent it. I don't feel you are misrepresenting it. But every 'trick' has a dark side. I'm not saying the tricks shouldn't employed 99% of the time. Just that in the other 1% you wish you could do it natively.

Even The Foundry is the first to say that their tools are great. Their tools save time. But they don't always work perfectly.

If you like the look of Magic Motion more than normal motion blur then it's a feature. If you don't, then you'll never get it perfect. You'll usually get it pretty damn near perfect but there'll always be artifacts.

You are making a critical assumption that the MNMB implementation is exactly the same as The Foundry's current products. Which is clearly not true as Jim has said "working on...". They are no doubt fine tuning an improved implementation that works much closer to true motion blur.

Another thing worth remembering - even if the implementation is exactly the same - the blur will take its effect on the X-track, which is detailed and blur-less. Helps immensely for synthetic blur.

 

[gbalaji]

Jim,

Could you post us some light meter readings for inside / outside / especially sky & mountains if you have recorded one and would like to share with us if possible, for us to get clear picture of how much latitude we are talking about. Image looks great both on Still & Motion but people cannot make out from web compressions and this would help them to understand the value of technology you people are working on.

HDRx rocks!

 

[Paul E. McCarthy]

Jim,
Incredible. Your all genius' in my book.

Please invite me to go the beach with your epic, a 6x6 ultra bounce, some matte silver show card....and, some french/italian models.
We'll shoot a "test".

it will be fun.
What do you say.

 

[Matt Drake]

Haha this is absolutely insane!!!!!! Congrats to the RED team on such a phenomenal accomplishment. This will go down in camera history.

 

[Neil B Smith]

This was shot with Peter Lik and Mick Plumridge in a ghost town near Las Vegas over the weekend. Both said "impossible" when we went into the barn. It was wicked dark in there and blistering noon desert-sun outside. They both fell over dead when they saw this. I called 911. They are OK now...

Jim

:001_tt1: WOW, WOW and more WOW. Loving the image and loving what RED is doing. This is how I've always dreamed of what a camera should do, and now its happening with RED!! Can't wait to get my stricky mits on a Scarlet and start shooting great images like this and make my clients go WOW.

 

[Jannard]

Jim,

Could you post us some light meter readings for inside / outside / especially sky & mountains if you have recorded one and would like to share with us if possible, for us to get clear picture of how much latitude we are talking about. Image looks great both on Still & Motion but people cannot make out from web compressions and this would help them to understand the value of technology you people are working on.

HDRx™™ rocks!

We didn't take a meter with us... ugh.

But it didn't take a rocket scientist to see that there was no way to hold the exposure inside and out. Peter has been shooting for almost as long as I have. Mick about half as long. There was no way to hold both inside and out with any camera that we have ever shot. This was a DARK barn and brilliant noon-day sun outside.

I told the guys that no one would believe this...

Jim

 

[Gavin Greenwalt]

The first law of image processing is that you can never manufacture data.

Easy way to break MNMB. Go fire some tracer rounds in the desert at night. The trackers won't be able to find a pattern.

Very fast moving objects might only be in frame for one a single frame. No amount of special sauce will be able to track that.

Just two quick examples. Almost every example would be an 'outlier' from normal shooting scenarios but I would like to see some HDRx™ footage of a grinder on steel. Nothing could track it reliably.

Also texture which is repeating tends to throw it for a loop. I've had forests for instance break every optical flow motion blur solution I've thrown at it. Again it's not an effort to knock The Foundry or RED. I love both. I'm a total fanboy for The Foundry's technology. But they also aren't wizards.

There is actually a pretty good discussion on the Nuke User mailing list right now (including input I believe from Foundry developers) discussing the merits of motion blur in post.

 

[Peter Weisberg]

We didn't take a meter with us... ugh.

But it didn't take a rocket scientist to see that there was no way to hold the exposure inside and out. Peter has been shooting for almost as long as I have. Mick about half as long. There was no way to hold both inside and out with any camera that we have ever shot. This was a DARK barn and brilliant noon-day sun outside.

I told the guys that no one would believe this...

Jim

It's almost quite comical really, you've stumbled into something so unbelievable that people are having a hard time believing it. Quite the paradox Mr Jannard.

 

[Deanan]

Haha this is absolutely insane!!!!!! Congrats to the RED team on such a phenomenal accomplishment. This will go down in camera history.

Anybody else notice the counting of the frame rate? Frame counts to 100 and counts in multiples of 4! Makes sense, but pretty bizarre at the same time! :crazy:

That's decimal time not frames

 

[Jannard]

The first law of image processing is that you can never manufacture data.

Easy way to break MNMB. Go fire some tracer rounds in the desert at night. The trackers won't be able to find a pattern.

Very fast moving objects might only be in frame for one a single frame. No amount of special sauce will be able to track that.

Just two quick examples. Almost every example would be an 'outlier' from normal shooting scenarios but I would like to see some HDRx™ footage of a grinder on steel. Nothing could track it reliably.

Also texture which is repeating tends to throw it for a loop. I've had forests for instance break every optical flow motion blur solution I've thrown at it.

Gavin... while it is not possible for us to do ever test that someone suggests, I am quite sure that the tests will be done by owners of EPIC. The capability of HDRx will either stand the test of time... or not.

Jim

 

[Neil B Smith]

Now you can see why we want to put this in ALL our cameras!

Jim

Jim, you're a star!!! [Big hugs all round]

This is sooooo going to be worth spending an extra $1,000 for and suffering the excruciatingly long wait for my little peice of camera perfection.

Jim what is RED going to be giving us in 10 or 20 years time??????? Is there really much more you can do??

 

[Jeff Mustaman]

just amazing...

 

[Sanjin Jukic]

We have been thinking about everything!

I have been on the phone with Graeme about all the possibilities. Both of us, plus Jarred and Deanan, know exactly how big a breakthrough this is. Graeme is working 24/7 to give us the best post tools possible. The Foundry is also involved. Big stuff.

Jim

I'm a bit late on this but it's a great shot.

Also happy that RED (Graeme Nattress) is working with The Foundry to give us the best tools.

 

[Pascal Scheffers]

Gavin... just so you know, the Chief Scientist of The Foundry, Simon Robinson, said... "Frankly, with results like this we were amazed that Jim was even looking at further options."

Heh - then he clearly hasn't had as much time as we had to lurk and post on these boards. There's always a need for more at RED

So far you've nailed resolution, RAW encoding, now DR and the Magick Motion stuff fits in there too.

What's up next... there are a bunch of other interesting tricks with light and sensors left. We have polarization; angle of incidence, which the FX guys would love; weird lenses; fixing the diffraction limit... pretty sure you'll surprise us.

 

[Roberto Lequeux]

A dense forest being a no-no would be a bummer, but it doesn't make sense to me Gavin. Are you also saying that it an extremely unlikely scenario? Was is compressed footage you were dealing with?

I'd like to know if the forest was compressed footage, and if there was a lot of motion blur on the trees.

The tracers can be added in post several times over with the money you saved on trying to get that kind of fill in the barn.

 

[RivaiC]

Oh yes, i feel it, i love the motion, i think you nailed it Jim, that's definitely filmic ! Well done...

What i love is not only about HDR-x but the fact you can manipulate those 2 independent stream. First thing comes to my mind, the dynamic range is greeeaatt !! But hey...i need to adjust a little contrast on outdoor and i dont want to do masking, i can do that and we have complete control over that !

Great Jim..my sincere congratulations for you, this is a must have camera !

Rivai

 

[Steven-Marc C.]

Gavin, which 18 stop sensor are you talking about?

Jim is showing us what can be achieved today. Even if it breaks down in 1% of situations (or more) it's still an fantastic new tool to have.

Well done, guys!

 

[Jannard]

Graeme is unbelievable. I am shocked every day at what he does. Simon at the Foundry is also a rock star. Between the two...

Jim

 

[Kris Anderson]

Perhaps Epic and Scarlet should ship with a defibrillator for all those DP's who are going to fall over dead?

With red paddles, of course.

Hey Mr EOSHD....

http://www.youtube.com/watch?v=xGcwQNwU498

 

[Gavin Greenwalt]

Gavin... while it is not possible for us to do ever test that someone suggests, I am quite sure that the tests will be done by owners of EPIC. The capability of HDRx™ will either stand the test of time... or not.

Jim

I still feel like you're being too black and white. I think HDRx will stand the tests of time. But I also am relatively certain there are going to be times when MNMB will flat out kick the can and be unusable.

It's not an either/or situation.

Part of selling professional tools is to know your tool's limits and work around them. It can stand the test of time and have limitations to be conscious of.