Playing with the Forged Carbon Fiber Weapon and it's AWESOME!

[John Marchant]

Yep, Weapon on the top. My ugly mug too, sorry everyone.

The intangible thing here is that everyone who laid hands on the camera for the brief period we had it loved the image and commented on how clean it was. There may not be much in it measurably, but the impression in use was significant. That counts for plenty in its own right.

I suspect the automatic blackshade is largely responsible. The subtleties of a good or indifferent blackshade are important on Epic Dragon. Losing that headache and always being at an optimum has practical and image-based effects. Also, Weapon's target temperature is based on the sensor temp, which is an important distinction in terms of continually maintaining optimum blackshade.

 

[Kevin Marshall]

Yep, Weapon on the top. My ugly mug too, sorry everyone.

The intangible thing here is that everyone who laid hands on the camera for the brief period we had it loved the image and commented on how clean it was. There may not be much in it measurably, but the impression in use was significant. That counts for plenty in its own right.

Yeah, subjective observations do matter. And, facetiousness aside, I do see a difference between these two shots, again just not the multi-stop difference others are seeing - comparing Weapon 5000 ISO to 1600 or 1250 Epic ISO.

I suspect the automatic blackshade is largely responsible. The subtleties of a good or indifferent blackshade are important on Epic Dragon. Losing that headache and always being at an optimum has practical and image-based effects. Also, Weapon's target temperature is based on the sensor temp, which is an important distinction in terms of continually maintaining optimum blackshade.

Amen to that. Even with multiple storage banks on the Dragon, as an AC I still often have problems getting commitments on frame rates and/or shutter angles we might shoot at at the beginning of the day, which is fair - the DP and Director should be able to change their mind on what's best for the shot without considering things like black shading. And even when I do have things nailed down, and have done multiple black shade passes, if I'm the 2nd AC, I've never gotten the 1st AC or DP to actually utilize the appropriate black shade for the settings. This is one ofmy ffavorite improvements in the Weapon.

And if you're saying Weapon's temperature controls are based on sensor temp instead of core temp, then that is awesome. During a shoot, the sensor temp always seems to be 2-3 degrees above where it was when I black shaded during prep, ddespite the core temp being the same. Even trying to re black shade during the day has the sensor temp dropping back down again.

 

[Patrick Tresch]

Thanks for sharing that clip, looks to be completely identical to Dragon in sensitivity!

Both camera uses Dragon sensor. One is an Epic camera the other an Weapon. There is no such thing as a Dragon camera.

Pat

 

[Gunleik Groven]

Weapon on top, I hope.

Also see a bit of a strange flare and magenta edges to the burnouts, I frankly cannot remember to have seen a lot in Dragon footage. But this is a demanding shot when gained this much.

I must say that I think of the Dragon upgrade as primarily a system upgrade, more than an image upgrade, until I see things otherwise.
I am sure it's a bit cleaner and has some more DR, and I can really see a bit of (surprising) highlight and lowlight differences in these shots, but I don't know how they are developed.

The systemwide changes alone are good reasons for an upgrade if you do a lot of rentals or big-production stuff.

Heck, I have been shooting non-stop on different projects since april, and would love them myself.

Prores and automatic blackshading and ports where men with man-fingers can access them easily.
All great stuff!

The weapon I really want though, has 8k on its side.

As that's where I see the big image changes coming in.

I guess there could be a lot of improvements happening in the metadata/control department, too.

 

[Charles Bergquist]

Both camera uses Dragon sensor. One is an Epic camera the other an Weapon. There is no such thing as a Dragon camera.

Pat

Understood, but it was announced that the Weapon would be better in the shadows and eliminate the CMOS smear. Same sensor different processing of the raw data. They look almost identical in the shadows.

 

[Gunleik Groven]

Understood, but it was announced that the Weapon would be better in the shadows and eliminate the CMOS smear. Same sensor different processing of the raw data. They look almost identical in the shadows.

If it is the half-stop Jarred has mentioned, then that would be pretty similar.

I don't think it's day and time to pass judgement on this.

What I am sure is that perception of ease of use and its ability to integrate into multiple workflows, has been two of ARRIs (very good) strongpoints, and if RED has addressed those, that is really major.

And the thing looks sexy...

 

[Kevin Marshall]

I must say that I think of the Dragon upgrade as primarily a system upgrade, more than an image upgrade, until I see things otherwise.

Absolutely. There may not be any major image quality jumps, and only the Carbon gets the higher frame rates, but as an AC every change/improvement to Weapon over Epic had me nodding my head. Unfortunately it's harder to convince people to drop money without an image improvement to go with everything else.

Auto black shade with a huge range, better thermal management, quieter cooling, better fan and port placements, built-in scratch mics, LUTs, ProRes, better back focus location, integrated top plate, dumb side controls, built-in wifi, simpler OLPF swap (and smart)... So many quality of life improvements - many of which we can do now with the Epic, but only by tacking on a bunch of other boxes. Now it's all in one (even smaller) package.

 

[Zeb B]

Been traveling a bunch with limited access to interweb. No time to sort through RU to see if there are any R3D's posted from field users yet. Anything up yet?

Are the the couple R3D's posted on this thread the only ones anyone has uploaded?

Is it just me or does RED shadow noise have a very distinct DNA in that it creates colors that are not there? Why not just neutral luma noise . . . what's with the big grain chroma noise? I'm guessing her shirt was black and skin was pink.

 

[Nick Morrison]

All I know, is the Epic Dragon's I've shot can't shoot this clean at 4000 ISO:

And here's a comparison of Weapon and Epic Dragon at 3200 ISO (both with exact same settings):


WEAPON 3200 ISO

EPIC DRAGON 3200 ISO

To me, the Weapon at 3200 looks CLEAN. Incredibly usable. Whereas Epic Dragon has already started to get noisy. Comparing both these shots at 5000 ISO almost seems to bring out too much noise in both, so Weapon's "advantage" becomes less clear. Knock it back to 3200, and I think it becomes more readily apparent.

For me, what this means practically, is I'd be less worried about shooting 1600 now. We tend to shoot at 800 max on LLO, as I'm a fan of "clean" blacks. But on Weapon, I can see 1600 being "clean". Or even higher.

And that to me....would be a big deal.

I'm not so worried about 5000 ISO. I'm expecting noise there, and that's not where I'd shoot on a typical day. What I care about is noise at 1600-2500: lowlight situations that every now and then are unavoidable, especially when shooting slow-mo.

What these early examples from Weapon are telling me is...that in these more likely lowlight ranges...Weapon may well indeed be cleaner than Epic Dragon, in a way that could impact my shooting quite a bit.

So...color me impressed...and intrigued.

 

[Matt Ryan]

I took the Weapon shot and processed with Redlogfilm and a custom curve that didn't crush the low end so much. Used a bit of neat video and it is like glass in motion. A still frame is the worst it will ever look ;-) These shots aren't a great comparison because of the varying camera positions, distance, difference in focal length, and focus make it hard to decipher between the two. Differences in shots aside, I like the Weapon "look" better here. To me it has better color, slightly less noise, and slight sharpness improvement (pending a true side by side and without knowing where focus was set and if it was double checked for accuracy). I also noticed Weapon is slightly easier to denoise when I ran them both. Not to mention the Dragoncolor2 blue channel clipping isn't present in the Weapon footage. If you right-click and open in a new tab to view full size, with the same amount of NR, the Weapon D footage looks slightly cleaner than the Epic D footage. Tough to make a definitive assessment because the variables are slightly different. But it's fun to speculate anyways until I use one myself.

 

[Elsie N]

All I know, is the Epic Dragon's I've shot can't shoot this clean at 4000 ISO:

And here's a comparison of Weapon and Epic Dragon at 3200 ISO (both with exact same settings):


WEAPON 3200 ISO

EPIC DRAGON 3200 ISO

To me, the Weapon at 3200 looks CLEAN. Incredibly usable. Whereas Epic Dragon has already started to get noisy. Comparing both these shots at 5000 ISO almost seems to bring out too much noise in both, so Weapon's "advantage" becomes less clear. Knock it back to 3200, and I think it becomes more readily apparent.

For me, what this means practically, is I'd be less worried about shooting 1600 now. We tend to shoot at 800 max on LLO, as I'm a fan of "clean" blacks. But on Weapon, I can see 1600 being "clean". Or even higher.

And that to me....would be a big deal.

I'm not so worried about 5000 ISO. I'm expecting noise there, and that's not where I'd shoot on a typical day. What I care about is noise at 1600-2500: lowlight situations that every now and then are unavoidable, especially when shooting slow-mo.

What these early examples from Weapon are telling me is...that in these more likely lowlight ranges...Weapon may well indeed be cleaner than Epic Dragon, in a way that could impact my shooting quite a bit.

So...color me impressed...and intrigued.

I know this is about separation, but truth be told, I could live with either one of those shots... which for me bodes well for integration.

 

[Patrick Tresch]

It looks like the lenses are not the same. The coma on the Epic looks bigger.

Pat

 

[Charles Bergquist]

Epic Dragon looks half a stop brighter in the shadows. Maybe the half a stop cleaner is achieved by clamping the low end. Hope that Phil does one of his rad tests.

 

[Gunleik Groven]

Absolutely. There may not be any major image quality jumps, and only the Carbon gets the higher frame rates, but as an AC every change/improvement to Weapon over Epic had me nodding my head. Unfortunately it's harder to convince people to drop money without an image improvement to go with everything else.

Auto black shade with a huge range, better thermal management, quieter cooling, better fan and port placements, built-in scratch mics, LUTs, ProRes, better back focus location, integrated top plate, dumb side controls, built-in wifi, simpler OLPF swap (and smart)... So many quality of life improvements - many of which we can do now with the Epic, but only by tacking on a bunch of other boxes. Now it's all in one (even smaller) package.

If the difference is indeed as big s the different between R1 MX and Epic MX, it would be kinda big IMHO.

Yet there are those out here claiming there isn't any difference between the two, while after exensive testing my employer at the time and I agreed that they should mostly not be mixed in the same production unless lighting is optimal (where they are indeed similar).

From these first pictures that may look like it is the case.
Still I am a bit surprised with especially one Epic Dragon artifact, unless weapon indeed has a CA limiting function built in.

Been traveling a bunch with limited access to interweb. No time to sort through RU to see if there are any R3D's posted from field users yet. Anything up yet?

Are the the couple R3D's posted on this thread the only ones anyone has uploaded?

Is it just me or does RED shadow noise have a very distinct DNA in that it creates colors that are not there? Why not just neutral luma noise . . . what's with the big grain chroma noise? I'm guessing her shirt was black and skin was pink.

yup, I find this interesting when I look at stills, too, but I have a hunch this contributes to the "organic" looking noise in moving images. And it's easy to get rid of for stills.

Not in denial. Just an observation.

 

[Brian Boyer]

I think the Weapon shot is clearly cleaner than the Epic shot. With both at ISO5000, DRAGONcolor (to negate the artifact around the flashes), REDlogFilm and no D.E.B., the Weapon shot seems to hold to the claim of 1/2 stop gain in the shadows. The curtains, for example, are cleaner and have better color.

The diamond shaped reflections in the Weapon shot are weird, though.

Epic Dragon looks half a stop brighter in the shadows. Maybe the half a stop cleaner is achieved by clamping the low end. Hope that Phil does one of his rad tests.

I think the noise in the Epic shot may be giving the illusion of brighter shadows. It's similar to how analog tape hiss would give the illusion of high frequency detail. When Dolby (or dbx) noise reduction was engaged, material would tend to sound dull by comparison because the hiss fooled the brain into thinking it was actually information.

Stacey mentioned the shadows perhaps being crushed on Weapon. I agree that it may simply be a case of dark areas not being as affected by the noise and the Epic Dragon might look similar if it were as noise-free.

The biggest reason for the shadows looking a bit dark is probably due to it being fairly dark in the room. Raising the ISO to 12,800 only brings the best exposed faces up to about 60 IRE, with most facial highlights less than 80 IRE. Of course, I wasn't there and John can speak more to that.

I also don't know that cleaner blacks/shadow areas necessarily equal more detail. I think people sometimes expect to see detail in the blacks where there's little to no noise, not pure blacks. Detail-less blacks get interpreted as being crushed. But, at some point, there just isn't enough light to show detail in the shadows for a given medium's sensitivity.

 

[Stacey Spears]

It looks like the lenses are not the same. The coma on the Epic looks bigger.

They are not, one is a 35 and the other a 25. I don't believe this was meant as a shootout. They had a brief moment to capture the same thing on both cameras. Once more of us get the upgrade, we can do something more tightly controlled. The Xyla 21 makes it easy to test for the CMOS smearing. The front, brightest, chip causes the smearing to happen.

I think the Weapon shot is clearly cleaner than the Epic shot.

I see less noise in the Weapon version around the drapes and in the shirts. Overall I would describe it as if a veil of noise has been lifted. I always love it in a review of an audio component when they say a veil has been lifted. I have a box of veils in my closet from my years of upgrading A/V gear.

The diamond shaped reflections in the Weapon shot are weird, though.

There are the odd reflections in the Weapon version that I posted earlier in the thread. Could be related to the distance and the lens. The Weapon was closer to the subjects, which is why you see the Weapon in the EPIC shot.

On a side note, the Weapon is 4:1 and Epic is 5:1 as that was as low as it would go with the 1.8" SSD.

Is it just me or does RED shadow noise have a very distinct DNA in that it creates colors that are not there? Why not just neutral luma noise . . .

I think this is part of the debayer. At least that is my assumption. You can try the chroma noise reduction in the debayer settings to see if it helps reduce them.

I also don't know that cleaner blacks/shadow areas necessarily equal more detail.

Sometimes we perceive noise as detail.

While there is a lot less clipping on Weapon with DC2, there is still some clipping if you zoom in.

All in all, I am excited and can't wait for my upgrade notification.

 

[Björn Benckert]

If the lenses are not the same I would say it´s not really much to compare.

 

[Peter Lyons Collister ASC]

I think we should say thank you for trying to give us "something" in the way of comparison. But I think at this point we should drop this discussion and wait for more comparisons. No offense but this was never meant to be directed as much as it is being done.
We are all chomping at the bit for comparison r3ds but perhaps we just need to wait another week or so.

 

[Peter Lyons Collister ASC]

I meant "dissected" not "directed"

 

[Zeb B]

Is there a technical or production hold up on Weapon? How many of these are in the field now? Agreed that this is not the best A/B test and weird that there is nothing official posted yet. What's sup?