Welcome to our community

Be a part of something great, join today!

Register 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 :)

V-Raptor Sensor Issues

Would love some clarification on this as well.... Can this issue with V-Raptor be addressed with firmware possibly in the form of an update that provides a new protocol for black-shading that would address any specific change in color, temperature, or exposure of each of the two sensor halves in order to avoid seeing any differential between the two sides (i.e. erasing the perceived line)?

Is it only backlit, lower-contrast scenes with blooming (sun just out of frame) where this problem presents itself?

Another interesting question that I would like to confirm is that Monstro was indeed a continuous wafer sensor? I have seen people claim both in this forum, that Monstro was 2 sensors stitched together like V-Raptor and others mention that it was indeed a single continuous large format sensor....

I thought there was a lot of talk from Jared in the beginning as to the reasoning why Monstro was so expensive initially was that it was much harder for RED to manufacture and QA/QC the production of a sensor wafer so large without getting a few bad ones in each batch of production....
 
Last edited:
Definitely would love to hear an official Red response. I'm was on the verge of getting a Raptor, but since I'm constantly shooting into the sun and direct lights for music video stuff, a vertical line would be a big spoiler. Can this be fixed via firmware? Maybe improved stitching code?
 
Red Support was prompt and nice, they want to rule out that it's not hardware related but didn't ask for me to send my camera in. So, I'm not sure where this leaves the situation.


"Thank you for providing the camera log file and an image sample.

As is common with large format sensors in our industry, the RED V-RAPTOR 8K VV sensor is manufactured by aligning two sensors side-by-side to create one larger sensor. In rare situations, it is possible to notice the seam between the left and right halves of the image, however, it should not be visible under normal shooting conditions."

Since we all shoot in varying and subjective lighting situations, I imagine this will happen more and more. Not sure if they are internally planning a recall, but it definitely limits the type of shoots I'll be able to bring Raptor onto.
 
Charles Bergquist said:
Red Support was prompt and nice, they want to rule out that it's not hardware related but didn't ask for me to send my camera in. So, I'm not sure where this leaves the situation.


"Thank you for providing the camera log file and an image sample.

As is common with large format sensors in our industry, the RED V-RAPTOR 8K VV sensor is manufactured by aligning two sensors side-by-side to create one larger sensor. In rare situations, it is possible to notice the seam between the left and right halves of the image, however, it should not be visible under normal shooting conditions."

Since we all shoot in varying and subjective lighting situations, I imagine this will happen more and more. Not sure if they are internally planning a recall, but it definitely limits the type of shoots I'll be able to bring Raptor onto.
Click to expand...

In the world of things; there is no such thing as a normal lighting or shooting situation. Especially in this day and age when people are taking smaller cameras like these all over the world into all manner of environments and shooting. If they want to conquer the world with a camera, it has to be up to the task. The question is, is the stitched sensor, it's covering, or the manner employed to attach the covering to the sensor a weak point that needs to be remedied from an engineering or production viewpoint.
 
Charles Bergquist said:
Red Support was prompt and nice, they want to rule out that it's not hardware related but didn't ask for me to send my camera in. So, I'm not sure where this leaves the situation.


"Thank you for providing the camera log file and an image sample.

As is common with large format sensors in our industry, the RED V-RAPTOR 8K VV sensor is manufactured by aligning two sensors side-by-side to create one larger sensor. In rare situations, it is possible to notice the seam between the left and right halves of the image, however, it should not be visible under normal shooting conditions."

Since we all shoot in varying and subjective lighting situations, I imagine this will happen more and more. Not sure if they are internally planning a recall, but it definitely limits the type of shoots I'll be able to bring Raptor onto.
Click to expand...

Thank you sharing this. I was not aware of this flow in the sensor.
 
Michael Lindsay said:
I have a raptor and haven't seen anything... can someone tell me how to replicate the issue?

thanks
Click to expand...

...they easiest way to see it, is take the mount cap off and just move the camera around in room with any light. You'll see a split in the sensor. This is the easiest way. Obviously, no one would ever shoot like this so I understand this isn't a reasonable way to say "there's a flaw with the sensor". That said, you can put a lens on and flare it out with a point source, it can be light to the sun. You'll see a split in the recorded image on screen. I am seeming to get it more on cleaner lenses Angie EZ-Zooms for example. I think everyone needs to be careful and we need to all make sure that this gets fixed. Otherwise, there's going to be a lot of stuff that could get possibly ruined. For example, a black hole sun issue on a sensor, is ugly, but it really can be fixed, this is two different exposures on left and right side of the image, that's unfixable.
 
Here's some samples so people can see the issue : https://www.dropbox.com/t/NkhpzHyETuN57eBR

Same link that I sent to Red Engineering today, so I'll keep those that are interested in the situation updated. It does happen with a lens on and backlit scenarios.

I'll let everyone know if it's just my sensor that's a bad "stitch" or if this will affect all V-Raptors.
 
Thanks for providing these samples Charles.
I was bracing for something a bit worse but can see what people are talking about.
Yes I hope RED fixes this.

On another note, I'm pretty surprised by how well that sensor seems to push in low light.
Granted it's only one frame and not sure how the grain looks when moving but for a dark shot
like that to push with minimal noise at ISO 12800 is impressive.

Brian Timmons
BRITIM/MEDIA
 
Dynamic Range is nice, it’s about the same as the Monstro yes. This thread is addressing the sensor error that can be produced when shooting backlit. As a DP I couldn’t offer this to a Director or Production as is. I’m hoping it’s just a handful but Phil Holland has confirmed that he’s brought it up to Red. I’m afraid it’s an engineering defect on the chip matching.
 
I for one need it fixed. Its not if but when It will rear its ugly head and as Charles mentioned the Black hole sun issue was something that could be fixed in post al be it you don't want to have to do that and its now fixed and a thing of the past but having half your image a very slightly different value is not gong to be an easy paint or fix in coloring.
Imagine a marvelous 8mm R blue autumn sky and there's a line down the middle of your shot... or if you have the sun going down in a Time laps and there's 2 different values for each side of your screen.
NO BUENO!!
 
Adding to the difficulty regarding communication across various groups around the internet and multiple posts, I'll put Jarred's reply here from a Facebook post as it is useful:

"This most likely is not actually a sensor stitching issue.. every VV sensor we have made has been stitched. It seems to be an optical issue with the adapters but that analysis is a bit premature, we are still exploring."

The only thing I can add to this is echoing what I've said here on REDUSER and a Facebook reply. It doesn't occur with all lenses, I initially suspected it had something to do with back focus or even specific light sources, but then noticed it wasn't happening on certain lenses at all even. It's very, very hard to test every single variable. I'm fortunate to have a great deal of glass and adapters at my disposal, but I don't have everything when it comes down to it. But it sounds like they are working on it.

It does pertain to optics in some way, shape, or form. Whether that's lenses, adapters, or things within the sensor box; hard to tell. Possible individual assembly I imagine too, but this is all RED engineer work. We'll have to be patient to see what they uncover.

Several people haven't been able to replicate it, which is interesting. But it's too early to know if they have unicorns or just aren't using anything that exhibits the artifact.

I know this has occurred on Arri LF bodies and the general move was to send the camera in to Arri for servicing. As to what happens in that service dungeon, hard to know. Project was wrapped by the time that camera was turned around.
 
Hello ,
I think that redcode also divided image to 2 glued stream ..
RED One was first 4k digital raw cinema camera ...
And RED's can use dual HD stream for got resulting of 4k image
Same as Apple ( 5k retina ) and Dell with first 4k monitors was glued with 2 HD panels .
If You has some corrupted R3D you can see that picture will splited same as sensor.
 

Attachments

  • 1.jpg
    1.jpg
    58.1 KB · Views: 1
  • 2.jpg
    2.jpg
    90.4 KB · Views: 1
Oouuf, I seem to recall deep-aperture red-dot sensor glass reflections being an optical issue... That only took <checks notes> 14 years to button down.

Is it an adapter/optics issue, or an RF lens mount one? Cause Charles mentions that it's most easily repeatable without a lens on the mount... That implies the lens optics aren't the problem. Does something similar happen to Monstro/Dragon VV when there is no lens on it? And does it happen to Raptor more or less often when it's plain RF mount, when using an RF to PL adaptor, or RF to EF adapter is attached (but no lens)?

Phil Holland said:
I know this has occurred on Arri LF bodies and the general move was to send the camera in to Arri for servicing. As to what happens in that service dungeon, hard to know. Project was wrapped by the time that camera was turned around.
Click to expand...

See, but that implies it was a defect of that unit, which is why Arri called it back for repair, *not* a build/design flaw potentially affecting all units. In any case, we don't know yet and time will tell. Until then, most users might not notice it.
 
Last edited:
Malk Kamalko said:
2 glued stream ..
Click to expand...

Except that you would need 4X. But hey, still a chance things are written in tiles or whatever. In your case though that's just some sort of copy corruption unfortunately. Data got beaten on in the process of either formatting the media and recovering. Was anything rolled on the media after format? I've never had to use REDundead thus far, but I have recovered data from other things in the past. Most issues like that pertain to the data itself or a bad copy. There's a chance that the footage itself had a few written bits after you formatted the mag in this case. Truly hard to know.
 
I am a little confused by Jarred's quote on Facebook saying every VV sensor RED ever produced was stitched.... I know for a fact there was LOTS of discussion on REDuser years prior as to why specifically the Monstro sensor was so expensive to produce over say an 8K helium S35 as it was quoted that "bigger sensors yield less on the wafer during production" and that "you'll likely get 2-2.5X more in spec Helium sensors off the silicon wafer alone compared to Monstro".

Post #646

https://www.reduser.net/forum/red-digital-cinema/recon/181048-komodo/page44#post2739621

Post #52

https://www.reduser.net/forum/red-digital-cinema/recon/181048-komodo/page4#post2733741


If these statements indeed were true almost 2 years ago and the Monstro sensor was indeed "larger" than a S35 helium sensor to produce then it indeed could not be two helium sized sensors or larger stitched together.... Helium sensor is 29.90 x 15.77 (continuous) and Monstro sensor is listed at 40.96 x 21.60 (continuous or conjoined). To make a sensor of this size with two halves both sensors would end up being smaller than a helium sensor and ultimately cheaper to produce in high yield.

Could we get some clarification here as I would like an honest answer as to whether or not Monstro is a continuous sensor despite having to ask Jinnimag to take one apart.


Phil Holland said:
Adding to the difficulty regarding communication across various groups around the internet and multiple posts, I'll put Jarred's reply here from a Facebook post as it is useful:

"This most likely is not actually a sensor stitching issue.. every VV sensor we have made has been stitched. It seems to be an optical issue with the adapters but that analysis is a bit premature, we are still exploring."

The only thing I can add to this is echoing what I've said here on REDUSER and a Facebook reply. It doesn't occur with all lenses, I initially suspected it had something to do with back focus or even specific light sources, but then noticed it wasn't happening on certain lenses at all even. It's very, very hard to test every single variable. I'm fortunate to have a great deal of glass and adapters at my disposal, but I don't have everything when it comes down to it. But it sounds like they are working on it.

It does pertain to optics in some way, shape, or form. Whether that's lenses, adapters, or things within the sensor box; hard to tell. Possible individual assembly I imagine too, but this is all RED engineer work. We'll have to be patient to see what they uncover.

Several people haven't been able to replicate it, which is interesting. But it's too early to know if they have unicorns or just aren't using anything that exhibits the artifact.

I know this has occurred on Arri LF bodies and the general move was to send the camera in to Arri for servicing. As to what happens in that service dungeon, hard to know. Project was wrapped by the time that camera was turned around.
Click to expand...
 
Does anyone have a sample of this happening on a Red Monstro , Arri Mini LF / LF ? Actual image from the camera? I feel like Raptor owners are being a bit gaslit here by saying it happens to all Large Format sensors. Being an owner of multiple Mini LFs and a Monstro and shooting them for about 4,000 hours collectively and not seeing it.

We’re literally dealing with images so the “pics or it didn’t happen” mantra should be mandatory when making claims. Same goes so V-Raptor.

If it is just a handful of cameras affected and it’s easily solvable by warranty work by Red, then it’s just a small bump in the road of an exceptional camera design. If it’s part of the design, there’s no way anyone should be trusting it for real work and that’s a shame.

My thought is that the two sensors aren’t positioned correctly in initial design and fabrication or it’s something as simple as a better calibration at Red that needs to be run.

I will keep everyone updated on service time on this as well. I was already told “it’s normal” which it’s not, so we’ll see what the next step is with Red and how long it takes to fix.
 
I've never seen it on Monstro. Strangely if it would be an electronic problem of "both" sensors it would be an easy fix in color corection, but the artefact is not contiuous in the gamma curve we should apply to fix it... and also warying density depending on colors.
 
Charles Bergquist said:
Does anyone have a sample of this happening on a Red Monstro , Arri Mini LF / LF ? Actual image from the camera? I feel like Raptor owners are being a bit gaslit here by saying it happens to all Large Format sensors. Being an owner of multiple Mini LFs and a Monstro and shooting them for about 4,000 hours collectively and not seeing it.

We're literally dealing with images so the 'pics or it didn't happen' mantra should be mandatory when making claims. Same goes so V-Raptor.
Click to expand...

Agreed... That's kinda what I was hinting at in my last post on the previous page (and then trying to reduce variables to when it happens on Raptor).

Does it happen with both production units and Stormtroopers? (Were STs even made differently/by hand ala Epic-M?) And I'm guessing bringing the camera up to temp (for the black shade to fully kick in) doesn't reduce it at all, right?
 
You must log in or register to reply here.
Back
Top