URSA Mini 4.6k The Original™ Magenta Cast Issue

Without seeing any images its hard to gain much perspective from it. I seriously doubt those clips are as green as the defective 4.6K sensor images are magenta. There is no way to actually know until we can see the Alexa clips and compare them.

I cut (and do some coloring) on Alexa footage very often. Without adjusting the tint in camera, it does put out a very green image. I typically set the tint to -2 in the camera, and it gets rid of it for the most part. But even then, in scenes with high-kelvin sources, the green still comes out stronger.

I cut (and do some coloring) on Alexa footage very often. Without adjusting the tint in camera, it does put out a very green image. I typically set the tint to -2 in the camera, and it gets rid of it for the most part. But even then, in scenes with high-kelvin sources, the green still comes out stronger.

If you have some Alexa clips that look green (and you can confirm they were shot without an ND filter added), please share them for comparison. Thanks.

Can we get back to the matter at hand?
I couldn't care less if Alexa produces green clips.

My 4 year old BMCC by BlackMagic Design produces very neutral images...

My 4 year old BMCC by BlackMagic Design produces very neutral images...

That's surprising to me. Mine is on the green side for sure and I've never seen any RAW or ProRes files posted that didn't have a green tint either. Can you post a RAW file?

My BMPCC also produces very neutral images. And it's just a 1K$ camera.

[Joshua Helling]

Hey All,

Just checking back in. The build is still going through testing and I'm told it is looking positive.

No update on when the update will be public.

Just wanted to drop in and let you guys know we haven't forgotten and are still hard at work. I am looking forward to being able to give you a more substantial update soon!

Awesome thank you J

Thanks Joshua.

I sadly sent my 4.6k back for a refund. If this magenta issue gets sorted out I'll gladly re-order the camera.

Hey All,

Just checking back in. The build is still going through testing and I'm told it is looking positive.

No update on when the update will be public.

Just wanted to drop in and let you guys know we haven't forgotten and are still hard at work. I am looking forward to being able to give you a more substantial update soon!

Thanks for the communication Joshua. it is much appreciated

Hey All,

Just checking back in. The build is still going through testing and I'm told it is looking positive.

No update on when the update will be public.

Just wanted to drop in and let you guys know we haven't forgotten and are still hard at work. I am looking forward to being able to give you a more substantial update soon!

This is all it takes

Joshua, after long years in the BM desert over the years thru various firmware fixes that took months to arrive, this is SO welcome.

Thank you SO much for even just that simple message. It really makes a difference.

Thank you!

Hey All,

Just checking back in. The build is still going through testing and I'm told it is looking positive.

No update on when the update will be public.

Just wanted to drop in and let you guys know we haven't forgotten and are still hard at work. I am looking forward to being able to give you a more substantial update soon!

Thanks for the update Joshua! Any word on wether the magenta fix will be something operating in the background or user set and adjustable?

Just wanted to say thank you to Benton for starting this thread and being a gentleman throughout. Kudos

Just wanted to say thank you to Benton for starting this thread and being a gentleman throughout. Kudos

Thank you Donnell and you're very welcome! It's quite a ride to say the least! I just wanted (and still want) a good working camera. Hopefully, that goal is closer at hand and if this thread has helped toward that end for everyone else, I couldn't be happier!

I believe it's close. I think after the fix this camera will be their best selling camera for a while. There's a lot of people on the fence waiting to see how this plays out, myself included. I have all the accessories for the mini just sitting in my closet. Once everything is fixed you won't be able to keep these 4.6ks on the shelf.

Thank you very much Josh for the update!

Please please let us know when everything is finally fixed what was it that caused the problem (?)
I think on a pure educational perspective such an explanation would be very welcome.

And Indeed Benton you sir are a gentleman.

Finally able to begin testing the 50mm APO and with the Mini 4.6K PL mount. Preliminary tests in raw at all apertures from T2.1 through T22 look good. No hint of a severe fall off or magenta corners or uneven colouration. Okay, so I got to first base with my first lens, but the 25mm and 32mm APO primes will be the real test. That won't happen for half a year or more.


Sent from my iPad using Tapatalk

That's awesome Rick ..by the time you get those primes , they'll have a fix for the issue. Can you post a frame of what your new 50mm prime looks like on the 4.6k?

Do you mean a DNG file or a normal graphic image?


Sent from my iPad using Tapatalk

Just a normal jpeg. I wanted to see the look of that APO prime on the 4.6k ..also Rick let's see a pic of that baby on your rig. That prime looks gorgeous

Good news Rick! Here is keeping our fingers crossed no other issues crop up. Sounds like you got one of the "good copies". I think, as others have reported, part of the "uneven looking" or co,or as tissues is somewhat lens related too. Seems the results are better with the more expensive purpose built Cine lenses, compared to the modified "Still Photography" DSLR lenses. But, it is enough of an issue anyway, that BM is checking it out.

Good news Rick! Here is keeping our fingers crossed no other issues crop up. Sounds like you got one of the "good copies". I think, as others have reported, part of the "uneven looking" or co,or as tissues is somewhat lens related too. Seems the results are better with the more expensive purpose built Cine lenses, compared to the modified "Still Photography" DSLR lenses. But, it is enough of an issue anyway, that BM is checking it out.

I've noticed that when the more expensive cine glass is used you don't seem to get the issue at all. I have Schnidar xenar 3's PL. They're telecentric. If I had a mini 4.6k to test. I don't think i would have the magenta issue.

I think Donnell is right, if the lens is telecentric, then I dont think youll see the problem at all.

I think Donnell is right, if the lens is telecentric, then I dont think youll see the problem at all.

I mean no offense here, but I have to disagree with that statement. The 4.6K camera I was shipped had sensor issues regardless of the lenses and aperture used. Even without a lens attached the images that sensor produced showed a pronounced uneven magenta cast that was heaviest on the right side. As far as I know, no one has empirically proven the telecentric lens theory. Have you seen a test with one of the defective 4.6K cameras where the issue disappeared when a telecentric lens was put on the same camera body? If you have images from a test like that, please post them. Without that, the assertion that telecentric lenses make any difference at all is simply speculation.

Jamie, for sure. If I was seeing what you have described, then i would completely agree with you.

What Ive seen, from about 20 or 30 cameras worth of DNGs now, is clean images from the no lens paper test, and clean images at, say f/2.8 and focal lengths like 18mm or 135mm.

I wonder if yours is afflicted differently.

Either way, I completely agree with your conclusions given the data youre seeing from your cam, but its not the same as what Im seeing overall? I cant draw a conclusion from that, but it certainly raises questions!

If you're curious to see for yourself, here's a link to an original DNG and a couple of stills from Resolve.

https://drive.google.com/folderview?id=0B1bnfGXKHqpcZzVpeldIZk8zbjA&usp=sharing

This is with a Tokina 11-16 set to 11mm and f5.6 (hardly an abnormal aperture). In the screenshot from Resolve you'll see that I had to pull the tint control in the raw tab all the way down to -32 to balance the white card on the left side of the image while the right side of the image still holds a magenta cast. There is also a comparison where the left and right sides are pasted next to each other. I can't speak for other 4.6K cameras, but on the unit I was shipped the magenta cast was far from minor and was present in all combinations of f-stop, lighting, and lenses that I tested.

I think Donnell is right, if the lens is telecentric, then I dont think youll see the problem at all.

I mean no offense here, but I have to disagree with that statement. The 4.6K camera I was shipped had sensor issues regardless of the lenses and aperture used. Even without a lens attached the images that sensor produced showed a pronounced uneven magenta cast that was heaviest on the right side. As far as I know, no one has empirically proven the telecentric lens theory. Have you seen a test with one of the defective 4.6K cameras where the issue disappeared when a telecentric lens was put on the same camera body? If you have images from a test like that, please post them. Without that, the assertion that telecentric lenses make any difference at all is simply speculation.

Jamie what I stated is speculation on my part. I'm making that assumption based on JB's footage as well as some others I saw on Vimeo that used high end cine glass did not show the issue at all. I also agree with Kyle that some cameras including yours may exhibit the issue way more than others. Like you said without proper tests, it's just me speculating. But it is also possible that my point on high end cine glass used on most cameras that exhibit this issue moderately may not show any issues at all.

I can confirm that I saw the magenta issue on my Mini 4.6 PL mount with a Zeiss CP.2 35mm super speed at f/8 and above. Below f/8, no magenta. 50mm and 85mm (also CP.2 SS) no magenta.

hey Ivon thanks.. I had the Cp.2s as well. From what i understand they are Still photography glass re-housed, in that new body. Does it make a difference? I don't know. Is there a difference when using purposely built cine glass, and does coating or non coating on the glass make any difference in contributing to/ or eliminating the problem?

Just a normal jpeg. I wanted to see the look of that APO prime on the 4.6k ..also Rick let's see a pic of that baby on your rig. That prime looks gorgeous

To get you a few sample pics, I recorded six brief raw 4608x2592 clips shot in the shadier parts of the yard (T22 since I am waiting for my Heliopan step-up ring to be able to use the SLR Magic Vari-ND). Unfortunately I had dual card recording set ON.

Resolve 12.5 is doing something crazy. I expected six folders on each CFast 2 card and that's what I saw when I selected the card in Media Manager, but when I move those clips into my media pool, Resolve has created hundreds of folders/shots, such as C0010_S001, C0010_S002, each with hundreds of DNG files. Taking forever to complete the task.


Sent from my iPad using Tapatalk

Rick try putting all those clips from both cards in 1 folder on your desktop.. Then when you go into resolve look for that folder on your desktop and open it. Your clips should be in sequence and should play fine.

Hey guys, ok so I'm back and with more testing.

Today we just got a new lens, a Canon 55-250 EF f4.6. I'm considering returning the lens due to the fact that this lens has, in today's tests at least it definitely showed some slight vignetting. and IMO that caused a lot of the problems with the magenta cast. Interesting to notice however, is the fact that where the lens tended to vignette, the magenta was far stronger. I've also used a Nikkor 55-300 which showed zero magenta on any focal length and aperture, so in this case I think it's a lens problem.

In any case, I would really appreciate feedback on what to do in this situation. Should I:

1) Return the glass if further testing shows magenta still, or
2) Keep the glass, and wait until the magenta software update in hopes that the magenta issue will be fixed.

Footage:

Password: magenta
All shots @ 800 ISO
ProRes HQ
Full sensor

I'm leaning more towards the first option as I have no idea whether or not the issue will be fixed with a software/firmware update. If the issue persists past the update, then I'm definitely dealing with BM directly as my camera is not fully operational, even though I've gotten some clean shots. I'll post those as soon as I can.

Thanks guys.

Hey guys, ok so I'm back and with more testing.

Today we just got a new lens, a Canon 55-250 EF f4.6. I'm considering returning the lens due to the fact that this lens has, in today's tests at least it definitely showed some slight vignetting. and IMO that caused a lot of the problems with the magenta cast. Interesting to notice however, is the fact that where the lens tended to vignette, the magenta was far stronger. I've also used a Nikkor 55-300 which showed zero magenta on any focal length and aperture, so in this case I think it's a lens problem.

In any case, I would really appreciate feedback on what to do in this situation. Should I:

1) Return the glass if further testing shows magenta still, or
2) Keep the glass, and wait until the magenta software update in hopes that the magenta issue will be fixed.

Footage:

Password: magenta
All shots @ 800 ISO

I'm leaning more towards the first option as I have no idea whether or not the issue will be fixed with a software/firmware update. If the issue persists past the update, then I'm definitely dealing with BM directly as my camera is not fully operational, even though I've gotten some clean shots. I'll post those as soon as I can.

Thanks guys.

To add to this, I would also like to ask you guys which of the footage do you consider acceptable/passable. I think that the wide open shots (f/4.6) are definitely the most useful, with the f/22 shots having some usability, especially when zoomed in, and the f/14 shots being completely DOA due to the stronger magenta.

Again, thanks for any and all feedback guys.

Canon APS-C sensors feature a 1.6x crop factor. Nikon and Sony feature a 1.5x crop.

Isn't the BM Ursa Mini sensor larger than 1.6x?

If the BM Ursa Mini sensor is larger than 1.6x that is more than likely why the Canon 1.6x 55 - 250 lens is vignetting.

I think the 4.6k is a 1.4 crop which is why that sensor would have a vignette.

I wouldn't say the purple is a lens problem, but rather an interaction between this lens and the camera being a problem. I doubt you get purple shading with that lens on a Canon camera.

Rick try putting all those clips from both cards in 1 folder on your desktop.. Then when you go into resolve look for that folder on your desktop and open it. Your clips should be in sequence and should play fine.

Donnell, I was able to reconstruct the six folders with all the proper content in my Capture folder on the Pegasus RAID. Then I deleted the hundreds of messed up redundant folders in the Capture directory. In Resolve, I deleted all of the messed up clips from the Media Pool etc. Then I added the six correct clips from the Capture folder to my Media Pool and created a new timeline. Problem solved. Thanks.


Sent from my iPad using Tapatalk

Awesome Rick.

[quote=“Donnell Henry"]Awesome Rick.[/quote]

Here’s the first raw frame shot yesterday which looks fine at full resolution, reduced greatly in size to HD to post here, but shot on the Mini with the 50mm APO at T22 with no grading:

Deer Me C0016_000001.png (851.33 KiB) Viewed 10177 times

Wowzers Rick that 50mm looks good and sharp. Kinda reminds me of Ziess. It's a lens I can get behind for sure. Thank you for posting. That deer is ready for its close-up.

Wowzers Rick that 50mm looks good and sharp. Kinda reminds me of Ziess. Thank you for posting.

Thanks, Donnell. That’s on my shoulder in a hurry to set aperture and guess at focus to get the shot of the deer who had been in our front yard when I fetched the camera and followed it across the road. Looking forward to seeing the 50mm APO perform on a tripod with time to prepare.

Wowzers Rick that 50mm looks good and sharp. Kinda reminds me of Ziess. Thank you for posting.

Thanks, Donnell. That’s on my shoulder in a hurry to set aperture and guess at focus to get the shot of the deer who had been in our front yard when I fetched the camera and followed it across the road. Looking forward to seeing the 50mm APO perform on a tripod with time to prepare.

SLR Has a winner on their hands with this 50mm. Can't wait to see what you do with it graded in a controlled environment. I'm sure judging from the jpeg it's going to look really good.

Just want to let folks know that I was able to successfully return my 4.6K (which had strong magenta vignetting) to B&H for a full refund. I did so within 30 days of receiving it. I look forward to this all getting sorted out so I can re-buy this awesome camera.

In the meantime, I'll continue using my URSA Mini 4K happily (albeit at ISO400 only )

Here’s the first raw frame shot yesterday which looks fine at full resolution, reduced greatly in size to HD to post here, but shot on the Mini with the 50mm APO at T22 with no grading:

Deer Me C0016_000001.png

I really really hate to be the one to say it Rick, but I'm clearly seeing magenta on the lower right side and bottom corner. The center of the driveway looks much more neutral than the bottom right corner. Maybe at wider apertures, it will disappear.

Next week I'll be able to use my SLR Magic Vari-ND. I was at T22 to get a good exposure but I don't want to shoot like that normally. We shall see what you think then.


Sent from my iPhone using Tapatalk

Next week I'll be able to use my SLR Magic Vari-ND. I was at T22 to get a good exposure but I don't want to shoot like that normally. We shall see what you think then.


Sent from my iPhone using Tapatalk

Have you considered using a narrow shutter angle to get much closer to an ideal stop while you wait for the SLR Magic ND? Since these are tests examining frames, you don't need to be concerned with the stuttered look of a narrow angle that would otherwise look horrible. I'm keeping my fingers crossed that the new firmware will take care of this whole magenta thing! I'm really looking forward to when this thread becomes a curious relic buried deep in this forum.

Next week I'll be able to use my SLR Magic Vari-ND. I was at T22 to get a good exposure but I don't want to shoot like that normally. We shall see what you think then.


Sent from my iPhone using Tapatalk

Have you considered using a narrow shutter angle to get much closer to an ideal stop while you wait for the SLR Magic ND? Since these are tests examining frames, you don't need to be concerned with the stuttered look of a narrow angle that would otherwise look horrible. I'm keeping my fingers crossed that the new firmware will take care of this whole magenta thing! I'm really looking forward to when this thread becomes a curious relic buried deep in this forum.

+1

RMA Update

It has been 6 Weeks since German Distributor sent it to BMD UK.

1) The UK Office has confirmed the Magenta Issue as being an incompatibility with certain lenses.
And that they will not be replacing my camera as it is 'OK' but my lenses are incompatible. They suggest I take the camera back and shoot at apertures above 8.0 until they can gather enough information on which lenses are incompatible and solve that with a firmware fix. I have tested with 4 different Full Frame Canon Mount Lenses. 1x Zeiss 50mm, 1x Zeiss 100m, 1x Canon 70-200mm, 1x Nikon 70-300. All exhibiting Magenta.

2) Apparently the UK Office never actually looked at my camera and just classified it in the above mentioned category. Utterly failing to recognize that aside from above, it suffers from a faulty sensor with at least uneven current AND a broken View Finder. Both of which they suggest I take back.

So much for sending it to BMD for an RMA....

I have no idea how to react to such a statement and I find this situation highly absurd. I dont think I have ever experienced worse support that in this case. Twice I called the UK Office directly and in both cases they stated that the Support Manager is looking into the case and checking the camera. This proved to be false.

So much misinformation - so much waiting and now this...

I think I will void the deal and take my 8k EURO add another 8k and invest them into a RED. But at least then I know that I have support and a camera which can shoot below aperture 8 and deliver natural looking non-magenta images.

What a sad sad scenario !