URSA Mini 4.6k The Original™ Magenta Cast Issue

I don't know about you, but it's a flaw I've never seen actually visible in footage from any other camera. Do you want the camera that requires "fixing" in a way that no others do? I just want a sensor that performs to spec and doesn't have flaws that I'd have to, or worse a client would have to, always grade out. Like I said, the pocket was right next to it under the same circumstances and the grade was the same and the results were different as you can see.


PS - Did you pump up the saturation? Yours looks considerably flatter than my grade.

PPS - Your cat is hilarious.

True, it shouldn't need that kind of fixing. But since it is fixable, maybe it's just the metadata that is incorrect.

I don't think I added any saturation to that particular version, but it's not horrible even when I do.

If Fahnon's camera suffers from irregular magenta corners (which hasn't been fully shown yet) like some cameras have been getting, it is NOT fixable in post. If you can bear to read though all 13 pages of this thread, you will see that there are cameras that exhibit NO issues under any circumstances and cameras that show irregular magenta corners under just about all conditions with or without a lens.

If what you are seeing is non linear falloffs (3D) then just RMA it as BMD have proven they are absolutely able to deliver cameras which have linear falloff.

If you get a camera that has linear falloff, then you probably have a better camera than any other camera you have ever owned...unless of course you happen to have in the cupboard a $70,000 Alexa.

Hi guys,

Need your help here. I can see magenta corners on some of the footage (mostly wider lens, situations with allot of light), but the tests seem to show a camera like many others around here.

Can anyone check these dng's and provide a more informed opinion? The turret-cap test seems fine, but the white wall test does seem to have color issues.

(have contacted BM but, still waiting for a response).

White wall no lens 1: https://www.dropbox.com/s/0mdgy5flzot3u ... 2.dng?dl=0
White wall no lens 2: https://www.dropbox.com/s/lx2llyl19dt9m ... 8.dng?dl=0
turret-cap black image: https://www.dropbox.com/s/21ngbiiqkcu7p ... 9.dng?dl=0

Thanks in advance!

If what you are seeing is non linear falloffs (3D) then just RMA it as BMD have proven they are absolutely able to deliver cameras which have linear falloff.

If you get a camera that has linear falloff, then you probably have a better camera than any other camera you have ever owned...unless of course you happen to have in the cupboard a $70,000 Alexa.

Benton did exactly that and has his new camera back already. We're going to test his alongside mine tomorrow.

If what you are seeing is non linear falloffs (3D) then just RMA it as BMD have proven they are absolutely able to deliver cameras which have linear falloff.

If you get a camera that has linear falloff, then you probably have a better camera than any other camera you have ever owned...unless of course you happen to have in the cupboard a $70,000 Alexa.

Benton did exactly that and has his new camera back already. We're going to test his alongside mine tomorrow.

This will be very interesting to see...good stuff !

Hi guys,

Need your help here. I can see magenta corners on some of the footage (mostly wider lens, situations with allot of light), but the tests seem to show a camera like many others around here.

Can anyone check these dng's and provide a more informed opinion? The turret-cap test seems fine, but the white wall test does seem to have color issues.

(have contacted BM but, still waiting for a response).

White wall no lens 1: https://www.dropbox.com/s/0mdgy5flzot3u ... 2.dng?dl=0
White wall no lens 2: https://www.dropbox.com/s/lx2llyl19dt9m ... 8.dng?dl=0
turret-cap black image: https://www.dropbox.com/s/21ngbiiqkcu7p ... 9.dng?dl=0

Thanks in advance!

Your camera is fine, i balanced your DNGs in Photoshop, its just a uniform cast.
Your corners are darker that's why you see the magenta, you can use davinci to change the tint and the wb

Hi guys,

Need your help here. I can see magenta corners on some of the footage (mostly wider lens, situations with allot of light), but the tests seem to show a camera like many others around here.

Can anyone check these dng's and provide a more informed opinion? The turret-cap test seems fine, but the white wall test does seem to have color issues.

(have contacted BM but, still waiting for a response).

White wall no lens 1: https://www.dropbox.com/s/0mdgy5flzot3u ... 2.dng?dl=0
White wall no lens 2: https://www.dropbox.com/s/lx2llyl19dt9m ... 8.dng?dl=0
turret-cap black image: https://www.dropbox.com/s/21ngbiiqkcu7p ... 9.dng?dl=0

Thanks in advance!

Your camera is fine, i balanced your DNGs in Photoshop, its just a uniform cast.
Your corners are darker that's why you see the magenta, you can use davinci to change the tint and the wb

Ahhhh...might seem like a dumb question but given BMD make the camera and Resolve, can they sorta figure out this stuff so it happens automatically on import ?

So if the color issues are all lens on the UM4.6 and the CRA of the sensor is 0° then one of these lenses should work perfectly:

While the schneider's are known to be a telecentric design, I've not found them to be a lens I love.

Like the Zeiss they tend to have a very bland or neutral look (depending on what you like) and they are very very very heavy. Did I mention how heavy they are ?

JB

So if the color issues are all lens on the UM4.6 and the CRA of the sensor is 0° then one of these lenses should work perfectly:

i have this exact lens and a 75 mm as well just waiting for the PL to come in. When i get the PL i'll be happy to test it

While the schneider's are known to be a telecentric design, I've not found them to be a lens I love.

Like the Zeiss they tend to have a very bland or neutral look (depending on what you like) and they are very very very heavy. Did I mention how heavy they are ?

JB

You did mention you didn't like these in another thread, you also gave me advice on choosing the mount on these since they are heavy. And yes they are very heavy I tend to like neutral looking lenses though. And i really like these because they have less breathing than the cooke's and seem to handle CA better than the Cooke's as well. Sorry to go off topic..ok There's Magenta everywhere ...carry on

While the schneider's are known to be a telecentric design, I've not found them to be a lens I love.

Like the Zeiss they tend to have a very bland or neutral look (depending on what you like) and they are very very very heavy. Did I mention how heavy they are ?

JB

You did mention you didn't like these in another thread, you also gave me advice on choosing the mount on these since they are heavy. And yes they are very heavy I tend to like neutral looking lenses though. And i really like these because they have less breathing than the cooke's and seem to handle CA better than the Cooke's as well. Sorry to go off topic..ok There's Magenta everywhere ...carry on

Let me say that they are MAGNIFICENT lenses. Very well designed and made. They are spiritually like Master Primes really. It's more a question of taste for me. The Cookes are certainly more "flawed" but that's also what a lot of people like

It's not a reflection on the lenses themselves, just what people like. I'm certain they will be a great match for the UM4.6K and will live and be appreciated well beyond this camera.

JB

While the schneider's are known to be a telecentric design, I've not found them to be a lens I love.

Like the Zeiss they tend to have a very bland or neutral look (depending on what you like) and they are very very very heavy. Did I mention how heavy they are ?

JB

You did mention you didn't like these in another thread, you also gave me advice on choosing the mount on these since they are heavy. And yes they are very heavy I tend to like neutral looking lenses though. And i really like these because they have less breathing than the cooke's and seem to handle CA better than the Cooke's as well. Sorry to go off topic..ok There's Magenta everywhere ...carry on

Let me say that they are MAGNIFICENT lenses. Very well designed and made. They are spiritually like Master Primes really. It's more a question of taste for me. The Cookes are certainly more "flawed" but that's also what a lot of people like

It's not a reflection on the lenses themselves, just what people like. I'm certain they will be a great match for the UM4.6K and will live and be appreciated well beyond this camera.

JB

Absolutely I agree. And thank you for the advice.

On a positive note. I have not been able to get my Ursa Mini 4.6k to produce any disturbing magenta, green, or weird sensor artifacts. I have a first run B&H production model. NORMAL (almost imperceptible except for on scopes) blue and red channel elevations that are even across the sensor plane before expectable minor WB and tint adjustments. I have minor corner gradation vignetting at full frame (4.6k) in the far corners and nearly nonexistent vignetting at 4K windowed. No color weirdness with my Sigma at 18 or 35. Clean images with my Zeiss CP.2 Superspeed 35mm/T1.5 and Zeiss EF and ZF.2s. Zeiss and Sigma are easily matched with scopes.

Adam Roberts' lens off, white paper (no ND) window test; white board daylight (overcast day in the shade with lenses) and chart tests all look good. I have not done the turret cap test, though I wonder about its validity.

Clearly, there are some sensors out there that are troubled. Considering the various difficulties I see with RED cameras as a colorist and the, normal, green tint of the Alexa, looks like BMD is doing okay out of the gate with some bumps. However, I am sure if you have a troubled sensor, it does not feel like a bump.

Nevertheless, keep this in mind: I had a bum #2 card slot with a mechanical problem and BMD fixed it and got my camera back to me in 4 days.

...beside the magenta issue, this is what I just found:

WARNING:
We cannot recommend this product for professional work. We reached this decision after repeated failures in the field, experienced both first-hand by LR-folks and second-hand via rental clients. Though it might work much of the time, it is our opinion that this product’s performance is too unpredictable to be trusted on high-value projects.

Its a warning from an equipment rental..
https://www.lensrentals.com/rent/video/ ... ni-4.6k-ef

...beside the magenta issue, this is what I just found:

WARNING:
We cannot recommend this product for professional work. We reached this decision after repeated failures in the field, experienced both first-hand by LR-folks and second-hand via rental clients. Though it might work much of the time, it is our opinion that this product’s performance is too unpredictable to be trusted on high-value projects.

Its a warning from an equipment rental..
https://www.lensrentals.com/rent/video/ ... ni-4.6k-ef

thanks for just finding it, but this has already been discussed at length all week.

"it is our opinion that this product’s performance is too unpredictable to be trusted on high-value projects."

I guess they must know what they're talking about.

JS1.jpg (148.91 KiB) Viewed 8565 times

JS2.jpg (117.44 KiB) Viewed 8565 times

JS1.jpg (148.91 KiB) Viewed 8565 times

And this.... viewtopic.php?f=2&t=46732

You're also about a week late on discussion that company's comments but good work taking this thread OT. (that company have had the same warnings on all BM cameras and also put as a warning that it doesn't come with a lens or a hot shoe.)

JB

...beside the magenta issue, this is what I just found:

WARNING:
We cannot recommend this product for professional work. We reached this decision after repeated failures in the field, experienced both first-hand by LR-folks and second-hand via rental clients. Though it might work much of the time, it is our opinion that this product’s performance is too unpredictable to be trusted on high-value projects.

Its a warning from an equipment rental..
https://www.lensrentals.com/rent/video/ ... ni-4.6k-ef

Hi Andreas,

I am wondering how your UM46 is performing - do you have the a flawed one ?

Hi guys,

Need your help here. I can see magenta corners on some of the footage (mostly wider lens, situations with allot of light), but the tests seem to show a camera like many others around here.

Can anyone check these dng's and provide a more informed opinion? The turret-cap test seems fine, but the white wall test does seem to have color issues.

(have contacted BM but, still waiting for a response).

White wall no lens 1: https://www.dropbox.com/s/0mdgy5flzot3u ... 2.dng?dl=0
White wall no lens 2: https://www.dropbox.com/s/lx2llyl19dt9m ... 8.dng?dl=0
turret-cap black image: https://www.dropbox.com/s/21ngbiiqkcu7p ... 9.dng?dl=0

Thanks in advance!

Your camera is fine, i balanced your DNGs in Photoshop, its just a uniform cast.
Your corners are darker that's why you see the magenta, you can use davinci to change the tint and the wb

It may appear uniform on the white wall test, but on some of the footage, when i white balance the edges of the frame, the center gets a wrogn white balance.

It may appear uniform on the white wall test, but on some of the footage, when i white balance the edges of the frame, the center gets a wrogn white balance.

Yes.. That is/could be an indicator of a non-uniform result. The edges are always (even without a lens) going to produce a fall off. And given the fact that less light (Power) produce a 'diff' color than brighter lit parts of the image, your center will be off if neutralizing the edges... and vice versa..

In most cases the result should be a similar color. So if your wall is white and balance the center for white, the edges should be falling off evenly and produce in case of i.e. an afternoon shot, a slightly more 'orange' result. That is if the Color Matrix of the camera is calibrated as such. The BM camera matrices tend to, instead of producing a orange/blue cast(Temperature), produce a green/magenta cast(Tint). So your edges could also wind up magenta/green if the center is balanced to gray on an EVENLY lit wall...

What should NOT happen is that the left side of the camera when center on evenly lit wall is balanced to neutral, produce a magenta result while the right side produces a green or orange result. The four edges should remain identical. Again the wall NEEDs to be evenly lit. If you light the wall off from center mass - the 4 edges ARE going to produce diff. colors.

[
I guess they must know what they're talking about.

JS1.jpg

JB

JB,

What's the cage/connection lock you are using? And the connection for powering the BMD camera? I face a lot of problems on powering up the camera as the connection tends to get loose with every camera operation!

Gave my UM46K a run yesterday.

Shot with an LED that has a magenta cast so I'm assuming that exacerbated the cast. Good news is it's a cast across the sensor similar to how previous cameras had a green cast. Colour match tool in Resolve helped balance the image though.

Shot RAW 3:1, SLR Magic 50mm APO (no IR CUT)
Camera's a beast though!

Gave my UM46K a run yesterday.

Shot with an LED that has a magenta cast so I'm assuming that exacerbated the cast. Good news is it's a cast across the sensor similar to how previous cameras had a green cast. Colour match tool in Resolve helped balance the image though.

Shot RAW 3:1, SLR Magic 50mm APO (no IR CUT)
Camera's a beast though!

Looking good Olan ...looking Good

Gave my UM46K a run yesterday.

Shot with an LED that has a magenta cast so I'm assuming that exacerbated the cast. Good news is it's a cast across the sensor similar to how previous cameras had a green cast. Colour match tool in Resolve helped balance the image though.

Shot RAW 3:1, SLR Magic 50mm APO (no IR CUT)
Camera's a beast though!

olan, was her shirt really pink, Because it certainly is in the shots here ?

Gave my UM46K a run yesterday.

Shot with an LED that has a magenta cast so I'm assuming that exacerbated the cast. Good news is it's a cast across the sensor similar to how previous cameras had a green cast. Colour match tool in Resolve helped balance the image though.

Shot RAW 3:1, SLR Magic 50mm APO (no IR CUT)
Camera's a beast though!

olan, was her shirt really pink, Because it certainly is in the shots here ?

yeah - it was a pink jumper

[quote="olan_collardy"][/quote]

Great - thx olan

It may appear uniform on the white wall test, but on some of the footage, when i white balance the edges of the frame, the center gets a wrogn white balance.

Yes.. That is/could be an indicator of a non-uniform result. The edges are always (even without a lens) going to produce a fall off. And given the fact that less light (Power) produce a 'diff' color than brighter lit parts of the image, your center will be off if neutralizing the edges... and vice versa..

In most cases the result should be a similar color. So if your wall is white and balance the center for white, the edges should be falling off evenly and produce in case of i.e. an afternoon shot, a slightly more 'orange' result. That is if the Color Matrix of the camera is calibrated as such. The BM camera matrices tend to, instead of producing a orange/blue cast(Temperature), produce a green/magenta cast(Tint). So your edges could also wind up magenta/green if the center is balanced to gray on an EVENLY lit wall...

What should NOT happen is that the left side of the camera when center on evenly lit wall is balanced to neutral, produce a magenta result while the right side produces a green or orange result. The four edges should remain identical. Again the wall NEEDs to be evenly lit. If you light the wall off from center mass - the 4 edges ARE going to produce diff. colors.

Many thanks Morten! the wall was evenly lit, i think that's good news for me (maybe)? No green and orange result, only magenta. But on some shots (skies for ex: the uneven magenta makes it hard to grade)

Many thanks Morten! the wall was evenly lit, i think that's good news for me (maybe)? No green and orange result, only magenta. But on some shots (skies for ex: the uneven magenta makes it hard to grade)

Hi - That could also be a camera matrix issue. I find that with the current camera matrix the camera has severe issues describing any object which are green or magenta by nature. My camera is as said before an RMA case so I cannot judge it properly until I get the replacement.

Check the following 3 images - the first two are from the UM46 in a 'tough' to repro-right situation, for any camera. Magenta/pink on subject mixed with Green(Grass) in direct sunlight. The third is from the Canon 5DMk3. Color Temperature was matched to approximate time/condition in camera.

I loaded up the UM46 image into Resolve and set decode to clip and Rec709. Nothing further.
Image 1 shows severe magenta saturation on my daughters blouse but at the same time the grass in background is matted somehow and not at all coming thru' the way grass is supposed to. (I spent about 5 years making films and shooting fotos for Golf&Country Clubs) so I know what grass should look like

Image 2 - is after turning Color Boost Slider all the way off or negative 100%. The entire image, aside from the blouse goes monochrome.

Image 3 - is the same subject shot with the canon 5dmk3 - nothing done to that image aside from loading it up into Lightroom and grabbing a screenshot off it...

The canon image represent what I am accustomed to be that an iPhone or DSLR. The UM46 in this case looks completely off and would pose a serious challenge to grade right. One would have to desat just the magenta part of the image which basically means increase saturation on green channel only or desaturate red and blues only. A very diff. task indeed.

Recalibrating the Camera Matrix from the Factory (If Possible) would be the right way of doing it.

Mind you, my camera is undergoing RMA due to problems - so it could be that when I get the replacement that all those problems vanish.

magentaSatCB.jpg (590.44 KiB) Viewed 8079 times

magentaSat.jpg (531.82 KiB) Viewed 8079 times

magentaSatCan.jpg (595.91 KiB) Viewed 8079 times

So BM just told my i have only a even magenta cast. But, according to some shots i made, this isn't so. What can i do? I think it's quite obvious by these screenshots that the cast is not even.

So BM just told my i have only a even magenta cast. But, according to some shots i made, this isn't so. What can i do? I think it's quite obvious by these screenshots that the cast is not even.

Can u post a link to a CinemaDNG of the upper image of the Road ?

So BM just told my i have only a even magenta cast. But, according to some shots i made, this isn't so. What can i do? I think it's quite obvious by these screenshots that the cast is not even.

Can u post a link to a CinemaDNG of the upper image of the Road ?

I don’t have dng’s from those previous screenshots (shot in prores) but heres is a sample from that prores: https://www.dropbox.com/s/a0xjhpfoug6rw ... 4.mov?dl=0

I don’t have dng’s from those previous screenshots (shot in prores) but heres is a sample from that prores: https://www.dropbox.com/s/a0xjhpfoug6rw ... 4.mov?dl=0

Nuno,

a quick n dirty...

1 = Original
2 - Neutralized
3 - Quick Mood

org.jpg (368.88 KiB) Viewed 8038 times

neutralized.jpg (431.34 KiB) Viewed 8038 times

mood.jpg (650.67 KiB) Viewed 8038 times

Nuno,

I am NOT saying that nothing is wrong with your camera - it is up to you and BMD to decide.

I AM saying that I would NOT have been able to achieve those results with my camera which IS broken !

Nuno,

I am NOT saying that nothing is wrong with your camera - it is up to you and BMD to decide.

I AM saying that I would NOT have been able to achieve those results with my camera which IS broken !

Morten, you've been a great help! Can you tell me how you neutralized magenta (2)?

Really, thanks!

Does the new color science in 3.2 fix the magenta issues


Sent from my iPhone using Tapatalk

Does the new color science in 3.2 fix the magenta issues


Sent from my iPhone using Tapatalk

If it's a new color science it is quite possible it's a fix without actually saying it. Download and give it a try.
Post back with your results

Tested ours today and it had the uneven magenta corners... top right was much worse than the other corners. Our reseller sent it back to Holdan today for testing.

Tested ours today and it had the uneven magenta corners... top right was much worse than the other corners. Our reseller sent it back to Holdan today for testing.

Neil did you download the new update 3.2?

It does say new color science. I'm going to do a quick test soon hopefully


Sent from my iPhone using Tapatalk

Tested ours today and it had the uneven magenta corners... top right was much worse than the other corners. Our reseller sent it back to Holdan today for testing.

Neil did you download the new update 3.2?

Nope I didn't but I doubt the problem is fixable via software...

Tested ours today and it had the uneven magenta corners... top right was much worse than the other corners. Our reseller sent it back to Holdan today for testing.

Neil did you download the new update 3.2?

Nope I didn't but I doubt the problem is fixable via software...

They did say it's a new color science ..so hopefully soon someone with a 4.6k that has seen the magenta issue download update 3.2 will let us know if it's still there or if its gone. Maybe when Morton's camera come back from RMA he'll let us know if it came with the 3.2 update already installed, and also if the issue has been resolved.

They did say it's a new color science ..so hopefully soon someone with a 4.6k that has seen the magenta issue download update 3.2 will let us know if it's still there or if its gone. Maybe when Morton's camera come back from RMA he'll let us know if it came with the 3.2 update already installed, and also if the issue has been resolved.

I'm not that technically minded, but it seems everyone's magenta issues are slightly different. Some like myself have a more prominent cast in the top right, whilst others have a more even cast. I'm guessing the people with an even cast could be fixed with an update, but how do you fix an issue that is isolated like mine?

Again I may be being a little silly, but they'd have to make a specific adjustment for specific people would then not?

Tested out 3.2 right now.
No fix.

Tried with 12 lenses, as soon as they reach f8.0, they start giving this:

Hmmmm. Sorry It didn't resolve the issue. So I wonder what is the look of this new color science.

I might misunderstand but it was to my understanding the new color science is for the 4k to better match the 4.6k in multi camera situations...


A

Sent from my Nexus 5X using Tapatalk

They did say it's a new color science ..so hopefully soon someone with a 4.6k that has seen the magenta issue download update 3.2 will let us know if it's still there or if its gone. Maybe when Morton's camera come back from RMA he'll let us know if it came with the 3.2 update already installed, and also if the issue has been resolved.

I'm not that technically minded, but it seems everyone's magenta issues are slightly different. Some like myself have a more prominent cast in the top right, whilst others have a more even cast. I'm guessing the people with an even cast could be fixed with an update, but how do you fix an issue that is isolated like mine?

Again I may be being a little silly, but they'd have to make a specific adjustment for specific people would then not?

To be honest Neil I don't know how that will work. Some people exhibit the cast more than others. Some can globally elimate it and some can't. Then there are those that don't get the magenta issue at all. So far I see about 6 people on the forum posting about the issue. I wonder if some of our friends here were unlucky and just recieved bad units. Hopefully that is the issue and it's not something that is a characteristic of this particular sensor. It would be great to just RMA the unit and get one without issues. Only time will tell how extensive this issue may be, or if it's just kinks of a first production run of a brand new camera.

I might misunderstand but it was to my understanding the new color science is for the 4k to better match the 4.6k in multi camera situations...


A

Sent from my Nexus 5X using Tapatalk

It says:
Added Version 3 BMDFilm 4K and BMDFilm 4.6K color science

That sounds to me like a new color science for the 4.6k

I might misunderstand but it was to my understanding the new color science is for the 4k to better match the 4.6k in multi camera situations...


A

Sent from my Nexus 5X using Tapatalk

It says:
Added Version 3 BMDFilm 4K and BMDFilm 4.6K color science

That sounds to me like a new color science for the 4.6k

In the mail from Grant:

This update includes new color science for both URSA Mini models so the 4K sensor models now have a similar look to the 4.6K sensor models. This means the two different sensors are now a much better match, which is good if you use both cameras on the same shoot, or a live production situation. DaVinci Resolve will read the metadata from the URSA Mini 4K automatically and select the correct color science.

--

"Both models" as EF/PL or 4k/4.6k?

Some lines above he talks EF/PL so for me it would be natural to think it's for these two..

Maybe not

A

Sent from my Nexus 5X using Tapatalk

I might misunderstand but it was to my understanding the new color science is for the 4k to better match the 4.6k in multi camera situations...


A

Sent from my Nexus 5X using Tapatalk

It says:
Added Version 3 BMDFilm 4K and BMDFilm 4.6K color science

That sounds to me like a new color science for the 4.6k

In the mail from Grant:

This update includes new color science for both URSA Mini models so the 4K sensor models now have a similar look to the 4.6K sensor models. This means the two different sensors are now a much better match, which is good if you use both cameras on the same shoot, or a live production situation. DaVinci Resolve will read the metadata from the URSA Mini 4K automatically and select the correct color science.

--

"Both models" as EF/PL or 4k/4.6k?

Some lines above he talks EF/PL so for me it would be natural to think it's for these two..

Maybe not

Hopefully the moderators can clear that up for us
Either way it's great for the 4K users.

Ah - added magenta cast to the 4K:)

I see there's a tint slider in the menu for the upcoming new OS software for the 4.6k. Would that help with the magenta issue?