URSA Mini 4.6k The Original™ Magenta Cast Issue

As frustration as this is, I'm sure BM are working to identify the issue and produce a fix.

My camera does not appear to have the un-even cast issue but to be sure I ran some additional tests this morning and thought it worth sharing as it shows that not all cameras have the un-even cast issue. From looking at all the footage and DNGs that have been posted it seem that the un-even cast shows up in skies. So I've shot clear blue skies in my tests.

There is a very slight magenta shift the the corners but nothing as extreme as some have shown and I would not consider this an issue in "real world" footage.

Shot early morning in the UK on a rare day with a beautiful clear skies. Sun rising from the left of frame and red brick buildings on the right of frame.

These were shot on the BM URSA 4.6K, running firmware 3.2. Recorded in 4.6K RAW.

Exposure was controlled with shutter angle. Shot with and without Hoya IR Cut filter.

Edited and Graded in a 1080p Timeline in DaVinci Resolve Studio 12.5 Beta.

Colour balance was done in the RAW panel and then an s-cuved and saturation curve was applied in the grade.

Lenses:
Sigma 18-35mm f/1.8
Sigma 50mm f/1.4

Filters:
Hoya UV / IR Cut

DNGs: https://www.dropbox.com/s/6gvxxhns9y6df ... t.zip?dl=0

Shot early morning in the UK on a rare day with a beautiful clear skies. Sun rising from the left of frame and red brick buildings on the right of frame.

If you get the opportunity to perform that test again, it would be nice if you got a shot rotated 180° to see if any perceived cast followed the sensor or was simply an artifact of the location.

Likewise, those shooting blank walls or paper might want to take a shot with the camera rotated 180° as well.

If you managed to keep the center the same in both shots you could rotate one image 180° and subtract out the background to leave mostly sensor artifacts.

Shot early morning in the UK on a rare day with a beautiful clear skies. Sun rising from the left of frame and red brick buildings on the right of frame.

If you get the opportunity to perform that test again, it would be nice if you got a shot in the opposite direction to see if any perceived cast followed the sensor or was simply an artifact of the location.

Likewise, those shooting blank walls or paper might want to take a shot with the camera upside down as well.

If you managed to keep the center the same in both shots you could subtract out the background to leave mostly sensor artifacts.

I did shoot both directions with the Sigma 18-35 (at 35mm) tho not exactly 180º for each other as the are obstacles like trees and building in my back yard. Shot about 30min apart.

Here is a DNG: https://www.dropbox.com/s/wtqceh74h6aoi ... 5.dng?dl=0

I did shoot both directions with the Sigma 18-35 (at 35mm) tho not exactly 180º for each other as the are obstacles like trees and building in my back yard. Shot about 30min apart.

Sorry, I meant to say rotated 180°. Think of it as a dark/white field subtraction as would be performed in astrophotography.

I did shoot both directions with the Sigma 18-35 (at 35mm) tho not exactly 180º for each other as the are obstacles like trees and building in my back yard. Shot about 30min apart.

Sorry, I meant to say rotated 180°. Think of it as a dark/white field subtraction as would be performed in astrophotography.

As in the camera pointing in the same direction (north) but turning the camera upside down?

Direction is not as important as rotation around the center of the scene 180°. It's not as fussy if the scene is a uniform gradient. When shooting the sky just point straight up, shoot, rotate around vertical axis and shoot again.

For less vertical shots you could mount a vidio head on a video head to achieve the same effect at other angles.

If you had perfect optical conditions, theoretically you could shoot a busy scene and if rotation was on the optical axis obtain complete subtraction except for sensor artifacts.

If I were designing a camera I would rig the camera during design so i could rotate the sensor to leave all other things like baffle and mount the same. That way i could isolate where image artifacts are coming from.

Hi Adam,
I see magenta corners on second clip very noticeable when used with IR cut filter.
Anyway, I had similar results like you when shooting at f5.6 with the same lens.
Try repeating the same test with lens stopped above f8.0, you should see way more magenta at corners (and sides)

I saw it too.
thanks cos I thought I was seeing things

it is there just like on my camera
purple haze comes to mind

I hope we get something from BMD soon, something official.
I think that'd leave many people happy...
I'm sure they won't let us down though.

I started the RMA process today.
Awaiting response.

adamroberts, can you possibly repeat the test with closed aperture like f16 or f22 ? The magenta seems to be more visible when stopped down.
Either way, it seems a IR filter would greatly improve the color of UM46.

My camera does not appear to have the un-even cast issue but to be sure I ran some additional tests this morning and thought it worth sharing as it shows that not all cameras have the un-even cast issue.

I felt the same way until I tested with three 35mm lenses stopped down all the way, in various conditions. This seems to produced the most dramatic magenta at the edges, plus strong green in the center.

I have now returned my second 4.6k.

Well, I started the RMA process today... Camera is going back to B&H. Weird thing is B&H said they had NEVER heard of this issue and weren't aware of getting any of these cameras returned for that purpose. Seemed really weird. Im guessing this customer service person either had no idea what they were talking about and were genuinely unaware or they were being purposely dishonest.

Either way, now I have to get my camera situation figured out all over again...

It's certainly valid to stop a lens down to T16 or T22 doing these tests, but in actual shooting I would think often people would not shoot slower than T5.6 or T8 to avoid diffraction, so it may be more important to note how the camera performs in the range of apertures you are likely to use.


Sent from my iPad using Tapatalk

It's certainly valid to stop a lens down to T16 or T22 doing these tests, but in actual shooting I would think often people would not shoot slower than T5.6 or T8 to avoid diffraction, so it may be more important to note how the camera performs in the range of apertures you are likely to use.

In controlled situations, I agree with you Rick. However, we are not always shooting controlled situations. When shooting verite documentary you need to get the shot, I think there are valid cases when you're forced to trade off a little diffraction to make sure you get an exposure. 4.6K really should just work at all apertures.

It's certainly valid to stop a lens down to T16 or T22 doing these tests, but in actual shooting I would think often people would not shoot slower than T5.6 or T8 to avoid diffraction, so it may be more important to note how the camera performs in the range of apertures you are likely to use.

f8-11 are uncommon?
Also, in a camera with no built-in ND filter and 800 base ISO it's pretty hard to advertize it for documentary when you can't shot slower than T5.6, imho.

Could someone take a different camera (Canon, Sony, Red, whatever) and do the same gray card test (white balance the middle, increase saturation, check for edge and corner casts)? Is there anything similar happening with those? Personally I haven't seen tests like these with other cameras, so maybe this is something that happens to all sensors to a degree.

Could someone take a different camera (Canon, Sony, Red, whatever) and do the same gray card test (white balance the middle, increase saturation, check for edge and corner casts)? Is there anything similar happening with those? Personally I haven't seen tests like these with other cameras, so maybe this is something that happens to all sensors to a degree.

In this test between two UM4.6k's, we also shot with a Canon 6D and BM Pocket Camera side by side. Both UM4.6k's produced a magenta cast, but neither the Canon nor the Pocket Camera did using the exact same lenses. viewtopic.php?f=2&t=46898&hilit=hunting

Could someone take a different camera (Canon, Sony, Red, whatever) and do the same gray card test (white balance the middle, increase saturation, check for edge and corner casts)? Is there anything similar happening with those? Personally I haven't seen tests like these with other cameras, so maybe this is something that happens to all sensors to a degree.

In this test between two UM4.6k's, we also shot with a Canon 6D and BM Pocket Camera side by side. Both UM4.6k's produced a magenta cast, but neither the Canon nor the Pocket Camera did using the exact same lenses. viewtopic.php?f=2&t=46898&hilit=hunting

Yes, I saw that. I'm just saying to a degree. There are some gray card tests with the 4.6K that only show the magenta when pushing saturation to extreme highs. I'm curious if that's common with all sensors or if it only happens with the 4.6K. Obviously there are some 4.6K's that exhibit the magenta corners much worse, but also some don't. If it is something that happens with other sensors/cameras, then there are in fact a bunch of good 4.6K cameras out there. If this is something entirely unique to the 4.6k then there's potentially a larger issue than just a few bad apples.

Jamie and Alessandro, no argument there are situations where you may want or need to go beyond T8 all the way to T22. I was thinking more of controlled narrative. And I agree the image should not suffer from magenta corners going to a smaller aperture.


Sent from my iPad using Tapatalk

Could someone take a different camera (Canon, Sony, Red, whatever) and do the same gray card test (white balance the middle, increase saturation, check for edge and corner casts)? Is there anything similar happening with those? Personally I haven't seen tests like these with other cameras, so maybe this is something that happens to all sensors to a degree.

In this test between two UM4.6k's, we also shot with a Canon 6D and BM Pocket Camera side by side. Both UM4.6k's produced a magenta cast, but neither the Canon nor the Pocket Camera did using the exact same lenses. viewtopic.php?f=2&t=46898&hilit=hunting

Yes, I saw that. I'm just saying to a degree. There are some gray card tests with the 4.6K that only show the magenta when pushing saturation to extreme highs. I'm curious if that's common with all sensors or if it only happens with the 4.6K. Obviously there are some 4.6K's that exhibit the magenta corners much worse, but also some don't. If it is something that happens with other sensors/cameras, then there are in fact a bunch of good 4.6K cameras out there. If this is something entirely unique to the 4.6k then there's potentially a larger issue than just a few bad apples.

Gotcha. None of my processed frames had the saturation pumped up. I think Fahnon may have added a little bit, but nothing extreme. You can download all the original DNG and RAW files and play with them if you'd like. I believe Fahnon, did pump up the Canon 6D to try and make it show magenta and it did not. This is not to say it won't start looking ugly, but it didn't push magenta more than any other color. But as a whole and admitted by BMD, the camera does have an overall magenta bias or cast, but that can be easily dialed out with white balance.

Gotcha. None of my processed frames had the saturation pumped up. I think Fahnon may have added a little bit, but nothing extreme. You can download all the original DNG and RAW files and play with them if you'd like. I believe Fahnon, did pump up the Canon 6D to try and make it show magenta and it did not. This is not to say it won't start looking ugly, but it didn't push magenta more than any other color. But as a whole and admitted by BMD, the camera does have an overall magenta bias or cast, but that can be easily dialed out with white balance.

I know. That's not what I'm asking. I'm wondering if other cameras also have something similar to "magenta corners" (maybe green on Alexa and Red), but just not as severe as some of these 4.6K's. It's clear that other manufacturers like Red and Arri have better QC. They may also see sensors with similar issues that they don't allow to go into their cameras, but maybe allow some up to a certain extent. This "magenta corners" issue could be a common characteristic with all sensors, but is showing more severely on some 4.6K sensors, like yours and not as severely (and sometimes not noticeable under normal circumstances) on many others. That's what I'm asking.

This makes me think of part of the reason why Red made the Scarlet MX. They had a bunch of sensors that didn't meet their criteria and QC for the Epic MX 5K, but were fine at 4K, so they came out with the Scarlet MX which could only shoot up to 4K. Could be a very similar situation.

Gotcha. None of my processed frames had the saturation pumped up. I think Fahnon may have added a little bit, but nothing extreme. You can download all the original DNG and RAW files and play with them if you'd like. I believe Fahnon, did pump up the Canon 6D to try and make it show magenta and it did not. This is not to say it won't start looking ugly, but it didn't push magenta more than any other color. But as a whole and admitted by BMD, the camera does have an overall magenta bias or cast, but that can be easily dialed out with white balance.

I know. That's not what I'm asking. I'm wondering if other cameras also have something similar to "magenta corners"

I thought a Canon 6D and the BM Pocket camera counted as other cameras. But if you specifically meant cameras in the class of REDs and Arri's I don't know. But I believe at least the RED's have had there share of issues, but I don't think it was anything like the magenta corners on the UM4.6.

I have an opinion that some may or may not find interesting or even possible,
I find it odd that the magenta is strongest in the corners and maybe this is due to the mounting/adhesive
they used for the sensors? I think it could possibly be drawing heat at the mounting points?

Anyway I canceled my pre-order long before all this magenta stuff,
I hope it gets corrected soon, good luck everyone.

But as a whole and admitted by BMD, the camera does have an overall magenta bias or cast, but that can be easily dialed out with white balance.

Benton, I don't mean to contradict you here as you were making a larger point, but I don't think we should be buying the whole "Alexa inherently leans green and the BMD 4.6K inherently leans magenta" line. Where are the facts to support it? I think it is completely untrue and misguided for two reasons:
One, people are confusing how the Alexa looks with ND filters (which often cause a green cast on every camera) versus how the Alexa looks without ND filters. The Alexa simply isn't green on its own. If someone has raw Alexa files shot in daylight that they can confirm were captured without ND yet they show a green cast, please share them.
Two, if you take the 4.6K raw sample files shared by BMD and put them into Resolve with the raw settings temp set to "as shot" and the color + gamma both set to REC709, those files balance absolutely perfectly. There is zero magenta cast and no need to adjust the tint at all. Please test it for yourselves. Do you see any hint of the BMD sample images leaning magenta?

Bottom line is that there shouldn't be any color cast to the 4.6K sensor in the cameras BMD is shipping. If we set the camera's kelvin temp to match the lighting, the camera should yield a balanced image, just like BMD's raw sample images. What they are actually selling needs to match what they've advertised.

But as a whole and admitted by BMD, the camera does have an overall magenta bias or cast, but that can be easily dialed out with white balance.

Benton, I don't mean to contradict you here as you were making a larger point, but I don't think we should be buying the whole "Alexa inherently leans green and the BMD 4.6K inherently leans magenta" line. Where are the facts to support it? I think it is completely untrue and misguided for two reasons:
One, people are confusing how the Alexa looks with ND filters (which often cause a green cast on every camera) versus how the Alexa looks without ND filters. The Alexa simply isn't green on its own. If someone has raw Alexa files shot in daylight that they can confirm were captured without ND yet they show a green cast, please share them.
Two, if you take the 4.6K raw sample files shared by BMD and put them into Resolve with the raw settings temp set to "as shot" and the color + gamma both set to REC709, those files balance absolutely perfectly. There is zero magenta cast and no need to adjust the tint at all. Please test it for yourselves. Do you see any hint of the BMD sample images leaning magenta?

Bottom line is that there shouldn't be any color cast to the 4.6K sensor in the cameras BMD is shipping. If we set the camera's kelvin temp to match the lighting, the camera should yield a balanced image, just like BMD's raw sample images. What they are actually selling needs to match what they've advertised.

+1

Gotcha. None of my processed frames had the saturation pumped up. I think Fahnon may have added a little bit, but nothing extreme. You can download all the original DNG and RAW files and play with them if you'd like. I believe Fahnon, did pump up the Canon 6D to try and make it show magenta and it did not. This is not to say it won't start looking ugly, but it didn't push magenta more than any other color. But as a whole and admitted by BMD, the camera does have an overall magenta bias or cast, but that can be easily dialed out with white balance.

I know. That's not what I'm asking. I'm wondering if other cameras also have something similar to "magenta corners"

I thought a Canon 6D and the BM Pocket camera counted as other cameras. But if you specifically meant cameras in the class of REDs and Arri's I don't know. But I believe at least the RED's have had there share of issues, but I don't think it was anything like the magenta corners on the UM4.6.

I'm specifically talking about doing the same gray card test that's being done on the 4.6K on other cameras to see if they show something like magenta corners with the contrast and saturation are boosted to unrealistic levels.. You didn't do a gray card test when the other cameras, so your test is not a good example of what I'm talking about..

I'm specifically talking about doing the same gray card test that's being done on the 4.6K on other cameras to see if they show something like magenta corners with the contrast and saturation are boosted to unrealistic levels.. You didn't do a gray card test with the other cameras, so your test is not a good example of what I'm talking about..

Actually, I think Benton's test can answer your question as he did shoot with other cameras.

If it's useful for comparison you can also take a look a few pages back in this thread where I also posted DNG files that are comparative shots of the BMCC and the Ursa Mini 4.6K. They are shot using the same lens and with DSC Labs color charts in the frame which can help you gauge appropriate saturation.
When I pulled them into Resolve, the BMCC did not develop a magenta cast on the edges when saturated to the same level of as my 4.6K Ursa Mini. You can download the DNG files from the link I provided and see how they look to you in Resolve.

Yesterday i was with a friend who just received the UM 4.6K, he told me that the camera didn't have any magenta issue, but when i put the aperature above f8 the magenta was there sitting in the corners
But in real world no one will shoot a wihte wall above f8

Yesterday i was with a friend who just received the UM 4.6K, he told me that the camera didn't have any magenta issue, but when i put the aperature above f8 the magenta was there sitting in the corners
But in real world no one will shoot a wihte wall above f8

Please don't take it the wrong way but,
I think that's irrelevant, the camera shouldn't produce magenta (or any color ) corners regardless of the circumstance/aperture... my BMCC doesn't.

Yesterday i was with a friend who just received the UM 4.6K, he told me that the camera didn't have any magenta issue, but when i put the aperature above f8 the magenta was there sitting in the corners
But in real world no one will shoot a wihte wall above f8

Please don't take it the wrong way but,
I think that's irrelevant, the camera shouldn't produce magenta (or any color ) corners regardless of the circumstance/aperture... my BMCC doesn't.

You are 100% right this is a problem, but some people really don't care like my friend loool, he didn't even know about it

Sent from my FP1U using Tapatalk

Bottom line is that there shouldn't be any color cast to the 4.6K sensor in the cameras BMD is shipping. If we set the camera's kelvin temp to match the lighting, the camera should yield a balanced image, just like BMD's raw sample images. What they are actually selling needs to match what they've advertised.

Exactly.

Any (RAW)-Camera's image, in all cases where one dominant temperature prevails, should be able to deliver image-data which can be color-cast-neutralized by adjusting nothing BUT the RGB-Components which are temperature (Red-Blue) & tint (Green) controls.

If that is not possible - and it is not fixable by redefining the Matrix (DNA of how the colors are presented), the camera is not performing to the standards of ANY digital camera (At the least, the ones I have had the pleasure to work with).

I must have gone thru' about 60 digital cameras the past many years and never have I encountered ONE which was not able to yield a neutral image by adjusting further than I mention above.

Perhaps, although I strongly doubt it, I have just been lucky with all those other cameras and what we see here is indeed performing according to standards. If that is the case then it all comes down to one choice.

Buy a Camera and get busy Shooting & Creating

or

Buy a Camera and get busy Testing & Grading

Gotcha. None of my processed frames had the saturation pumped up. I think Fahnon may have added a little bit, but nothing extreme. You can download all the original DNG and RAW files and play with them if you'd like. I believe Fahnon, did pump up the Canon 6D to try and make it show magenta and it did not. This is not to say it won't start looking ugly, but it didn't push magenta more than any other color. But as a whole and admitted by BMD, the camera does have an overall magenta bias or cast, but that can be easily dialed out with white balance.

I did add some saturation to show the issue is there, but nothing outside the realm of an actual saturated grade.

The 6D was uniform throughout as a camera should be.

But as a whole and admitted by BMD, the camera does have an overall magenta bias or cast, but that can be easily dialed out with white balance.

Benton, I don't mean to contradict you here as you were making a larger point, but I don't think we should be buying the whole "Alexa inherently leans green and the BMD 4.6K inherently leans magenta" line. Where are the facts to support it? I think it is completely untrue and misguided for two reasons:
One, people are confusing how the Alexa looks with ND filters (which often cause a green cast on every camera) versus how the Alexa looks without ND filters. The Alexa simply isn't green on its own. If someone has raw Alexa files shot in daylight that they can confirm were captured without ND yet they show a green cast, please share them.
Two, if you take the 4.6K raw sample files shared by BMD and put them into Resolve with the raw settings temp set to "as shot" and the color + gamma both set to REC709, those files balance absolutely perfectly. There is zero magenta cast and no need to adjust the tint at all. Please test it for yourselves. Do you see any hint of the BMD sample images leaning magenta?

Bottom line is that there shouldn't be any color cast to the 4.6K sensor in the cameras BMD is shipping. If we set the camera's kelvin temp to match the lighting, the camera should yield a balanced image, just like BMD's raw sample images. What they are actually selling needs to match what they've advertised.

Good point Jamie. The leaning magenta explanation could be a smoke screen, but they actually go so far to say that it's part of the color science. There is an interview from the floor of NAB where a representative says this. I completely agree with you that the camera should be 100% neutral straight out of the box, but if I had to shift the global tint one way or another to achieve that, while less than ideal, I could live with it, but of course the magenta corners issue makes that impossible.

...I completely agree with you that the camera should be 100% neutral straight out of the box, but if I had to shift the global tint one way or another to achieve that, while less than ideal, I could live with it, but of course the magenta corners issue makes that impossible.

I have never experienced an image coming off a camera that did not need a slight cast-adjustment in post. That is perfectly normal and to be expected. I.e. D65 is an approximation of a temperature in sunlight. However, sunlight-intensity varies from where on the planet you are located. So i.e. D65 is a starting point and one tweaks the temp/tint sliders from there to fine adjust the cast.

What is not normal is that an image has 7 different random local temperatures. That makes it impossible to adjust with temp/tint controls as those are global. There are issues of course with natural dual-light temperatures. I.e. a Wedding in the winter just before it turns dark outside. Inside there is this cozy warm candle light atmosphere while outside the window it is cold and blue. In this case balancing for either would make the other look wrong and it would have to be corrected with local masks.

In any case where one even-colored or neutral object is being lit by one global source like the sun - it should be possible to neutralize any cast with mere temp/tint controls. IMO

Actually, I think Benton's test can answer your question as he did shoot with other cameras.

If it's useful for comparison you can also take a look a few pages back in this thread where I also posted DNG files that are comparative shots of the BMCC and the Ursa Mini 4.6K. They are shot using the same lens and with DSC Labs color charts in the frame which can help you gauge appropriate saturation.
When I pulled them into Resolve, the BMCC did not develop a magenta cast on the edges when saturated to the same level of as my 4.6K Ursa Mini. You can download the DNG files from the link I provided and see how they look to you in Resolve.

No his test cannot answer the question. We need to see a clear field to really test if there's even a slight amount of corner casting on other sensors. Obviously we haven't seen the corner cast from other cameras to the degree that Benton's 4.6K shows. The question is if there is even a slight corner cast from other cameras, since there are some 4.6K's (maybe most for all we know) that barely show the corner cast. The only way we'd be able to tell is with a gray card test, not a color chart test.

If the issue is sensor color shading I don't know if shooting a grey or white card will show the full extent of the problem. Color shading is the sensor misreading color in part of the image. There's no color to misread on a white card so when balanced to neutral the magenta vignette mostly goes away. Shoot a blue sky etc and the error caused by the sensor misreading color can be seen.

If the issue is sensor color shading I don't know if shooting a grey or white card will show the full extent of the problem. Color shading is the sensor misreading color in part of the image. There's no color to misread on a white card so when balanced to neutral the magenta vignette mostly goes away. Shoot a blue sky etc and the error caused by the sensor misreading color can be seen.

+1

If the issue is sensor color shading I don't know if shooting a grey or white card will show the full extent of the problem. Color shading is the sensor misreading color in part of the image. There's no color to misread on a white card so when balanced to neutral the magenta vignette mostly goes away. Shoot a blue sky etc and the error caused by the sensor misreading color can be seen.

Obviously you haven't looked closely at the gray card tests, as those do show the 4.6K magenta corners issue in them when you increase the contest and saturation levels. In fact, it's the best way to show the issue since it's a neutral color all the way around, so if there is a cast anywhere in the frame it would definitely show up. A blue sky isn't necessary uniform through the entire frame as you can see the color of the sky changes slightly from east to west because of the sun. You also introduce IR contamination which could skew the results.

Obviously you haven't looked closely at the gray card tests, as those do show the 4.6K magenta corners issue in them when you increase the contest and saturation levels. In fact, it's the best way to show the issue since it's a neutral color all the way around, so if there is a cast anywhere in the frame it would definitely show up. A blue sky isn't necessary uniform through the entire frame as you can see the color of the sky changes slightly from east to west because of the sun. You also introduce IR contamination which could skew the results.

Aharon was referring to this ->

If the issue is sensor color shading

The big question "is there any UM 4.6k magenta free?"

The big question "is there any UM 4.6k magenta free?"

Someone must tell us))

The big question "is there any UM 4.6k magenta free?"

My dealer has sold 25 and thus far, I am the only one to have complained about it.
Judging by the movie John shot, it certainly appears that there are cameras not exhibiting those issues.

Also it is worth noting that more than 50% don't bother to post or read on a forum let alone know it exists.

It is all speculation but I would NOT think that any company would sign off on shipping a camera which produces an image so unpleasant as those coming from the UM46 suffering from what is being discussed here.
So I would think that as BMD signed off on shipping that the cameras leading to that decision were yielding wonderful images.

No one would sign off a shipping-go, knowing it would lead to RMA cases on so many cameras. It wouldn't be profitable !

The big question "is there any UM 4.6k magenta free?"

I've been through 5 of them now and all 5 had the magenta corners @ 35mm and f/8.

My dealer is very patient and I will look through more of them next week.

I think I have seen good frames at 35mm and f/8 in footage on the internet, but I haven't seen it with an actual camera in front of me yet, and that's scaring me.

0-5 so far. I will let you all know what i find out.

Meanwhile, if anyone can shoot and post clean DNGs ISO 800 from the 4.6k cam shot at f/8 and 35mm, especially if you use the Sigma 18-35mm lens on the 35mm side, and shoot a white wall that's lit so that you aren't all the way up the histogram (lower the shutter angle if you have to) that would be great.

I want to believe there are good ones are out there.

The big question "is there any UM 4.6k magenta free?"

I've been through 5 of them now and all 5 had the magenta corners @ 35mm and f/8.

My RMA is on the third week now. If you have gone thru' 5 the person handling your RMAs is MUCH more efficient than mine. How long did your RMAs take ?

The big question "is there any UM 4.6k magenta free?"

Not many. I am waiting on my 3rd 4.6k, it's been almost 3 weeks so I have no idea what the deal is with that. But several of my friends and coligues have received theirs and all 4 suffered from the issue. And all 4 were returned (not sure if any received their new camera). I'm in LA, and yes, there are a lot of professionals here, but 6 returns just in my group of friends is insane. I can't even imagine the overall number, this can't be good for business, and I hope they fix it soon. A local agency that I shoot for has recently adopted a no Blackmagic policy which if I ever get my replacement camera is going to hurt me. So BMD please pull your stuff together so individuals can regaine confidence in your products. Thanks.

John

I returned mine to my dealer, as far as I know BM isn't doing RMAs for this issue.

My dealer is interested too, and so we are opening boxes and testing them as we are able. 0-5 so far, more next week.

I think all of them have this issue, now if you use a PL lens and shoot at T3 or T4 the problem will not appear, on the EF side the problem is bigger i think if you put any EF lens at aperture f5.6 and beyond there is magenta

I returned mine to my dealer, as far as I know BM isn't doing RMAs for this issue.

My dealer is interested too, and so we are opening boxes and testing them as we are able. 0-5 so far, more next week.

Mine is going directly to BMD. According to dealer it IS on its way back. Whether repaired or Replaced, I don't know. But BMD IS doing RMAs on this particular issue. Even support stated that they would auth. the RMA after seeing my images.

Perhaps your dealer just got a guilty batch. That would explain the 5 to Nil Scenario.