URSA Mini 4.6 - Repro Case| Sensor Bug -Magenta Edge Shift

John,
The images I see from Benton and Morten are not quite examples of consistent colour cast. These are problems and very serious ones. I am trying to figure out if this can be handled in Resolve, but I doubt.

That was John's point.

Many are posting images that had an overall cast. This is not an issue. Easy to balance in post... Just like on RED or Arri footage.

Some are posting images with inconsistent casts. This IS an issue, no one is questioning that, and they need to contact BM Support to work through the issue. That's what BM Support are there for.

If you think you have a sensor issue contact BM Support so they can sort it out for you.

John, yes. I have personal experience too. Not always in shooting, but tons in post. Our post house does about one film per month, most of them are Red. Just don't forget that our country produces the largest number of feature films in the world, so we have the problem of seeing things in bulk.

Btw, I do find BMPC4K a bit greenish too, much like the Alexa. *grin*

GREAT !

So the actual point...

Every camera has a cast. It's not abnormal to have a cast.

JB

Exactly. In fact they don't have a cast in all situations but if the light and angle of sun alters slightly from the kelvin set (what is expected from that exact number) the cast occurs.

This cast, though, is uniform and I have been dealing with Uniform Casts as long as I have had digital cameras. Nothing out of the ordinary there. But if filming a scene where the lighting guy did not forget to turn off the big magenta search light just right off the camera I have NEVER seen a non-uniform color cast resulting in the artifacts which the UM46 is capable of producing.

The Scene I captured yesterday with the center gray card on a white rock background had the sun @ 100deg vertical and center mass horizontally. The result was 100% correlating to the noise pattern captured with the turret cap on. That is a clear indication of a sensor or hardware issue. Not a standard color cast which I have been dealing with the past 2 decades almost.

Why would I "waste" my good time writing all this and doing all these experiments if all I had to do to resolve it is to iterate over a technique which I have been employing successfully the last 20 years

In the other Magenta Thread I was taking your stance, John. Implying to the folks in trouble that they should 'learn' how to proper white balance their imagery. Reason being that I simply could not imagine that this would be as issue. It was though and I stand corrected...

You've got me wondering though...

In the old thread about Magenta Casts - at a time where the camera was still in beta, you yourself - confirmed that the camera did in fact have the magenta cast issue and that it was known and should not be taken seriously as the camera was in beta...

So my question is this to you -> Given the nature that the images back then were suffering from the exact same issue as mine and those of others' are today - how come that today you claim that it is a tinting issue but back then you confirm that the camera has a magenta cast issue being looked into ? The symptoms are 100% identical !

Does that mean that the issues back then were also 'mere' color casts ?

Thats a very good point from Morten.

Has BM accepted there is a problem yet and it is being looked into or are most customers and potential customers still being told that any problem they find is normal and we should learn how to white balance properly or told to look at obscure examples from other cameras that do something funny too?

I haven't said that there isn't a problem.

I only point out...again....


1. It's not unusual to have a cast.
2. It's not unexpected to have fall off in corners.

Many of us are veterans of issues like this. There was Orbgate on the pocket. I was accused of exactly the same stuff, denying people had problems and being an apologist for BM. If you go back and look at the posts many of those that posted their pocket footage with orbs were posting normal out of focus highlights. Meanwhile, BM fixed the orb problem in very short time. Despite having a bad name for QC they seem to have turned a camera that had people out with pitchforks into a camera that has changed people's lives.

http://mattscottvisuals.com/blog/2016/1 ... ifechanger

So all I am trying to point out whilst not denying that some may have a problem, is the above two points....It's normal to have darker corners than in the middle. And that an overall cast is also pretty normal.

JB.

I haven't said that there isn't a problem.

I nearly point out...again....


1. It's not unusual to have a cast.
2. It's not unexpected to have fall off in corners.

Many of us are veterans of issues like this....

JB.

I agree 100%. As do all the others on this forum
We all know that casts are happening and that corners have fall-offs.

And nothing wrong with mentioning that - in threads about white balancing

But if you mention the obvious in a thread about the non-obvious it will inevitably be interpreted as if the non-obvious is in fact just the obvious being overlooked. In this thread and in this case, Color Casts and Edge Falloff are just as relevant as Saturation or sharpening

http://mattscottvisuals.com/blog/2016/1/28/bmpcclifechanger

Thanks for the link.. really great read! Hits close to home..

Aslong as it gets fixed. The one thing I can say about BM is that I rarely see them stumble across problems which they don't at some point address and leave everybody happy. I remember the uproar with the shortcomings for the original BMCC camera, but over time they pretty much made that little box into one incredible tool with little to be desired.

As long as it gets fixed. The one thing I can say about BM is that I rarely see them stumble across problems which they don't at some point address and leave everybody happy. I remember the uproar with the shortcomings for the original BMCC camera, but over time they pretty much made that little box into one incredible tool with little to be desired.

Or the problems in normal use end up being not so much of a problem ?

Anyone else remember the "split sensor" problem on the BMCC ?

viewtopic.php?f=2&t=4604

This fault is still there today !! Go check your bmcc but no one ever talks about it. The lens cap test would reveal it. But is it actually an issue ? A lens cap test will always by the way, reveal all kinds of sensor sins that have little bearing on the utility of the sensor.

And that's what I also mentioned in an another thread. I posted the TERRIBLE flat field image from my Leica. I was surprised at how bad it was but I'd never noticed it having a problem on the images.

Again, I'm not saying AT ALL that some might not have issues, but once you "see" magenta you're going to "see" it every where the same way we all "see" orbs everywhere and we "see" moire once it's pointed out to us.

I mean no one has even METERED the CT of their white walls and I guarantee an IMAC screen would measure about 9000K and if it's LED backlit, not be a very "nice" CRI light source either. You're moaning about the camera not being white balanced when you haven't even ensured you actually have a known light source.

Let's hold BM to account for sure, but let's also at least accept there are also other circumstances that can generate the same end result in these examples AND also work on a methodology that reveals what is and isn't an issue.

jb

I mean no one has even METERED the CT of their white walls and I guarantee an IMAC screen would measure about 9000K and if it's LED backlit, not be a very "nice" CRI light source either. You're moaning about the camera not being white balanced when you haven't even ensured you actually have a known light source.

1) You are assuming that I only measured the iMac center mass.. I actually placed the lens on 6 diff. locations on the iMac screen - EACH shot revealing identical result.

2) When shooting 3 Different White Walls and 6 diff. locations on a white iMac screen, one turret cap utter black on all diff. ISOs all results revealing identical non-uniformness, I think this qualifies as a pretty thorough test and not just some chump moaning off about his light-source being out of bounds !

-- Morten

1) You are assuming that I only measured the iMac center mass.. I actually placed the lens on 6 diff. locations on the iMac screen - EACH shot revealing identical result.

2) When shooting 3 Different White Walls and 6 diff. locations on a white iMac screen, one turret cap utter black on all diff. ISOs all results revealing identical non-uniformness, I think this qualifies as a pretty thorough test and not just some chump moaning off about his light-source being out of bounds !

-- Morten

If we're going to get into nitty and gritty, the light source has to emit the full spectrum of colour to be able to have a chance of being photographically represented in an even way. Daylight would be a good start or a tungsten light source as an alternative as both have a very flat spectral output (though expect a sniff more noise and less DR with tungsten)

Even better would be someone actually using a CT meter to say what the the prevailing lighting white point is.

JB

If we're going to get into nitty and gritty, the light source has to emit the full spectrum of colour to be able to have a chance of being photographically represented in an even way. Daylight would be a good start or a tungsten light source as an alternative as both have a very flat spectral output (though expect a sniff more noise and less DR with tungsten)

Even better would be someone actually using a CT meter to say what the the prevailing lighting white point is.

JB

Nitty and Gritty Nice.

No background is ever going to be evenly lit. Never. Been shooting digital for almost 20 years - never had this issue in any camera at any price... But hey, it is probably just because the sun started to shine differently all over the world and light the backgrounds unevenly all of a sudden Or perhaps the day I did the tests I (and all the others) were affected by local sun storms - we'd better try again today and check astrophysics prediction prior - so that we can ensure even spectrum lighting

I can pull out any other digital camera I have and shoot all the white walls in Hamburg Germany, all day long and perfectly grade them all in which ever software I please.... With the URSA Mini 4.6 I can't even shoot ONE white wall and grade that half-way satisfactorily.

But I am happy to learn and if I am the one making the mistake - I will stand corrected and that with a big smile on my lips too.

If we're going to get into nitty and gritty, the light source has to emit the full spectrum of colour to be able to have a chance of being photographically represented in an even way. Daylight would be a good start or a tungsten light source as an alternative as both have a very flat spectral output (though expect a sniff more noise and less DR with tungsten)

Even better would be someone actually using a CT meter to say what the the prevailing lighting white point is.

JB

Nitty and Gritty Nice.

No background is ever going to be evenly lit. Never. Been shooting digital for almost 20 years - never had this issue in any camera at any price... But hey, it is probably just because the sun started to shine differently all over the world and light the backgrounds unevenly all of a sudden Or perhaps the day I did the tests I (and all the others) were affected by local sun storms - we'd better try again today and check astrophysics prediction prior - so that we can ensure even spectrum lighting

I can pull out any other digital camera I have and shoot all the white walls in Hamburg Germany, all day long and perfectly grade them all in which ever software I please.... With the URSA Mini 4.6 I can't even shoot ONE white wall and grade that half-way satisfactorily.

But I am happy to learn and if I am the one making the mistake - I will stand corrected and that with a big smile on my lips too.

Haha this made me lol

Nitty and Gritty Nice.

No background is ever going to be evenly lit. Never. Been shooting digital for almost 20 years - never had this issue in any camera at any price...

Well that's not my experience at all. I guess you must have got lucky.

I'm simply suggesting that we try to eliminate variables to make sure we truly eliminate other possible explanations. You're making fun.

I want to get to the bottom of what's going on and you already know.

JB

Nitty and Gritty Nice.

No background is ever going to be evenly lit. Never. Been shooting digital for almost 20 years - never had this issue in any camera at any price...

Well that's not my experience at all. I guess you must have got lucky.

I'm simply suggesting that we try to eliminate variables to make sure we truly eliminate other possible explanations. You're making fun.

I want to get to the bottom of what's going on and you already know.

JB

Not wanting to get involved, but I see both sides here as a bystander to the dispute. I think maybe your replies have an air of disbelief in his issue and the tone sometimes comes across as condescending. If you really want to figure it out maybe have him Skype/FaceTime you and see his setup and what he's doing first hand and maybe take delivery of some files he has shot and see what you think? This back and forth doesn't seem to be helping anyone.

Nitty and Gritty Nice.

No background is ever going to be evenly lit. Never. Been shooting digital for almost 20 years - never had this issue in any camera at any price...

Well that's not my experience at all. I guess you must have got lucky.

I'm simply suggesting that we try to eliminate variables to make sure we truly eliminate other possible explanations. You're making fun.

I want to get to the bottom of what's going on and you already know.

JB

No I don't know. Why do you think I am doing all this ????

And I am not making fun. Do you think it is funny to pay 9kEuro for camera you can't use ?

Do you think I made all theses experiments just so that I could get to write on a forum ?
I don't need this here. Any more than I need rain on day where I want to throw beef a on barb.

So I beg you to remember that before inferring that I am too green(no pun intended) to distinguish a simple magenta color cast from a non-uniform color cast which only one camera on this planet that I know of is capable of producing

OK. You guys win. You know it all.

I'll leave you to it.

Good shooting.

JB

OK. You guys win. You know it all.

I'll leave you to it.

Good shooting.

JB

That's a mature response... Come on he is asking for help on the issue. If you think it's a natural issue with the sensor can you provide examples? Can you show us how to get rid of the issue in resolve? Can you fix the issue your end?

The attachment imacWhite.jpg is no longer available

Morten Carlsen

Morten, I found I got a very different type of uneven magenta cast with different light sources. When I used a florescent transparency viewer light box (calibrated at D5000), I got a horrendous cast, similar to your iMac image. When I used daylight shooting an open sensor through a PhaseOne supplied white plexi, the magenta cast showed up as a gradient from right to left. This was still not good or acceptable, but it did reveal that different light sources have different effects. Are you able to try a daylight source for your open sensor test? Shooting through a piece of white paper against a window should work if you don't have a translucent white plexi.

Florescent viewer saturation pushed 100%

Blackmagic URSA Mini_1_2016-03-30_2135_C0001_000018 1.jpg (411.16 KiB) Viewed 12055 times

Daylight through white plexiglas saturation pushed 100%

Blackmagic URSA Mini_1_2016-04-03_2325_C0014_000014 2.jpg (586.66 KiB) Viewed 12055 times

Have you guys contacted BMD about a RMA? There seem to be units out there that don't exhibit this problem.

Light that's evenly balanced across a spectrum won't show variance when individual channels are balanced to clean across a lighter-darker vignette. Unbalanced light will separate unevely across light to dark in just a vignette as it is balanced across just highlights for instance; a different color will emerge in the shadow area of the vignette.
To illustrate: using raw settings only the DNGs posted can be balanced and then lightened or darkened without seeing color in the vignette. That's because the the light is interacting with something that is single color and is balanced. Interacting with other colors (put a face in the vignette) will show the variance in the RGB channels at different places in the RGB parade as they get darker through the vignette.

In a unbalanced lighting scenario the light is uneven across the RGB parade, and uneven through the shadow of the vignette. Balancing for the center won't balance for the edges and vice versa. This is the color vignette we've been seeing.
Ie. Functioning normally: Unbalanced light (shooting outside with tungsten setting etc) The RGB parade is out of balance leaning towards a color cast. Balancing the channels will balance the entire image and eliminate the cast.
Functioning Abnormally. Uneven light represents in different values across RGB depending on luma. The same chart chip shot at the wrong color balance will land in different places on the RGB parade depending on where it's placed in the frame and the luma values of the vignette. As we correct the chart we see different colors emerge even though they where lit with the same intensity of unbalanced light.

At this point it's speculation but my rationale:
The 680nm wavelength is in the visible light spectrum...
The Schneider IR CUT IF489 has a spectrum response curve that I like in theory.

http://www.schneiderkreuznach.com/en...ir-cut-filter/

Sent from my iPad using Tapatalk

A very reasonable theory for sure, but the proof is always in the pudding. Are you able to exchange the 750 if it turns out you need to dig deeper toward visible light to cut any red pollution? Maybe they could let you test the 680, 715 and 750 and return the ones that didn't work as well? I recall the tests I saw for the original BMCC and it clearly needed the 680 to look it's best and the new 4.6k has been reported to have much of the 2.5k's DNA. BM cameras have always been very sensitive to IR and in past experience, need a pretty heavy IR hammer.

Thanks, Benton and James. I think further inquiry with Schneider is warranted and a test of the IF489 and True Cut 680/715/750 would be ideal. Today I have no idea if I can have access to any or all of these filters.

Assumptions:
The red or brown or muddy effect we are told is due to the sensor's sensitivity to infrared wavelengths.

Near-infrared refers to wavelengths from 700nm to 1 micron and according to Wikipedia in test situations some people can detect 'light' in the near-infrared range.

Observations:
Spectrum charts of the performance of 680nm IR Cut filters show a very steep drop in sensitivity from around 680nm and almost total cutting by 700nm.

Thesis:
680 cut on a digital image is simply too soon and must affect the colour perception of deep red.

Methodology:
Practically speaking, the best proof available to me is just using filters with different response curves on the 4.6K sensor to see what looks best to most people.

If and when I can, I'll try to test it, but I and everyone else should keep an open mind because skin colour perception may vary among a group of people or even vary between your left eye and right eye. Therefore which IR Cut approach is 'right' for the 4.6K sensor largely becomes a subjective choice.

Wishlist:
BMD should be doing the objective tests. I think they should be very curious about the behaviour of their 4.6K sensor since they're both engineers and artists.

I haven't said that there isn't a problem.

I only point out...again....


1. It's not unusual to have a cast.
2. It's not unexpected to have fall off in corners.

If you use crap lenses, yes.
If you use high quality lenses as I do, then there is no vignetting at all.

So only tests with an unmounted lens should be considered in the magenta test.

I don't have a horse in this race, but this thread is a sad read. I expected more support from JB.

I haven't said that there isn't a problem.

I only point out...again....


1. It's not unusual to have a cast.
2. It's not unexpected to have fall off in corners.

If you use crap lenses, yes.
If you use high quality lenses as I do, then there is no vignetting at all.

So only tests with an unmounted lens should be considered in the magenta test.

There is still fall off in the corners. Even without a lens. Even on other cameras. Look at my A7s image in the other thread:
viewtopic.php?f=2&t=46112&start=250#p268591

I haven't said that there isn't a problem.

I only point out...again....


1. It's not unusual to have a cast.
2. It's not unexpected to have fall off in corners.

If you use crap lenses, yes.
If you use high quality lenses as I do, then there is no vignetting at all.

So only tests with an unmounted lens should be considered in the magenta test.

It's funny, I never considered Zeiss lenses to be crap. But I guess they are since they are among the lenses here that show vignetting and a uneven magenta cast posted by some users here.

It's funny I never considered Zeiss lenses to be crap. But I guess they are since they are among the lenses here that show vignetting and a uneven magenta cast posted by some users here.

My 50mm Zeiss has done me well the past 6 years I have had it. It aint the most expensive (700 EURO) but is is great. Never sick at sea.My 100mm Makro from Zeiss as well. Great lens. My OTUS 28mm as well.

All giving me wonderful images on any camera I have, on todays jewelry-shoot as well with the BM4k and Canon 5DMKIII. But all three lenses mentioned above must be REALLY crap because they produce a wild color shift when put on the URSA Mini 4.6k which is impossible to neutralize.

The URSA Mini for all I care may be able to deliver wonderful images without a lens. But WITH a lens, at least all 3 above mentioned aren't capable of yielding a result which I would consider satisfactory on this cam. And who cares about a camera shooting nice images but only without a lens

I did try your suggestion shooting in daylight already - with a lens on the results are messed up. Perhaps the turret is crooked, perhaps the filters aren't attached properly, perhaps there are current irregularities perhaps perhaps perhaps... I will ask my dealer to supply me a new one and try with that. 1 Test 1 Background. If it fails, I'm off...

Just a thought... I must have spent days literally in camera stores - bringing my own storage card, testing Lenses and cameras and looking at the results when I get back... The shots were taken in puke-ugly fluorescent lighting within the store and a few outside - never ONCE did I have a result remotely similar to this problem I and many others on this forum are having. Boy, must I have been lucky - and that for almost 20 years of doing it !

This issue is as real as it gets. And while there is a chance that my brain died overnight and the devil is carrying over me a broken array of Magenta LEDs flashing up every time I hit record - I seriously doubt it

Anyway, I am thru' with testing. All test are yielding exact same result !

I'll report back when/if I get the new camera !

It's funny I never considered Zeiss lenses to be crap. But I guess they are since they are among the lenses here that show vignetting and a uneven magenta cast posted by some users here.

My 50mm Zeiss has done me well the past 6 years I have had it. It aint the most expensive (700 EURO) but is is great. Never sick at sea.My 100mm Makro from Zeiss as well. Great lens. My OTUS 28mm as well.

All giving me wonderful images on any camera I have, on todays jewelry-shoot as well with the BM4k and Canon 5DMKIII. But all three lenses mentioned above must be REALLY crap because they produce a wild color shift when put on the URSA Mini 4.6k which is impossible to neutralize.

The URSA Mini for all I care may be able to deliver wonderful images without a lens. But WITH a lens, at least all 3 above mentioned aren't capable of yielding a result which I would consider satisfactory on this cam. And who cares about a camera shooting nice images but only without a lens

I did try your suggestion shooting in daylight already - with a lens on the results are messed up. Perhaps the turret is crooked, perhaps the filters aren't attached properly, perhaps there are current irregularities perhaps perhaps perhaps... I will ask my dealer to supply me a new one and try with that. 1 Test 1 Background. If it fails, I'm off...

Just a thought... I must have spent days literally in camera stores - bringing my own storage card, testing Lenses and cameras and looking at the results when I get back... The shots were taken in puke-ugly fluorescent lighting within the store and a few outside - never ONCE did I have a result remotely similar to this problem I and many others on this forum are having. Boy, must I have been lucky - and that for almost 20 years of doing it !

This issue is as real as it gets. And while there is a chance that my brain died overnight and the devil is carrying over me a broken array of Magenta LEDs flashing up every time I hit record - I seriously doubt it

Anyway, I am thru' with testing. All test are yielding exact same result !

I'll report back when/if I get the new camera !

Thanks for all your input on this issue Morten! I'm a believer in your capped sensor residual noise test. Good luck in getting a different camera!

Having seen the different test frames, I'm wondering if this is simply a manufacturing issue. I'm thinking something like uneven application of thermal compound between the sensor and the cooling unit.
Given the pressure which BMD likely is under to get the cameras to the streets, it would be conceivable that the QC had to take a step or two back - which seems to be backfiring...

Having seen the different test frames, I'm wondering if this is simply a manufacturing issue. I'm thinking something like uneven application of thermal compound between the sensor and the cooling unit.
Given the pressure which BMD likely is under to get the cameras to the streets, it would be conceivable that the QC had to take a step or two back - which seems to be backfiring...

I am thinking the same... When extreme contrast is applied to an image shot on the came there are weird artifacts on the image which I can reproduce on no other camera... Look analogous to the air pockets which can occur on the thin protection layer users can apply to iPhone glass.

Wonderful Camera in form factor and features etc but my sensor or something inside the cam. is out of whack for sure. Hopefully a replacement is going to fix it. Time will tell

Thanks for all your input on this issue Morten! I'm a believer in your capped sensor residual noise test. Good luck in getting a different camera!

Thanks Benton !

I'll be sure to update this topic as soon as something comes along !

So I'm happy to report that I shot with the camera for the first time yesterday and the footage exhibits no weirdness and looks as I'd expect at a glance. It wasn't the best test as when I switched the recording mode from raw to prores, the camera switched the gamma to video and I didn't realize. Blue skies and white walls looked as expected.

Having said that, I know that the problem may not present itself at all times and I might not have run into it yet, so this weekend I'm doing a battery of tests along side a BMCC and a Canon 6D in the real world and the white wall test. If this camera isn't as perfect as the other two, I'm sending it back. It's too large an investment to chance it.

Benton, since we live on opposite sides of the same park, we might want to test together if you're up for it.

I haven't said that there isn't a problem.

I only point out...again....


1. It's not unusual to have a cast.
2. It's not unexpected to have fall off in corners.

If you use crap lenses, yes.
If you use high quality lenses as I do, then there is no vignetting at all.

On large sensor (bigger than APSC/Super35), there can be vignetting even on great lenses. Ask any stills shooter using full frame. It's also sometimes dependent on aperture and focal length (if a zoom).

Either way, if it's not happening on any other camera and is happening to some on this one, something's up...

So I'm happy to report that I shot with the camera for the first time yesterday and the footage exhibits no weirdness and looks as I'd expect at a glance. It wasn't the best test as when I switched the recording mode from raw to prores, the camera switched the gamma to video and I didn't realize. Blue skies and white walls looked as expected.

Having said that, I know that the problem may not present itself at all times and I might not have run into it yet, so this weekend I'm doing a battery of tests along side a BMCC and a Canon 6D in the real world and the white wall test. If this camera isn't as perfect as the other two, I'm sending it back. It's too large an investment to chance it.

Benton, since we live on opposite sides of the same park, we might want to test together if you're up for it.

Absolutely Fahnon. I get a replacement camera tomorrow and want to put it to the test. Two different 4.6k cameras pointing at the same subject would be a very interesting test. I'll PM you.

I look forward to Benton's new test with the replacement!

Man waiting for the bank to clear so I can order is MURDER lol. But I mean If the longer wait means I get a clean unit whatever XD

So I'm happy to report that I shot with the camera for the first time yesterday and the footage exhibits no weirdness and looks as I'd expect at a glance. It wasn't the best test as when I switched the recording mode from raw to prores, the camera switched the gamma to video and I didn't realize. Blue skies and white walls looked as expected.

Having said that, I know that the problem may not present itself at all times and I might not have run into it yet, so this weekend I'm doing a battery of tests along side a BMCC and a Canon 6D in the real world and the white wall test. If this camera isn't as perfect as the other two, I'm sending it back. It's too large an investment to chance it.

Benton, since we live on opposite sides of the same park, we might want to test together if you're up for it.

Fahnon, I tried sending you a PM a while ago, but it's just sitting in my outbox and I can't find any way to motivate it to go on it's way to you. I think it has to show up in my "Sent" folder for it to be received by you, which so far, it hasn't done. If you don't get it, please try to PM me. Thanks.

Fahnon, I tried sending you a PM a while ago, but it's just sitting in my outbox and I can't find any way to motivate it to go on it's way to you. I think it has to show up in my "Sent" folder for it to be received by you, which so far, it hasn't done. If you don't get it, please try to PM me. Thanks.

I believe it sits in your Outbox until it's read by the other user. Then moves to your Sent box once it's been read.

Not Exactly Sure how you guys are processing this test to get these weird hues, but here is my URSA Sensor and My 4.6K Sensor. Also notice how on the 4.6K the blue channel is raised and on the 4K Original Ursa the red channel is raised.

Default Resolve Import

DefaultCap.jpg (151.39 KiB) Viewed 11625 times

Left at Rec 709 set to 5600K 0 Tint exposure set max iso plus the max 5 stops in the RAW Tab. As you can see the URSA 4K and my 4.6K mini both have an even uniform noise. I even boost saturation and color boost to introduce exaggerate colors and nothing.

CapTestGainBoost.jpg (233.09 KiB) Viewed 11625 times

Hi all, I just received mine yesterday... We did some quick tests and found that the overall image was slightly on the magenta side, especially when compared to the 2.5 BMCC, which I had great hopes for the two to match... But I can't find an uneven distribution yet... The image actually looks stunningly good. It just does not match my 4 BMCC's which is a bit of a bummer... Maybe our tests were not refined enough, but it seems like a working camera. I just find funny that you have to open the screen to turn it on, which seems a bit award when using the EVF.


Uberraum
Kai v. Ahlefeld
69, bd Richard Lenoir
75011 Paris

With respects to image quality, BM, and this post... Specifically relating to the comment about how the Pocket Cam's pricing sits compared to the Ursa 4.6...

Tested the BMMCC, bought one this week, and the Pocket, owned for about 1 year and 2 months now, to see if there were any magenta issues. In both cases, neither camera displays issues.

I appreciate this particular thread. Hope this info helps.

Just for comparison's sake, I've tested my 4.6K with 2 lenses so far and haven't had any magenta corners. I've had the slightest luminance change around the edges but I couldn't see it with my eyeballs. Had to use a color picker.

Xeen 24mm f/4 f/2.8
Canon 24-105 f/4
3840x2160 ProRes 422 BMD Film
ISO 800 and 400
60p full sensor

I can pull out any other digital camera I have and shoot all the white walls in Hamburg Germany, all day long and perfectly grade them all in which ever software I please....

You might just be lucky. I have started to research this in the last few days and have found it can happen on other cameras too with people complaining about it.

A7R



http://www.photoclubalpha.com/2014/04/28/sony-alpha-7r-the-swiss-army-knife-camera/

Z3



http://www.xperiablog.net/2014/10/27/xperia-z3-owners-reporting-issues-with-pink-shading-in-pictures/

Leica M9



http://diglloyd.com/blog/2013/20131013_1-SonyA7r-color-shading.html

Sony again



NEX7



http://diglloyd.com/blog/2012/20120105_4-SonyNEX7-ColorShading.html

I can pull out any other digital camera I have and shoot all the white walls in Hamburg Germany, all day long and perfectly grade them all in which ever software I please....

You might just be lucky. I have started to research this in the last few days and have found it can happen on other cameras too with people complaining about it.

Surely everyone here realizes that we are talking about even and balanced falloff versus the few cameras that show uneven balance and falloff.

Is like we have stepped back 10 years and we are going through the whole digital camera revolution again where people are just discovering RAW for video but the stills guys have been at it for a while ?

Reminds me of the day I bought my Nikon D90 and used it for video.....of course it didn't shoot high bit depth and bit rate RAW and the image was pre-processed before it came out of the camera.

Hmmmmm.....

You might just be lucky. I have started to research this in the last few days and have found it can happen on other cameras too with people complaining about it.

Any camera coming off an assembly line could potentially suffer from numerous kinds of flaws. No camera maker regardless of brand are immune to such. As I am sure Canon, Leika etc etc. have had to RMA 100s of malfunctioning cameras in their lifespan.

Thing is, when one gets a new camera he does test shots and if he has been buying, using & grading for a large number of years he would quickly notice if something wasn't right. Just as I did with the UM46. If that happens (it has never happened to me before) one has to take action and get a properly functioning camera.

I must have graded 1000s of portraits shot on white backgrounds for my own needs and for numerous of professional photographers and magazines. Not once did I ever receive an image exhibiting the flaws my UM46 has. Not to say that none of those photographers did never get hold of a bad camera. But if they did, the got an RMA. I have been dealing with Vignettes for just as long - nothing special there as they are always going to be present. But if one half of the image looks totally off from the other - then something is wrong.

If the shot was a rare one like a one in lifetime shot. Of course, those color shifts could be dealt with just as images for magazines are dealt with. 200 Photoshop layers and off you go. Nothing out of the ordinary. Simple touch-up stuff. Only - you don't wanna have to do that to every shot you take. Image doing a wedding video with my UM46 - LOTs of white in a wedding shot in low-light mostly. I doubt that the bride and groom unless carrying the surname Branson, would be able to afford a wedding video without Magenta & cyan Edges all over the place.

The UM46 is a new camera - it will have child-diseases. Not all(Probably not very many) UM46s are suffering from this. One user here has posted flawless images from his UM46 - so I am confident that this will become contained.

As far as my luck and white backgrounds go - if what I have experienced the past 20 years has been mere luck, then statistically, chances of me winning the lottery for 10 consecutive Saturdays are greater. So I had best get off this here Computer and go buy some tickets.

You might just be lucky. I have started to research this in the last few days and have found it can happen on other cameras too with people complaining about it.

Any camera coming off an assembly line could potentially suffer from numerous kinds of flaws. No camera maker regardless of brand are immune to such. As I am sure Canon, Leika etc etc. have had to RMA 100s of malfunctioning cameras in their lifespan.

Thing is, when one gets a new camera he does test shots and if he has been buying, using & grading for a large number of years he would quickly notice if something wasn't right. Just as I did with the UM46. If that happens (it has never happened to me before) one has to take action and get a properly functioning camera.

I must have graded 1000s of portraits shot on white backgrounds for my own needs and for numerous of professional photographers and magazines. Not once did I ever receive an image exhibiting the flaws my UM46 has. Not to say that none of those photographers did never get hold of a bad camera. But if they did, the got an RMA. I have been dealing with Vignettes for just as long - nothing special there as they are always going to be present. But if one half of the image looks totally off from the other - then something is wrong.

If the shot was a rare one like a one in lifetime shot. Of course, those color shifts could be dealt with just as images for magazines are dealt with. 200 Photoshop layers and off you go. Nothing out of the ordinary. Simple touch-up stuff. Only - you don't wanna have to do that to every shot you take. Image doing a wedding video with my UM46 - LOTs of white in a wedding shot in low-light mostly. I doubt that the bride and groom unless carrying the surname Branson, would be able to afford a wedding video without Magenta & cyan Edges all over the place.

The UM46 is a new camera - it will have child-diseases. Not all(Probably not very many) UM46s are suffering from this. One user here has posted flawless images from his UM46 - so I am confident that this will become contained.

As far as my luck and white backgrounds go - if what I have experienced the past 20 years has been mere luck, then statistically, chances of me winning the lottery for 10 consecutive Saturdays are greater. So I had best get off this here Computer and go buy some tickets.

Mate...you do what we all do with crap we buy, either choose to keep it or send it back and get a refund or if your are up for slightly more pain, then chase a replacement.

Come to think of it, absolutely sucked but there was a job I once had where they replaced me....

Mate...you do what we all do with crap we buy, either choose to keep it or send it back and get a refund or if your are up for slightly more pain, then chase a replacement.

Yeah... Only the UM46 is in no way crap. It is a fantastic piece of engineering !

Mate...you do what we all do with crap we buy, either choose to keep it or send it back and get a refund or if your are up for slightly more pain, then chase a replacement.

Yeah... Only the UM46 is in no way crap. It is a fantastic piece of engineering !

Yup...most people get a good one and for some reason a small percentage of people don't.

Having seen the different test frames, I'm wondering if this is simply a manufacturing issue. I'm thinking something like uneven application of thermal compound between the sensor and the cooling unit.
Given the pressure which BMD likely is under to get the cameras to the streets, it would be conceivable that the QC had to take a step or two back - which seems to be backfiring...

Think you are right as well.
A. Images of the sensor with the body cap on showing uneven noise distribution
B. The waveforms from .dng posted with magenta vignette show falloff in the corners but also overall uneven luma distribution, almost like the sensor is leaning in one direction.
C. A tilted sensor would create a disparity between the lens and sensor microlenses creating falloff in general but more falloff on the side where light leaving the lens hits the microlenses at a angle.
D. Have seen a post mentioning more sensor color shading at lower t-stops like 2.8 and less at higher t-stops like 8. If this issue is cause by sensor tilt it would fit well with the theory that this issue is sensor color shading as and vignetting on the micros lens scale. Themicro lenses aren't getting enough light or enough direct light, causing falloff and perhaps cross talk at the sensor edges.

My images look like this in most cases...however, every dng I have sent to tech support keeps coming back as fine...I do not this this is right...here is one that I know has issues...can anyone please tell me if this is normal..after 20 years of working with video and film..I have not seen this before.
Thanks!
Chad

https://www.dropbox.com/s/hj0hhv1o4sqnh ... 1.dng?dl=0

My images look like this in most cases...however, every dng I have sent to tech support keeps coming back as fine...I do not this this is right...here is one that I know has issues...can anyone please tell me if this is normal..after 20 years of working with video and film..I have not seen this before.
Thanks!
Chad

https://www.dropbox.com/s/hj0hhv1o4sqnh ... 1.dng?dl=0

That image would bug me big time.
Believe me or not, it's your decision. I have over 34 years of production experience and am just now seeing this kind of image flaw??? You know what, it kind of reminds me of the old color tube cameras that I had many moons ago. Those things were a mess.

Not all of us here are new to the game. I will hold off on the turret until I see that this has been resolved.

My BIG URSA does not have any issues like that!!