Blackmagic Micro Cinema Camera

[squig]

6 minutes ago, giostrante said:

Relax guys, I know what  I've done, it's just a stress test, I didn't want to awake the professor inside all of you, keep calm!

It's more than a bit I'm afraid

Compared to the 5D MK3 the BMMCC FPN is practically non-existent (if you expose correctly). It is a noisier camera than the 5D raw overall, but the noise looks more filmic.

 

[giostrante]

I agree, the noise is more organic than all the Dsrls I've ever used.

[Bioskop.Inc]

10 minutes ago, giostrante said:

Relax guys, I know what I've done, it's just a stress test,I didn't want to awake the professor inside all of you, keep calm! 

And as I said, even with more light is clearly visible at iso 1600 and even at 800 when you look at the footage of course, here is just a still image.

Yeah, I understood that it was an extreme stress test - I'm always calm.

I never use ASA 1600 - just too much noise & so you need a plugin like Neat Video.

Use the ETTR rules: put zebras to 90% in ProRes & 100% for RAW. When you get peaking on your screen dial it down to taste (ie. if you want to blow a light source, blow it etc...) & then you'll get less noise. Just remember that these cameras are not like DSLRs at all & so there's a learning curve involved - forget everything that you've learnt with a DSLR & start again.

Then again, there might be something wrong with these new cameras - but BM does come out with FW fixes, so don't panic.

[squig]

There's low light, and then there's shit light. It's okay in low light, but not as good as the 5D MK3 raw.

 

[Nikkor]

2 minutes ago, squig said:

There's low light, and then there's shit light. It's okay in low light, but not as good as the 5D MK3 raw.

+100000000000

Shitlight is where 99% of the camera tests take place, totally misleading.

[squig]

Testing in shit light is a good idea. There's always those times when you have to shoot someone running down a dark alley at night naked chasing someone with a knife (without a permit). In that situation fixed pattern noise is going to be a real concern.

[Mattias Burling]

Shitlight is what seperates cameras. All cameras look awesome in controlled light. Both low and bright.

Just look at the Zacuto test. An Alexa and GH2 both look good. It just took 50 more lights for the GH2. The Alexa made do with the shitlight.

[richg101]

I've seen the fixed pattern noise from the bmmcc that wouldn;t be apparent from the same shot in the bmpcc.  I'd actually say the bmmcc might have a slightly reduced usable dr because of this - you can;t pull detail from the shadows as much as you can with the bmpcc from what I've seen so far.   I say this with a pinch of salt.  the image is still lush, and the 60p capability is worth taking a 1 stop hit on dr.  It just means i can't pull shadows up as much as I like to on shots where I;m exposing to retain highlight information. 

[Nikkor]

Shitlight is always going to look shit, with and without noise. If shitlight performance were what seperates cameras they would shoot blockbusters with gopros because they don't have shitlights there.

[Mattias Burling]

19 minutes ago, Nikkor said:

Shitlight is always going to look shit, with and without noise. If shitlight performance were what seperates cameras they would shoot blockbusters with gopros because they don't have shitlights there.

Its just that you forgot that there is a genre called "documentary".
I stand by my statement that was also proven in the Zacuto test. Pretty much any camera looks good if the DP has the know how and resources needed.
When there isn't time, money, knowledge, what have you, its all down to the camera, lens and codec.

GH2 with loads of lights

Alexa with Shitlight

Iphone with shit loads of light

 

And BTW, if you think Hollywood Blockbusters don't need to keep cost down and that they can use an endless amount of lights, think again.
 

 

20 minutes ago, richg101 said:

I've seen the fixed pattern noise from the bmmcc that wouldn;t be apparent from the same shot in the bmpcc.  I'd actually say the bmmcc might have a slightly reduced usable dr because of this - you can;t pull detail from the shadows as much as you can with the bmpcc from what I've seen so far.   I say this with a pinch of salt.  the image is still lush, and the 60p capability is worth taking a 1 stop hit on dr.  It just means i can't pull shadows up as much as I like to on shots where I;m exposing to retain highlight information. 

Same thing with the BMPC4K. Lots of noise if pushed to hard. Still worth it for the image imo.
Solution was to not push, or light the scene.

[ade towell]

'shitlight' in this context is giostrante's 'test' not Zacuto which actually used lights - the Alexa would have looked shit with giostrane's test too

[Mattias Burling]

5 minutes ago, ade towell said:

'shitlight' in this context is giostrante's 'test' not Zacuto which actually used lights - the Alexa would have looked shit with giostrane's test too

Shitlight was referring to "99% of the tests" in the post I responded to.

[Nikkor]

That's not the shitlight I'm talking about. Shitlight is when they do iso/noise tests without any lightsource except lightpollution from under a door or some streetlight and believe it or not, that's the way they make most forum/youtubers tests. 

The idea behind most of these is that the or a camera that looks better in that condition will work better in real photo/video situations. But that's just misleading.

[giostrante]

2 hours ago, squig said:

Testing in shit light is a good idea. There's always those times when you have to shoot someone running down a dark alley at night naked chasing someone with a knife (without a permit). In that situation fixed pattern noise is going to be a real concern.

 

1 hour ago, ade towell said:

'shitlight' in this context is giostrante's 'test' not Zacuto which actually used lights - the Alexa would have looked shit with giostrane's test too

I think that all of you, dear professors, are not familiar with the concept of "Stress Test".

It's pretty simple, you test something in an exagerated way to see how far you can go before it breaks, think about helmet or crash tests for example, they push to the extreme to see how far they can go, to determinate the limit of an object. That's exactly what I've done, is strange that all of you, shining minds, can't catch it.

As I said already, vertical pattern noise is present all over the image in ANY LIGHT CONDITION at iso 1600, more visible in the shadows or underexposed images of course, still there at iso 800 on my camera, but less pronounced. I've just pushed to show it more clearly, seems obvious to me.

And this is not normal at all! Normal is noise at high iso, not a full grid of vertical lines, my test is to point out the problem, should be clear even for the luminaries of this forum.

1 hour ago, Mattias Burling said:

Shitlight was referring to "99% of the tests" in the post I responded to.

Here's someone less professor and more smart! 

[JazzBox]

5 hours ago, giostrante said:

 

I've recently bought an used BMPCC and discovered this problem too, it's clear at iso 1600 but also at 800 is there. 

How can anyone be so calm and quiet about it? It's simply an horrible problem, it destroys the wonderful image this camera can produce in low light, it's a total shame, why nobody is complaining about it? I've just written to BM and ask for a solution, it's unacceptable. People make a lot of noise complaining about things like the old sony a7s black hole sun, or the artificial blu light problem but this is definitely worse!  

We all should ask them a serious solution!!

Giostrante, do you have this problem on the Pocket or also on the Micro? I ask, because they should have different sensor, so hopefully the issue could be fixed via FW update.
BlackMagic is a "little" company (compared to Canon, Sony, Panasonic...) and probably they need inputs on problems like those from real world user, but, in my humble opinion, noise or not it is more convenient a BM or a GH4 then a Canon 80...

[giostrante]

I don't have a Micro, just a Pocket and as I showed to Blackmagic my footage they told me to send the camera to them, this is to answer to other professors here who know for sure that my camera has not any problem,all judging from a screenshot. Amazing, no, wrong word, ridiculous fit better!

[Zak Forsman]

I have both the BMPCC and BMMCC and I do not see the same vertical striations in the my Pocket footage. No trace of it. On the BMMCC it appears in shadow areas (but not in complete black). I rolled a few frames off with the lens cap on and it's not there. Only seems to appear in parts of the frame that register in the neighborhood of 5 to 30 IRE.

[giostrante]

Bad news!!! Here's another with a good pocket, I can't wait to have mine fixed back!

[JazzBox]

6 minutes ago, Zak Forsman said:

I have both the BMPCC and BMMCC and I do not see the same vertical striations in the my Pocket footage. No trace of it. On the BMMCC it appears in shadow areas (but not in complete black). I rolled a few frames off with the lens cap on and it's not there. Only seems to appear in parts of the frame that register in the neighborhood of 5 to 30 IRE.

Zak, do you think it is something that needs to send the camera back to BM to be fixed?

13 minutes ago, giostrante said:

I don't have a Micro, just a Pocket and as I showed to Blackmagic my footage they told me to send the camera to them, this is to answer to other professors here who know for sure that my camera has not any problem,all judging from a screenshot. Amazing, no, wrong word, ridiculous fit better!

Thank you Giostrante.

In my humble opinion lots of cameras have issues: the V-Log from GH4 is way noisy in dark environment (in my personal experience, probably some other people will be happy with that), the CineD has some strange color cast on skin tones, but I feel that I have amazing images with that, compared with the Canon 70D I had.  

Sony A7s is great in low light but I don't like the final file that much (again, my opinion), the amazing 6300 has this overheating problem... 

Every camera has some amazing features and we need to accept compromises in order to have cheap cameras. Or, we need a lot of money for an Alexa  

I hope BM could solve all those issue from Pocket, Micro etc... because their files are the most cinematic I see at those prices... (in my opinion they also should give us some better ergonomics... I always say that I love the BM files, but I love also the easiness of use of GH4). Things like ISO, shutter, WB etc... should be easily accessible.

[Bioskop.Inc]

5 minutes ago, giostrante said:

Bad news!!! Here's another with a good pocket, I can't wait to have mine fixed back!

Here, try this on for size & so you can compare with yours - from my pocket really low light test (its anamorphic so change the vertical height - divide by 1.5). Shot with a Nikon 24mm @ f2.8 & Isco 54. This is stressed & I would never shoot like this, grade or even use this.

https://dl.dropboxusercontent.com/u/90827273/Pocket%20Low%20Light.dng

I raised the first by 5 stops, raised the highlight compression to 16, moved brightness to 60, Local contrast to 10 & contrast to 5 - all done in RPP. The second, I did the same but took some contrast out. This is RAW so ISO/ASA doesn't count