Page 1 of 5

BMCC Killer?

PostPosted: Tue May 07, 2013 6:58 pm
by Rudy Satria
Will 5D Mark III with Magic Lantern could beat dynamic range of BMCC ?
Please let me know what you think about this video

Thanks :D

This is not a debate or arguing whos the winner between BMCC and 5D. Everyone has different thoughts so please respect to every opinion that comes up in this topic ;)

-Rudy

Re: BMCC Killer?

PostPosted: Tue May 07, 2013 7:10 pm
by Tom
No, for the following reasons:


1: Dynamic Range is better on the BMCC by a significant amount.
2: This is not true raw, these are 4:2:2 14 bit dng's - not Bayer pattern Uncompressed raw like on the bmcc.
3: Even if they get it to HD resolution at good frame rates - there is no avoiding the limited buffer issue. Even the head ML dev has stated that recording for more than a few seconds is not likely to ever be possible. Hence it is great for burst shooting and silent shutter, but not really ever feasible for proper video.
4: Lower Resolution than BMCC at proper frame rates.
5: Different colour science (I would argue inferior, but its down to taste I suppose)

Re: BMCC Killer?

PostPosted: Tue May 07, 2013 7:16 pm
by spike
BMD are killing the BMCC all by themselves!

Re: BMCC Killer?

PostPosted: Tue May 07, 2013 7:23 pm
by Rudy Satria
Tom wrote:No, for the following reasons:


1: Dynamic Range is better on the BMCC by a significant amount.
2: This is not true raw, these are 4:2:2 14 bit dng's - not Bayer pattern Uncompressed raw like on the bmcc.
3: Even if they get it to HD resolution at good frame rates - there is no avoiding the limited buffer issue. Even the head ML dev has stated that recording for more than a few seconds is not likely to ever be possible. Hence it is great for burst shooting and silent shutter, but not really ever feasible for proper video.
4: Lower Resolution than BMCC at proper frame rates.
5: Different colour science (I would argue inferior, but its down to taste I suppose)


I can see the point. I hope BMD could see this and always steps ahead. Thanks for your quick response, Tom.

Re: BMCC Killer?

PostPosted: Tue May 07, 2013 7:28 pm
by Rudy Satria
spike wrote:BMD are killing the BMCC all by themselves!


Hi Spike, do you think BMD forced us to buy their side products? could you make it clear for me? Thanks Spike :)

Re: BMCC Killer?

PostPosted: Tue May 07, 2013 8:05 pm
by spike
rudysatria wrote:
spike wrote:BMD are killing the BMCC all by themselves!


Hi Spike, do you think BMD forced us to buy their side products? could you make it clear for me? Thanks Spike :)


If I understood what on earth you were talking about I would certainly try.

Re: BMCC Killer?

PostPosted: Tue May 07, 2013 8:59 pm
by Trevor Zuck
Tom wrote:No, for the following reasons:


1: Dynamic Range is better on the BMCC by a significant amount.
2: This is not true raw, these are 4:2:2 14 bit dng's - not Bayer pattern Uncompressed raw like on the bmcc.
3: Even if they get it to HD resolution at good frame rates - there is no avoiding the limited buffer issue. Even the head ML dev has stated that recording for more than a few seconds is not likely to ever be possible. Hence it is great for burst shooting and silent shutter, but not really ever feasible for proper video.
4: Lower Resolution than BMCC at proper frame rates.
5: Different colour science (I would argue inferior, but its down to taste I suppose)


I have a qualm with #2, there is "raw" and then there is "RAW"

"raw" the word can mean footage straight from the camera, unedited, uncorrected footage
"RAW" signifying a camera/image RAW codec or file, like RED RAW, ARRI RAW, or DNG. all DNGs are an image RAW file, and most RAW formats can utilize compression.

Also really and truly if you want "true" raw you would be looking at a bayered image.

and 14-bit color depth is 4 times better than 12-bit, because math.
(for those that don't know color bit depth is calculated as 2^n colors where n = bit depth. so 2^12=4096, 2^14=16,384)

Re: BMCC Killer?

PostPosted: Tue May 07, 2013 9:40 pm
by Aaron Scheiner
Bit depth is one thing, whether or not you have the dynamic range to utilise that depth is another.

Also, DSLRs suffer from nasty aliasing due to the way the image is downscaled before being recorded, BMCCs don't suffer anywhere nearly as badly from this as the recorded resolution is the native sensor resolution (no rescaling takes place{cropping does}).

Re: BMCC Killer?

PostPosted: Tue May 07, 2013 10:10 pm
by Rakesh Malik
spike wrote:BMD are killing the BMCC all by themselves!


Yes, they are, sadly.

Re: BMCC Killer?

PostPosted: Tue May 07, 2013 10:21 pm
by Trevor Zuck
Aaron Scheiner wrote:Bit depth is one thing, whether or not you have the dynamic range to utilise that depth is another.

Also, DSLRs suffer from nasty aliasing due to the way the image is downscaled before being recorded, BMCCs don't suffer anywhere nearly as badly from this as the recorded resolution is the native sensor resolution (no rescaling takes place{cropping does}).


the depth helps when stretching those fewer stops of DR though, and at 3.5K resolution its not nearly as bad aliasing, and the low light performance is going to be better with that 5Dmkiii...

I personally think the 5D is way over rated and by no means a threat to the BM cams, but lets make sure our points are correct before bashing the competition, you know, to keep it classy.

Re: BMCC Killer?

PostPosted: Tue May 07, 2013 10:44 pm
by John Brawley
TZuck wrote:
Tom wrote:
I have a qualm with #2, there is "raw" and then there is "RAW"

"raw" the word can mean footage straight from the camera, unedited, uncorrected footage
"RAW" signifying a camera/image RAW codec or file, like RED RAW, ARRI RAW, or DNG. all DNGs are an image RAW file, and most RAW formats can utilize compression.

Also really and truly if you want "true" raw you would be looking at a bayered image.

and 14-bit color depth is 4 times better than 12-bit, because math.
(for those that don't know color bit depth is calculated as 2^n colors where n = bit depth. so 2^12=4096, 2^14=16,384)



Couple of things.

The bmcc sensor is more than 16 bit. The RAW sensor data is encoded into a 16bit LIN file which is then encoded into a 12 bit LOG file before becoming a DNG. You'll notice the DNG file unpacks as a 16 bit file in ACR and Resolve.

I'm not sure the distinction between raw and RAW is ratified anywhere. In my books, RAW means WB and ISO aren't encoded.

422 ProRes HQ is an encoded colourspace even if it's LOG. 4:2:2 ProRes is most certainly NOT a RAW format. Compression is another issue as well that is not really tied to RAW.

jb

Re: BMCC Killer?

PostPosted: Tue May 07, 2013 11:26 pm
by Aaron Scheiner
lol @ keep it classy ;)

Another interesting observation in DR was that the 5D MK III changed very little from its 3.5-year-old predecessor, the 5D MK II. Our measurements actually revealed a regression by two-tenths of a point...
from DXOMark's 5D Mark III review/sensor comparison.

And from the same page :
The 5D MK III scored 11.7 EVs (exposure value) compared to the D800 score of 14.4 EVs. Both reached their maximum EV score under low ISO settings of between 50 and 100 ISO. It was interesting to note that Nikon’s dynamic range continued to improve between ISOs 50 and 1000, while the 5D MK III’s performance began to stagnate.
Needless to say, 11.7 isn't the BMCC's 13.

The SNR on the camera appears to be even worse than a Sony PMW-EX3 (rated at 54dB) :
Image

A blog claims that :
…the (EX3) S/N Ratio 54 dB (Y-typical) … is about 9-bit precision.
don't know how true it is . So if 54dB is worth 9 bits of precision, the 5D MKIII has ? how many useful bits (when considering the SNR) ?

Compared to both of these cameras the BMCC does 88dB SNR ?

And then there's still the aliasing*... and the fact that you have to hack it** to get this functionality. I have nothing against hacking, in fact I love pulling things apart and have written code for my DSLRs, but I consider hacked firmware to be a nice to have... it's better to have sexy functionality out of the box rather than having to add it yourself.

*Aliasing; Even at 3.5k you still have to get that data out of a 5.76k image... you can't downsample it properly... so you'll either have to crop or line-skip.

And after all that it's more expensive than a BMCC :P . Also, it's logical that it's SNR/DR should be worse than a BMCC because the BMCC's sensor is actively cooled (that's how they get that amazing noise floor).

It's a complex subject.

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 3:07 am
by John Brawley
DXO are honestly a waste of time.

This is the same company that rated the Leica M9 the worst full frame camera they ever tested.

If you've had the pleasure of shooting with an M9 you'll know that it's specs over-ruling your eyes.

http://www.stevehuffphoto.com/2013/03/1 ... ults-ever/

http://petapixel.com/2013/03/08/dxomark ... befuddled/

I believe the results you've quoted are also for stills. Which are nothing to do with what kind of DR theses cameras do when shooting video.

So, DXO produce questionably useful test results in my opinion anyway, and the results you're quoting are with shooting stills, not video.

jb

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 7:04 am
by Rudy Satria
After read all your comments, i assume that BMCC still has better RAW quality rather than DSLR. Correct me if i'm wrong. In conclusion, do you think DSLR could achieve better image result rather than BMCC? or its just a matter of taste?

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 8:12 am
by John Brawley
spike wrote:
John Brawley wrote:DXO are honestly a waste of time.
jb


Wow Mr. Brawley you sure are an opinionated person, aren't you now. As we have apparently lowered this forums tone with flinging insults to people that are not even here to defend themselves, can I now please throw an insult at someone that IS here? Sorry I can't do as you have done as I think it's despicable... so here it is!

Mr JB, you are not a good cinematographer and randomly speak utter rubbish on a daily basis about BM. Do you remember when people were saying BMD were always late to market and you were backing them up saying they were not? Yeah, rubbish. Do you remember when you said BMD are trying their hardest? Yeah, rubbish. Do you remember when you said the BMCC was delayed due to thunderbolt certification? Yeah, rubbish as it was always a manufacturing issue wasn't it now?

Anyways, it's not nice to fling insults at people. Is it? Now DXO may well be what you say, but is it really worth someone in your position outright saying what you said? No. Maybe get them over here to have a discussion before bad mouthing them.


Actually I expressed an *opinion* about a company that provides a service that tests and measures sensors, not a person.

I backed up my opinion with my reasoning and links to others that feel the same.

I also pointed out why i thought it wasn't even relevant to the discussion.

I'd suggest you're the one flinging insults.

JB.

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 8:31 am
by Aaron Scheiner
JB wrote
I believe the results you've quoted are also for stills. Which are nothing to do with what kind of DR theses cameras do when shooting video.
Unfortunately no one seems to have actually done tests on the SNR with this camera in video mode (because it's primarily a stills camera)... and even if they had it would show tainted results because you're talking about uncompressed YUV output from the video buffer which would potentially show wider DR than H.264 output.

And JB, why do areas with 100% zebra on the them on the BMCC expand when the ISO setting is switched from 200 to 1600 ? It implies that sensor clipping has increased due to a change in ISO.

-edit- It appears that no one else has actually done DR/SNR tests on the 5D MKIII, in video or stills.

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 10:39 am
by Tom Sefton
rudysatria wrote:Will 5D Mark III with Magic Lantern could beat dynamic range of BMCC ?
Please let me know what you think about this video

Thanks :D



No.

The Magic Lantern plugin for this feature at present only works for 48 frames. That's 2 seconds. The storage media and processor in the camera won't be capable of creating RAW footage for a period of time that makes it useful.

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 11:28 am
by Rudy Satria
spike wrote:
John Brawley wrote:DXO are honestly a waste of time.
jb


Wow Mr. Brawley you sure are an opinionated person, aren't you now. As we have apparently lowered this forums tone with flinging insults to people that are not even here to defend themselves, can I now please throw an insult at someone that IS here? Sorry I can't do as you have done as I think it's despicable... so here it is!

Mr JB, you are not a good cinematographer and randomly speak utter rubbish on a daily basis about BM. Do you remember when people were saying BMD were always late to market and you were backing them up saying they were not? Yeah, rubbish. Do you remember when you said BMD are trying their hardest? Yeah, rubbish. Do you remember when you said the BMCC was delayed due to thunderbolt certification? Yeah, rubbish as it was always a manufacturing issue wasn't it now?

Anyways, it's not nice to fling insults at people. Is it? Now DXO may well be what you say, but is it really worth someone in your position outright saying what you said? No. Maybe get them over here to have a discussion before bad mouthing them.


Hi Spike, please don't get me wrong. We are here to discuss and share what is in our mind related to the topic. I'm just a new cinematographer and very stupid, thats why i came here. So if you think that you are smarter, please come with the lights. Just take it easy and stay focus at the topic.

- Rudy

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 11:31 am
by Christian Schmeer
Aaron Scheiner wrote:Needless to say, 11.7 isn't the BMCC's 13.


Comes close to the "Production" camera's 12 stops though :mrgreen:

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 2:00 pm
by John Brawley
spike wrote:What ARE you talking about? Where did I say I was smarter? I was simply implying that anyone can insult people that are not here. It's not polite to do that... I have not the first clue about being a DoP as I direct mostly. I do know what IS and IS NOT rude and I think jb was being rude to a company/site/person that was not here to defend themselves and by the sature he has here, that is unacceptable behaviour. Savvy?


You were the only one to make a personal attack. The OP has asked nicely to bring it back on track so you attack and insult them.

jb

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 2:02 pm
by Aaron Scheiner
Christian Schmeer wrote:Comes close to the "Production" camera's 12 stops though :mrgreen:
:P
To each his own.

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 2:06 pm
by Michael Coviello
spike wrote:
John Brawley wrote:DXO are honestly a waste of time.
jb


Wow Mr. Brawley you sure are an opinionated person, aren't you now. As we have apparently lowered this forums tone with flinging insults to people that are not even here to defend themselves, can I now please throw an insult at someone that IS here? Sorry I can't do as you have done as I think it's despicable... so here it is!

Mr JB, you are not a good cinematographer and randomly speak utter rubbish on a daily basis about BM. Do you remember when people were saying BMD were always late to market and you were backing them up saying they were not? Yeah, rubbish. Do you remember when you said BMD are trying their hardest? Yeah, rubbish. Do you remember when you said the BMCC was delayed due to thunderbolt certification? Yeah, rubbish as it was always a manufacturing issue wasn't it now?

Anyways, it's not nice to fling insults at people. Is it? Now DXO may well be what you say, but is it really worth someone in your position outright saying what you said? No. Maybe get them over here to have a discussion before bad mouthing them.


Spike, I think you're overreacting to JB's comment. I know it's easy to take things out of context on the internet but you're equating JB calling DXO a "waste of time" to "flinging insults to people" and that's not right. No where in JB's post did I ever get the feeling he was being aggressive, mean-spirited, or insulting. He was simply expressing his opinion about a company. A LOT OF PEOPLE DO THIS. Also, he actually gave supporting information for his claim (NOT MANY PEOPLE DO THIS). We're all opinionated people. I think you're nitpicking in trying to call out JB for something he clearly hasn't done. I'm sorry I even took this much time out of my day to respond to something so minute on the internet, but I feel you are doing JB wrong, when he hasn't even done anything.

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 2:10 pm
by John Brawley
Aaron Scheiner wrote:Unfortunately no one seems to have actually done tests on the SNR with this camera in video mode (because it's primarily a stills camera)... and even if they had it would show tainted results because you're talking about uncompressed YUV output from the video buffer which would potentially show wider DR than H.264 output.

And JB, why do areas with 100% zebra on the them on the BMCC expand when the ISO setting is switched from 200 to 1600 ? It implies that sensor clipping has increased due to a change in ISO.

-edit- It appears that no one else has actually done DR/SNR tests on the 5D MKIII, in video or stills.


DR tests are really tricky because it's actually very subjective. At what point is detail in the blacks actual detail or noise ? What some people find acceptable noise, others wont. Hence I've never felt comfortable quoting DR in terms of stops. And why you get such wild disagreement and weird terms like "useable stops" whatever that means. I have a way of doing DR tests that works for me and means I can understand what kind of range I can work with. Everyone has to find their own way. I find pointing cameras at the same subjects is a really good start.

A little old now but you get the idea....

http://johnbrawley.wordpress.com/2011/0 ... nge-tests/


and

http://johnbrawley.wordpress.com/2011/0 ... ly-online/


With regards to Zebras changing....

What I know is that the zebra is always showing you at the sensor level what is happening, independent of ISO....except....

BMD decided to make a slight difference in ISO1600. They actually allocate the bits differently to favour more information in the shadow detail so you'll find that there is a very slight difference between ISO1600 and all the other ISO's.

I never noticed it initially and the difference is small.

Let me know if you find the zebra behaves differently at ISO800 and ISO200

jb

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 2:51 pm
by Trevor Zuck
John Brawley wrote:
TZuck wrote:
Tom wrote:
I have a qualm with #2, there is "raw" and then there is "RAW"

"raw" the word can mean footage straight from the camera, unedited, uncorrected footage
"RAW" signifying a camera/image RAW codec or file, like RED RAW, ARRI RAW, or DNG. all DNGs are an image RAW file, and most RAW formats can utilize compression.

Also really and truly if you want "true" raw you would be looking at a bayered image.

and 14-bit color depth is 4 times better than 12-bit, because math.
(for those that don't know color bit depth is calculated as 2^n colors where n = bit depth. so 2^12=4096, 2^14=16,384)



Couple of things.

The bmcc sensor is more than 16 bit. The RAW sensor data is encoded into a 16bit LIN file which is then encoded into a 12 bit LOG file before becoming a DNG. You'll notice the DNG file unpacks as a 16 bit file in ACR and Resolve.

I'm not sure the distinction between raw and RAW is ratified anywhere. In my books, RAW means WB and ISO aren't encoded.

422 ProRes HQ is an encoded colourspace even if it's LOG. 4:2:2 ProRes is most certainly NOT a RAW format. Compression is another issue as well that is not really tied to RAW.

jb


wow did not know the bmcc sensor was more than 16-bit. good to know. still would like the control ACR gives you with that RAW image, but in Resolve, though i'm sure I'm missing something somewhere.

there hasn't been an official distinction, thus can make it confusing as to what people are really talking about. I've always seen RAW as that super controllable-in-post image file, and raw as just ungraded unconverted footage straight off the camera. Client: "Why does that image look so flat" Pro: "O its just the raw footage. tech/post guy: *confused* "but we shot in ProRes right?..."

yes I would agree 4:2:2 ProRes is not a RAW format, but when it comes to 4:2:2 14-bit DNG I would say it's just as RAW as the other RAWs whether it opens up to 16-bit or 14-bit.


Now if you want to see the tested Dynamic Range of the Canon's in Video mode,
watch this series:

http://www.zacuto.com/the-great-camera-shootout-2011
episode one goes over DR, episode 2 goes over sensor sensitivity.
(11 stops, favoring shadows)

there's also a 2012 one but cant remember if they went over DR or not.

and JB is right DR can be subjective as far as usability goes, and when you have to deal with compression artifacts it can really shorten usability (looking at you H.264)

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 4:49 pm
by Tom
TZuck wrote:I have a qualm with #2, there is "raw" and then there is "RAW"

"raw" the word can mean footage straight from the camera, unedited, uncorrected footage
"RAW" signifying a camera/image RAW codec or file, like RED RAW, ARRI RAW, or DNG. all DNGs are an image RAW file, and most RAW formats can utilize compression.

Also really and truly if you want "true" raw you would be looking at a bayered image.

and 14-bit color depth is 4 times better than 12-bit, because math.
(for those that don't know color bit depth is calculated as 2^n colors where n = bit depth. so 2^12=4096, 2^14=16,384)


there hasn't been an official distinction, thus can make it confusing as to what people are really talking about. I've always seen RAW as that super controllable-in-post image file, and raw as just ungraded unconverted footage straight off the camera. Client: "Why does that image look so flat" Pro: "O its just the raw footage. tech/post guy: *confused* "but we shot in ProRes right?..."

yes I would agree 4:2:2 ProRes is not a RAW format, but when it comes to 4:2:2 14-bit DNG I would say it's just as RAW as the other RAWs whether it opens up to 16-bit or 14-bit.


Now if you want to see the tested Dynamic Range of the Canon's in Video mode,
watch this series:

http://www.zacuto.com/the-great-camera-shootout-2011
episode one goes over DR, episode 2 goes over sensor sensitivity.
(11 stops, favoring shadows)

there's also a 2012 one but cant remember if they went over DR or not.

and JB is right DR can be subjective as far as usability goes, and when you have to deal with compression artifacts it can really shorten usability (looking at you H.264)


For me, when a professional describes a camera being capable of raw video capture or raw stills capture - I take it to mean bayer pattern data - straight from the sensor, no information being thrown away, no chroma sub-sampling, no lossy compression.

Otherwise its easy to call almost anything raw - I could say my 5d Mkii already does raw video - in that the footage straight from the CF card is raw 5d footage. In the end it is down to semantics (cue Mhood's quote).

What this Magic lantern hack is doing, is not providing raw sensor data, or bayer data, but 14 bit images with chroma sub-sampling - a far cry from the flavour of raw being provided by the bmcc. Which is why my original point of this "raw video" not being the same as BMCC raw video and thus not a point on which the magic lantern hack creates a "bmcc killer".

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 4:59 pm
by Rakesh Malik
Tom wrote:For me, when a professional describes a camera being capable of raw video capture or raw stills capture - I take it to mean bayer pattern data - straight from the sensor, no information being thrown away, no chroma sub-sampling, no lossy compression.



That's what RAW means. Capitalizing it really just allows you to be clear that you mean raw sensor data, avoiding confusion that could result from contexts such as telling a client that, "It's just raw footage from the camera, not the final product."

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 8:27 pm
by rick.lang
Tamerlin wrote:
Tom wrote:For me, when a professional describes a camera being capable of raw video capture or raw stills capture - I take it to mean bayer pattern data - straight from the sensor, no information being thrown away, no chroma sub-sampling, no lossy compression.



That's what RAW means. Capitalizing it really just allows you to be clear that you mean raw sensor data, avoiding confusion that could result from contexts such as telling a client that, "It's just raw footage from the camera, not the final product."


English can be an infuriating language in which the meaning of words changes significantly depending upon the context. Your example is perfectly correct in the context it is given: raw milk, raw footage; meaning non-pasteurized milk or unprocessed video, etc. However when we see words using all uppercase, it is generally accepted that the word is an acronym taking its spelling from the first letter of each word. IBM means International Business Machines, ACR means Adobe Camera Raw. RAW is not an acronym and should be written in lower case within a sentence such as the BMCC shoots uncompressed raw. I used to capitalize 'raw' too in order to make it clear, but it is not proper English.

Rick Lang
Sent from my iPad using Tapatalk HD

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 8:33 pm
by Rakesh Malik
rick.lang wrote:
English can be an infuriating language in which the meaning of words changes significantly depending upon the context. Your example is perfectly correct in the context it is given: raw milk, raw footage; meaning non-pasteurized or unprocessed. However when we see words using all uppercase, it is generally accepted that the word is an acronym taking its spelling from the first letter of each word. IBM means International Business Machines, ACR means Adobe Camera Raw. RAW is not an acronym and should be written in lower case within a sentence such as the BMCC shoots uncompressed raw. I used to capitalize 'raw' too in order to make it clear, but it is not proper English.

Rick Lang
Sent from my iPad using Tapatalk HD


Agreed. Except for the IBM part; that means Itty Bitty Morons. :)

You are correct, however; even though we do use "raw" in caps as if its an acronym, doing so is a compromise to avoid confusion.

Re: BMCC Killer?

PostPosted: Wed May 08, 2013 9:14 pm
by Trevor Zuck
rick.lang wrote:
English can be an infuriating language in which the meaning of words changes significantly depending upon the context. Your example is perfectly correct in the context it is given: raw milk, raw footage; meaning non-pasteurized or unprocessed. However when we see words using all uppercase, it is generally accepted that the word is an acronym taking its spelling from the first letter of each word. IBM means International Business Machines, ACR means Adobe Camera Raw. RAW is not an acronym and should be written in lower case within a sentence such as the BMCC shoots uncompressed raw. I used to capitalize 'raw' too in order to make it clear, but it is not proper English.

Rick Lang
Sent from my iPad using Tapatalk HD


I'm pretty sure you can "all caps a word" to show emphasis as to differentiate meanings and it still be "proper" English. or you know RAW could be an abbreviation for "Camera Raw Imaging Format" because if Digital Negative can be DNG, then why not?

and to bring this back to the original intent of this thread, if it's more difficult to make a case about how to type RAW and be unanimously "correct" than it is to make a case for the hacked 5D MkIII limited DNG video mode being a BMCC killer then I think you have a pretty clear answer... that being said the 5D has some advantages and it's "look" may be appealing to some.

Re: BMCC Killer?

PostPosted: Fri May 10, 2013 6:19 pm
by bhook
Tamerlin wrote:Except for the IBM part; that means Itty Bitty Morons. :)




Yea...a real pack of Itty Bitty Morons, huh?

Re: BMCC Killer?

PostPosted: Fri May 10, 2013 6:38 pm
by Rakesh Malik
mhood wrote:
Tamerlin wrote:Except for the IBM part; that means Itty Bitty Morons. :)




Yea...a real pack of Itty Bitty Morons, huh?


Did you miss the smiley?

IBM earned that monnicker (along with I've Been Moved) because of how they treated their employees, and how resistant IBM's management have been to things like innovation. I have lots of amusing stories about them... but that's because I grew up in an IBM family. Being a child of two IBM employees, it's hard to be fan of the company. :)

Re: BMCC Killer?

PostPosted: Fri May 10, 2013 6:42 pm
by Christine Peterson
mhood wrote:
Tamerlin wrote:Except for the IBM part; that means Itty Bitty Morons. :)




Yea...a real pack of Itty Bitty Morons, huh?

That's really, really cool.

Re: BMCC Killer?

PostPosted: Fri May 10, 2013 7:26 pm
by bhook
Tamerlin wrote:how resistant IBM's management have been to things like innovation.


Yea...nothing innovative about making movies at the atomic level, huh?

Re: BMCC Killer?

PostPosted: Fri May 10, 2013 7:35 pm
by JerryBruck
mhood wrote:
Tamerlin wrote:Except for the IBM part; that means Itty Bitty Morons. :)




Yea...a real pack of Itty Bitty Morons, huh?


Grow up, Marc.

Re: BMCC Killer?

PostPosted: Fri May 10, 2013 7:59 pm
by bhook
JerryBruck wrote:Grow up, Marc.


Check my avatar Jerry...I'm as grown up as I'm going to get. :lol:

Re: BMCC Killer?

PostPosted: Fri May 10, 2013 11:17 pm
by bhook
Christine Peterson wrote:That's really, really cool.


Amazing isn't it Christine? What a zoom lens! :D

Re: BMCC Killer?

PostPosted: Sat May 11, 2013 3:59 pm
by JerryBruck
mhood wrote:
JerryBruck wrote:Grow up, Marc.


Check my avatar Jerry...I'm as grown up as I'm going to get. :lol:


O yeah? Just so happens that since this thing started I've grown old as the hills, shot right past you if I ever was behind, so I can talk. My hair's fallen out, or the rest of it; teeth wobble back and forth at the touch of a tongue (anybody's); eyesight tobogganing & one lid starting to droop plus something has washed the recall of a long list of proper names right out of my head. That's just some of the things, ones I can remember at the moment. Improper-type words now starting to go. Thought a dictionary would help but how do you look up words you've completely forgotten?! Unless maybe there's one where the definitions come first, bet there isn't one though.

Do I blame anyone or anything for catalyzing this decline? Nope, not me. Latest developments have finally made it really too hard though, is my problem. I could grasp it when there was just one camera that never managed to appear. Now we have three cameras, which we also don't have. Will the new ones appear? Some say Yes, some No. Will the old one show up? Same thing!

There's more: once, for example, the promise of Metabones' SpeedBooster made the teeny-cam at least worth thinking about (for choosy persons) but now the MFT versions have been delayed, maybe by a lot, and the EF-EF by much more than a lot, maybe even like forever, nobody seems to know.

Recurring vision: it is all sinking back into the ooze, these glittering mesmerizing gizmos, the wires into Chile's copper pits, the sensor & processor back to a wave-pulverized chaos of seaweed-choked beach, the artfully molded plastic devolved into stinking crude oil. In their stead the angry fragmented rabble we've come to be -- I'm remembering the shipping thread here -- seasoned pros vs amateur Artists; company butt-kissers against the Commies; grown-ups/whiners & crybabies; Scandinavia/EU Europe; Europe vs USA; heartland vs The Cities, everyone against the gnomes of B & H and not least, surely the greatest victims of all, the MFT crowd who gave up early positions in the EF line to pursue the impossible dream, their cries rend an indifferent sky. Rude against polite. The stoopid against everyone. Why no t-bolt in better PC chipsets? I'll b alright Marc thank u & goodnite.

PS: What's this Marc-with-a-"c" thing? Are there French people in Texas?

Re: BMCC Killer?

PostPosted: Sat May 11, 2013 4:23 pm
by bhook
JerryBruck wrote:PS: What's this Marc-with-a-"c" thing? Are there French people in Texas?


:lol: My Mother wanted to name me Marcquis but my Dad insisted on something more macho...Marcus. I am told Marc-with-a-"c" was a compromise.

About the w8: At least the older you get, the quicker time passes. ;)

Re: BMCC Killer?

PostPosted: Mon May 13, 2013 2:28 am
by Christian Schmeer
Just wanted to post this - they have finally managed to get up to 24fps continuous RAW recording on the 5D Mark III: http://www.eoshd.com/content/10324/big- ... d-mark-iii

Yes the world just turned on its head and I have no idea what the implications for the market are going to be yet, other than that now Canon’s DSLRs, through no action at all from Canon themselves just became a whole lot more serious as cinema cameras.

There’s no doubt the raw images from the 5D Mark III exceed the quality you get from Canon’s much more expensive cameras, like the C300. The 1D C now also has some serious competition from much cheaper DSLRs. I feel Blackmagic could lose most from this though, and will be a shame if that happens. I’ve not yet tested them head to head but the sensor in the 5D Mark III is far more capable than the one in the Blackmagic Cinema Camera in many ways – generating clean moire free images especially in 1:1 crop mode and cleaner high ISO images from a much larger sensor. The BMCC is no longer as attractive to me.

The 5D Mark III is now top dog for video quality, far surpassing the GH2 and GH3. All this is rather tragic, because quite simply, Canon don’t deserve this success. It is entirely of Magic Lantern’s making, and they did it for us free.


Formats: RAW (currently regular DNG not Cinema DNG) and MJPEG

Resolutions:
Full frame mode -
1920 x 1280 (3:2) – great for anamorphic
1280 x 1280 (16:9)
1280 x 720 (16:9)
1280 x 1280 (1:1)

1:1 crop modes -
3592 x 1320 – I get some drop frames at this res
2880 x 1320
2560 x 1080
1920 x 1080 (full HD crop mode like GH2)
1280 x 720 (the lower the resolution the more telephoto the crop)

Re: BMCC Killer?

PostPosted: Mon May 13, 2013 2:30 am
by John Brawley
Christian Schmeer wrote:Just wanted to post this - they have finally managed to get up to 24fps continuous RAW recording on the 5D Mark III: http://www.eoshd.com/content/10324/big- ... d-mark-iii



Is it RAW though ? Isn't it still line skipped ?

jb

Re: BMCC Killer?

PostPosted: Mon May 13, 2013 2:31 am
by Christian Schmeer
John Brawley wrote:
Christian Schmeer wrote:Just wanted to post this - they have finally managed to get up to 24fps continuous RAW recording on the 5D Mark III: http://www.eoshd.com/content/10324/big- ... d-mark-iii



Is it RAW though ? Isn't it still line skipped ?

jb


I am guessing it would be line skipped in Full Frame mode, but not in 1:1 crop mode (?), though I don't know much about sensors.

Re: BMCC Killer?

PostPosted: Mon May 13, 2013 2:35 am
by Rakesh Malik
Christian Schmeer wrote:
I am guessing it would be line skipped in Full Frame mode, but not in 1:1 crop mode?


It's probably doing line skipping for all of its video modes.

Re: BMCC Killer?

PostPosted: Mon May 13, 2013 2:37 am
by John Brawley
Tamerlin wrote:
Christian Schmeer wrote:
I am guessing it would be line skipped in Full Frame mode, but not in 1:1 crop mode?


It's probably doing line skipping for all of its video modes.


And is it actually RAW ??

jb

Re: BMCC Killer?

PostPosted: Mon May 13, 2013 2:45 am
by Christian Schmeer
John Brawley wrote:
Tamerlin wrote:
Christian Schmeer wrote:
I am guessing it would be line skipped in Full Frame mode, but not in 1:1 crop mode?


It's probably doing line skipping for all of its video modes.


And is it actually RAW ??

jb


I am not sure, but if I've understood correctly, they're still getting the images out of the camera's live view buffer. I guess it's not RAW, but "raw".

The raw format is 14bit YUV 422 uncompressed.


The results look incredible though. Will be great for low light stuff...

Re: BMCC Killer?

PostPosted: Mon May 13, 2013 3:02 am
by John Brawley
Christian Schmeer wrote:
I am not sure, but if I've understood correctly, they're still getting the images out of the camera's live view buffer. I guess it's not RAW, but "raw".




Well that's not RAW or raw is it. (what's the difference ?)

If that's what it is then it's encoded video. It's not sensor data. Not knocking what they're doing, but I don't see how it you can call this raw or RAW. It's more like LOG isn't it ?

jb

Re: BMCC Killer?

PostPosted: Mon May 13, 2013 4:27 am
by Csaba Nagy
If they have been able to enable what was once thought to be a virtually impossible feature on the Mark III ( and even potentially other cameras like the cheap 600D! ) BMD should to do the same in regards to features we have been demanding. ( higher frame rates, 2.5k Pro Res, etc. )

Yes, I have read into the details and for now its far from a reliable release ( could even raise some issues that we don't know about ATM ). At the same time from what we know now, it has the some features even the BMCC doesn't. Allowing the use of what I would think is a better "hardware" camera ( larger sensor, Superior low light sensitivity, smaller body, etc..)

This hack could have the potential to change a lot of peoples opinions about the BMCC and the Pocket cinema camera.

Besides Grant even said in an early interview that the BMCC is really targeted at those who used and worked with DSLRs like the MARK III.

There is no way to confirm this hack will ever leave a "alpha" or even "beta" stage before it crashes.
I'm not trying to side with this new development, I'm a grateful owner of a BMCC.

But I like to use this as an example of how hackers ( not canon engineers ) have been able to unlock a capability many people doubted could ever happen. This should be used to leverage BMD to officially give us more features on the new line of cameras, that to this day are said to be not possible. ( features that are still within reason)

Schoolpost,

Re: BMCC Killer?

PostPosted: Mon May 13, 2013 4:46 am
by Christian Schmeer
I still think an SDK would be a great thing for the BMCC/BMPC/BMPCC.

Re: BMCC Killer?

PostPosted: Mon May 13, 2013 4:59 am
by Steve Lee Jean
They're taking sequences of DNG stills aren't they? I'm assuming they're RAW. Even then I'm happy for what the ML guys are doing, as competition is good. The 5d has HUGE pluses over the BMCC, but bottom line for me is, it's still more expensive, doesn't give Resolve/Scopes, and has that utterly annoying 135 sensor.

It's a wonderful stills camera though.

Re: BMCC Killer?

PostPosted: Mon May 13, 2013 5:08 am
by Rakesh Malik
innerspark wrote:They're taking sequences of DNG stills aren't they? I'm assuming they're RAW. Even then I'm happy for what the ML guys are doing, as competition is good. The 5d has HUGE pluses over the BMCC, but bottom line for me is, it's still more expensive, doesn't give Resolve/Scopes, and has that utterly annoying 135 sensor.

It's a wonderful stills camera though.


DNG doesn't necessarily mean raw sensor data, which is what raW (rAw? ;)) means. If the sensor is using line skipping in order to keep up with the demands of video production in contrast to stills, then it's not capturing raw video like the various Black Magic cameras do, or like the still cameras do when capturing stills, for that matter.

Re: BMCC Killer?

PostPosted: Mon May 13, 2013 5:10 am
by Christian Schmeer
Some new demo videos were posted:

"14-bit RAW" (which may not be RAW?) 1080P at 24fps
http://www.youtube.com/watch?feature=pl ... HR92jzWUd4

Regular "I-frame" 1080p at 24fps (Technicolor CineStyle)
http://www.youtube.com/watch?feature=pl ... vCRQ-haYYc

Exposure was the same.