Blackmagic Forum

http://cheesycam.com/rokinon-85mm-t1-5- ... gic-cinema

Just saw this one on my cellphone.

Reading the comments a lot of people are saying the BMCC footage doesn't look sharp and has less dynamic range then the dslrs Emm and Olivia used on this test.

Haven't got the chance to look at it on a proper screen but could it be that they're just perceiving the aliasing and moire in the dslr footage as better resolution?

What do you guys think happened (or is it happent) here?

All those cameras have "picture profiles" that are designed to give the image a bit of "punch" and have probably got sharpening left on. The BMCC has no sharpening in camera, you would do that in your grade. Would be interesting to see the raw footage rather than the compressed YouTube footage.

I did like the image from the GH3 tho.

This is odd. Yes, the BMCC appears to have less dynamic range in this test. The GH3 and the BMCC have blown-out highlights. Clearly, the BMCC is unlikely to have blown-out highlights if exposed and graded correctly. The Canon's looked great to me, in comparison in this video. Either the lighting or the color grade (or lack thereof) were off on this.

I also agree with Adam's comments about sharpening.

They filmed in Prores Video mode.... no more comments

AndreasK wrote:They filmed in Prores Video mode.... no more comments

Shouldn't even the prores footage have better sharpness and dynamic range then the dslrs, because of the higher bitrate and bitdepth?

Didnt see any mention of an ISO. If they were fluctuating ISO's then the test is completely invalid...at least when it comes to DR and side by side comparisons....

Max M. wrote:Didnt see any mention of an ISO. If they were fluctuating ISO's then the test is completely invalid...at least when it comes to DR and side by side comparisons....

Agreed. I think user "error" is likely here in the sense of their settings. Otherwise, I actually like their demonstration of the lens (which is their purpose), though it seems targetted at a consumer level.

I watched the video.. I also know that Olivia and Cheesycam just got the camera a day or so ago. and this very well might be the first time they shot video with it. there was no color correction or sharpening in post. But Emm, the camera op, mentioned that she had an idea what happened that made the bmcc look the way it did.. and was going to 'try' something. So it's possible there was something she did wrong in editing or ingesting the video. (she might have transcoded it to something weird)

Emm has posted something again.

Wasn´t prores sharper when John Brawley shot something in () that? Or is that just me...not really impressed by this:

Was the difference between raw and prores always this big?

so the glass makes it or breaks it ?

Margus Voll wrote:so the glass makes it or breaks it ?

Isn't it that that shouldn't matter that much since BMCC uses the center 'sweetspot' of most lenses?

I'd like to know how those Rokinon and Samyang compare to other primes. The 24mm might be promising. http://www.bhphotovideo.com/c/product/8 ... NE_ED.html

Heh its funny how they use T instead of F, sounds like they are trying too hard to sound like film. But I'd still like to know how the glass holds up.

Complete mishandle of BMCC, if they are going to compare these together they need to normalize, you can use DSLR with profiles (including surely sharpening) then take the BMCC as is. Additionally, I think someone simply forgot to check the focusing

MatthijsLiethof wrote:Emm has posted something again.

Wasn´t prores sharper when John Brawley shot something in () that? Or is that just me...not really impressed by this:

Was the difference between raw and prores always this big?

1. Brawley is a pro DP with tons of XP, which helps quite a bit.
2. Brawley's footage is graded, Olivia's is pointedly not.
3. Brawley was shooting between takes on sets. Sets are lit. And with God as gaffer, dusk is almost always beautiful.

danallen82 wrote:I'd like to know how those Rokinon and Samyang compare to other primes. The 24mm might be promising. http://www.bhphotovideo.com/c/product/8 ... NE_ED.html

Heh its funny how they use T instead of F, sounds like they are trying too hard to sound like film. But I'd still like to know how the glass holds up.

Hmm well it is a Cine lens so I can accept using the proper term "T" stops.

Some folks found the 24mm Cine to be wonky.

I was going to dive in but after a few bad reviews from real world users, Ill find something else in that size.

To add I have the Cine 35mm and 85mm.

Love the 35mm and still figuring out the 85mm.

Using these along with a Tokina 11-16mm on my Scarlet.

MatthijsLiethof wrote:Emm has posted something again.

Wasn´t prores sharper when John Brawley shot something in () that? Or is that just me...not really impressed by this:

Was the difference between raw and prores always this big?

the footage that Olivia had shot is fine that is more less how raw look most the time. Yes it could be alot sharper so I tested that prores footage she uploaded and did a quick color test in avid not even resolve and it looked ten times better then what most people were posting.
main facts are most people are new to raw or a log style footage, for those who shoot on dslr's and use cinestyle are more familar with this look and are able to work with much better to get a better dynamic range then standard

I was thinking about relative sharpness not the look. Of course raw is raw but softness in image
is not so good if compared to JB footage.

There may be ton of factors doing it but it is most probably the glass.

I have considered 27-70 canon L glass mark II and 70-200 L mark II for that reason.
They look on charts pin sharp.

And of course cheap glass can not be as sharp as the more expensive one ?

Olivia's sample prores color corrected with colorista 2 + little contrast + little vibrance

just a single frame

http://ryushare.com/vp0ne2yhaile/olivia.png

sharpened with Sapphire Blur+Sharpen

same frame

http://ryushare.com/yiela231pny7/olivia2.png

sorry but I cannot take anything usable from a 176Mbps 4:2:2 prores

malbrand wrote:... sorry but I cannot take anything usable from a 176Mbps 4:2:2 prores

I don't understand what you mean by this, regrets.


-

at beginning of this post, people were complaining about Olivia's comparison with gh3 and canons, but, from my experience, when it comes from compressed mode, in this case 176Mbps 4:2:2 prores, i could not take anything that make it looks better than other cameras, color correction and sharpening it "in post" was not enough. thats what I mean, Peter.

malbrand wrote:at beginning of this post, people were complaining about Olivia's comparison with gh3 and canons, but, from my experience, when it comes from compressed mode, in this case 176Mbps 4:2:2 prores, i could not take anything that make it looks better than other cameras, color correction and sharpening it "in post" was not enough. thats what I mean, Peter.

Thank you for that clarification.

This particular BMCC ProRes clip by Olivia's camera operator looks odd to me, too, but I'm not sure why. That's why I've asked them about it in the comments section on the Vimeo page. Hopefully they'll respond to my questions:


There are what I think are good-looking examples of BMCC ProRes footage available, such as by James Tonkin & Den Lennie:
viewtopic.php?f=2&t=3385




... and Simon Beer:


... and Greg Ferris:


... and John Brawley:
viewtopic.php?f=2&t=374
http://johnbrawley.wordpress.com/2012/0 ... a-footage/
http://johnbrawley.wordpress.com/2012/1 ... rty-blues/

It appears that BMCC ProRes footage can look quite good if shot with care.

But as you know, many factors might contribute to a particular clip looking not-so-good, such as focus, lens sharpness, filter quality, white balance, aperture-related diffraction, and so forth. I don't know if any of those were a factor with Olivia's sample clip. Cheers.

I think it's less about the gear and more about the composition that makes this one hard to do anything with.

It's not that the shot is "wrong", it just looks like a boring ENG shot with no photographic thought applied.

Alas that I do not have my camera so I can stop relying on other people to provide footage and do my own tests!

I agree that the composition was a bit lacking..
I had a thought that might explain the poor quality of the BMCC prores clip from Olivia.

I know that when you upload a clip to youtube that is encoded to a higher bit rate than the suggested H.264 5000Kbit/s they encode the clip using a higher compressed codec than if you use the suggested settings. Does vimeo do that as well? If so that might explain why.. if so, If you uploaded a 175mb/s proresHQ clip.. they'll compress the heck out of it making it look way worse than if you encoded to the recommended H.264 5000Kbit/s..

Just sayin'

Bill Rich wrote:Does vimeo do that as well? If so that might explain why.. if so, If you uploaded a 175mb/s proresHQ clip.. they'll compress the heck out of it making it look way worse than if you encoded to the recommended H.264 5000Kbit/s..

Just sayin'

hi Bill, the version available for download on vimeo is ProRes 176Mbps 4:2:2 , it is not their fault, it is 922MB of video that I couldnt make look better than a dslr with picture profile, maybe is my fault, maybe it requires some magic (black magic) to extract something from it. best regards

There's lots of examples of "thought about" shots that make the ProRes very attractive as an alternative.

malbrand wrote:

Bill Rich wrote:Does vimeo do that as well? If so that might explain why.. if so, If you uploaded a 175mb/s proresHQ clip.. they'll compress the heck out of it making it look way worse than if you encoded to the recommended H.264 5000Kbit/s..

Just sayin'

hi Bill, the version available for download on vimeo is ProRes 176Mbps 4:2:2 , it is not their fault, it is 922MB of video that I couldnt make look better than a dslr with picture profile, maybe is my fault, maybe it requires some magic (black magic) to extract something from it. best regards

Perhaps you're missing the point I'm making: I think there's something about the way Olivia's camera operator shot that clip that makes it look "off" (at least to me).

You typically can't make bad video look good if it's too far "off".

As can be seen by the other ProRes clips I linked to in my post above, BMCC ProRes footage can look really, really good -- better than typical DSLR footage. It mostly depends on how it's shot, not how it's graded in post.

Until Olivia's camera operator answers the questions I asked on the Vimeo page we can't know what's different about their clip.


I don't think there's an issue with the BMCC. It's obvious that BMCC ProRes footage can look great.

I agree with Peter it almost look as if the camera was pull right out of the box and some one dropped in a drive put on a lens and started to shoot. there was no thought behind the clip there is no real dynamic range to display or anything. it cool to mess around with the clip for people how have free time but its a horrible example of what the camera can do

my .2

Perhaps its that the BMCC has a far higher native iso of 800 compared to the 160 of a 5D,

They were shooting them all at the same aperture with the no ND filters.

If they were using iso 100, then that means that the BMC is being overexposed compared to the other cameras by between 2 and 3 stops. It also looks slightly off focus.

Add 0.9ND and it probably would have looked great.

malbrand writes,

hi Bill, the version available for download on vimeo is ProRes 176Mbps 4:2:2 , it is not their fault, it is 922MB of video that I couldnt make look better than a dslr with picture profile, maybe is my fault, maybe it requires some magic (black magic) to extract something from it. best regards

Thanks for your response Malbrand!
The fact that they uploaded a ProRes 176Mbps 4:2:2 file to vimeo actually resulted in vimeo having to add far more compression in the encoding process resulting in a poorer quality video. Plus I believe the focus was a touch soft on Olivia.

Higher bit rate upload = highly compressed online video
suggested bit rate upload = less compressed better quality online video

Cheers!

Bill Rich wrote:

malbrand writes,

hi Bill, the version available for download on vimeo is ProRes 176Mbps 4:2:2 , it is not their fault, it is 922MB of video that I couldnt make look better than a dslr with picture profile, maybe is my fault, maybe it requires some magic (black magic) to extract something from it. best regards

Thanks for your response Malbrand!
The fact that they uploaded a ProRes 176Mbps 4:2:2 file to vimeo actually resulted in vimeo having to add far more compression in the encoding process resulting in a poorer quality video. Plus I believe the focus was a touch soft on Olivia.

Higher bit rate upload = highly compressed online video
suggested bit rate upload = less compressed better quality online video

Cheers!

The download link it to the original file, not a file recompressed by Vimeo.

danallen82 wrote:Heh its funny how they use T instead of F, sounds like they are trying too hard to sound like film. But I'd still like to know how the glass holds up.

Well it is their Cine lens and if the lens is a true cine lens and has been celebrated and tested the T stops would be the correct measure.

T stop is a more accurate measurement of light passing through a lens than an F-stop because it takes into account light being absorbed by the lens. So that value would also be of more value to a cinematographer.