nahua

This is an image of the Canon 24-105 F4L at 50mm on a 5D3 and FS700+Speed Booster blended 50/50. Max aperture set on 5D3 (F4) and FS700+SB (F4 full-frame DOF with F2.8-like light transmission performance).
[attachment=393:FS700+SB_5D3_DOF.jpg]
 
Close up of bokeh:
[attachment=394:bokeh_match.jpg]
 
The FS700 S35 sensor with Speed Booster gives full-frame performance with a 1.09 crop, approximately same DOF and FOV performance as full frame, and effectively 1-stop faster light transmission performance. In this example, F4 DOF for both cameras, with the FS700+SB providing F2.8-like light transmission. Where a T-stop is exactly the same as an f-stop with 100% light transmission, then a lens rated with a T-stop of 4 on the 5D3 would have a T-stop rating of 2.8 with the FS700 + Speed Booster.
 

Sharpness loss at the edge of frame is mostly a function of the lens itself.  It's only apparent to the smaller sensor after using this adapter because it could never see it before, it only saw the "sweet spot".  In point of fact, sharpness at the edge of frame is better through the adapter on a smaller sensor than the same lens mounted to a FF camera.
 
Seems lots of people still aren't bothering to really investigate or grok what they've read so that they can add meaningful discussion.
 
edit: it cracks me up that, even though the adapter IS performing a kind of magick, its critics now feel compelled to criticize the magick itself as if it just weren't miraculous enough for their standards.... 
 
"Yeah, you turned water into wine...but you didn't raise anyone from the dead.  Fail.  You suck."

Love Spielberg's reference back to Mifune when talking about Eastwood's Man With No Name movies.
 
For those who don't know the history:
http://en.wikipedia.org/wiki/A_Fistful_of_Dollars
http://en.wikipedia.org/wiki/Yojimbo_(film)
http://en.wikipedia.org/wiki/Man_with_No_Name

looks better than most cheap steadicam samples you see online.  i think this would be perfect if you used the same technique alongide a cheap handheld steadicam as well, the camera and your processing will smooth out all the slight issues with the cheaper stabiliser i imagine.    

I can't figure out if you were laughing maniacally while writing this piece, or not.

It may be a little early but I think it was lafilm poo-poo'ing the very notion of non-professional cameras and solutions by smaller filmmakers making any sort of impact at Sundance or the Academy or other such hoity-toity institutions that only allow the right kind of people with the right kind of equipment to play their reindeer games.
 
LOL

@Axel thanxs I'm already working with MPG streamclip so I try that setting.
 
@Gravi... Yesterday I went to a film festival and I bet all those films where digital so ...
I also do use the word film in some cases. Let's not talk words though and stay o the topic, Who cares if we call it video making or filming. Everybody does understand what we're talking about!   ;)

See all this crap is why I usually avoid forums. Thanks Burnett for answering my questions.
 
Slightly less thanks to OverCranked for initially answering my question then going on a rant about something un--asked for.
 
Props to Tony for his always impeccable spelling.

LOL, you obviously haven't read the actual tests being done (click the link above).  Resolution with the same lens on the adapted NEX is measurably higher than a 5DmkII in the center and averaged over the entire lens, only dipping slightly in the corners.  Right there, sharper than still the best value in full-frame motion picture shooting, so good still that Canon had to take it off the market to make their newer, more expensive cameras more attractive to Canon buyers.
 
Also, you haven't read how Slumdog Millionaire was shot with those cameras either. 
 
Please, continue to punt though.

Strictly speaking, you are getting a wider and faster lens, that produces an image on a smaller sensor that is equivalent to the image produced by that lens on a larger sensor. If the metabones adapter were optically perfect, the two images would be mathematically indistinguishable. Creating a lens that is both wider and faster is a big deal. Video readout from the smaller sensors is currently more complete (higher resolution, less aliasing) than the video readout from larger sensors, which means the ultimate image coming from the smaller speedboosted sensor will be superior to the ultimate image coming from the larger sensor.

I don't exactly understand what you're really trying to say here? This is how many undiscovered directors have got their start. Crews of 100+ are not WHY films are good. Good films come from talent, not crew-size and money. Well produced films typically HAVE allot of money and crew, but it's not "why" they are good.
 
Seems like you're bitter about something. Threatened by the innovation and falling costs/sizes of production? But then again, maybe some people should be... the days of hacks who could never make it in the creative aspects of film-making, yet still getting paid $1500 a day to hand people lenses, are coming to an end.
 
Technology is finally freeing the creatives from the "brush-holders" of the industry. This is great! Should be celebrated!

@lafilm
 
Barrel distortion is a function of the lens.  If you see it with this adapter it would show up on a FF camera.  Next.  (forgetting the fact that strict prohibition of barrel distortion of any magnitude is kinda silly outside stodgy industrials and ENG stuff that tends to be flavorless in every way anyhow...).  Oh, and your "not as sharp" comment is 100% incorrect.
 
Your second comment makes no sense at all.
 
Wide options exist for M4/3 and BMCC type films.  I seem to remember wide shots in Slumdog Millionaire.  Oh look, it's even smaller than the BMCC.  Oh look, it won the Academy Award for Best Cinematography.  Who'dathunkit?  Not folks that handicap based on a spec list and their preconceived notions instead of the observation of reality and results.

modern nikon is probably the most unexciting thing in the world.  When they get their focusing rotation round the right way I am sure people will get more interested.  

not really cynical i reflect what is going on in this weird post modern world.
i reflect using bad bad english grammar : )


on the subject of your adapter i would give it 2 weeks before canon baliffs are at your door with a warrant and an nda contract.
you will sign stating that you have never owned or do not understand the term speed booster.
they may not even get you to sign they might just erase your memory on the doorstep.
and programme your brain to buy a nice big new canon camera.

mk ultra tavistock institute very easy to programme a mind.
all you wanted was a little adapter and you ended being a mind controlled assassin.
get rid of it asap i beg you : )

GH2 will be getting the same field of view as the 1DC shooting in 4K mode.  That's very exciting.
 
Where I'm excited for s35 stuff, like the bigger dog cameras, is it means we'll potentially be seeing a return of an aesthetic that doesn't exist in current motion pictures, in the theater.  Outside the occasional indie shot on a 5D, like Rubber perhaps, the cool, small-format, experimental stuff is going to be limited mostly to online and festival venues.  
 
Applying this technology to the likes of a Genesis or Alexa or RED means we could see a new era of epic aesthetics show up in the next mega-million tent pole picture (that odds are would have been made in pedantic S35 spherical otherwise, unless it was from one of the few filmmakers still out there with enough interest and enough pull to insist on anamorphic).  
 
There's only a few directors, like Nolan or PT Anderson maybe, willing to dust off one of those Smart Car sized large format film cameras to shoot in 65mm or 8-perf 35mm.  Something like this means digital for big budgets could finally aspire to something beyond just mimicking or replacing the most generic, least quality 35mm format that's been employed for major motion pictures since the introduction of color nearly, and without major new investment either.  Everybody wins.

By this time I was already frustrated with you.  You want reality, everything after the first paragraph in your first response to this thread was bullshit and everything I've written after it was a response to said bullshit.  Now, if English is a second language or something then I'll say, "sorry, I'm being too hard on you, you're handicapped from making a good point here."  
 
Otherwise, you failed to address the actual query of the original author so that you could go on some kind of rant, either original and from your own head or cobbled together from other stuff you read on the internet somewhere (which may have also been a bloated, non-answer to someone's specific topic), because you wanted to participate more than you could actually contribute.
 
There.  Face your realities.  Or would you like me to give you line item "WTF are you talking about?" replies to your first post here?  I was going to do that but then thought it might come off rude so I just offered a gentle nudge to stay on topic instead.  My mistake.  There's a reality for me.

@Blanche
View some tutorials on color grading with the on-board tools of FCP X (skip the ones that suggest you can do without the scopes or use automatic functions or match colors). I disagree with every single post that says you create one preset and paste it. Why? Because your already good images can be so much improved by thorough grading. Do it as the last step, after editing is finished. It should take a minute or two for every clip.
 
NEVER 'share' your film directly 'for vimeo', because that means, that the inferior Quicktime H.264 encoder is used, and the film is simplified too much. The fade in at the beginning shows 'temporal banding'. You can avoid it completely, if you put a very subtle amount of grain on top of this clip, export as ProRes master and encode an x264 mp4 (for the vimeo upload) with the free x264 encoder from the high quality master.

Wow that is some really great footage!  Must have been quite a trip!
 
I guess the question is how much post-processing work you can handle.  At the standard 0,0,0,0 settings, you will get moire and aliasing.  You also get banding in the skies.  However, if you have Adobe Premiere CS5.5 or CS6 you can do some simple corrections.  The most common is Fast Color Corrector and Sharpen. 
 
Fast Color Corrector can add contrast, saturation, and you can set levels.  The best thing is it can playback in realtime.  When you lower the contrast and saturation in camera, you get a "flat" desaturated image that you can work with in post.  But I suggest not lowering it too much otherwise there might not be enough information to bring back in post.  I've seen people go from -2 to -4.  I've been using -5 contrast and -3 saturation, and then I can play with gamma +.3 to +.5, and saturation 130% to 150%.  I suggest you test it out and see how much you can work with. 
 
Moire and aliasing are based on the in-camera sharpening.  I have dialed sharpening to -5 and it helps a lot.  In fact I rarely if ever see anymore moire or aliasing.  You can add sharpening in post with the Sharpen filter.  I suggest adding anywhere from 10-30.  But again, when you add sharpening even in post you risk adding aliasing, especially to sharp edges like lamp posts, roof lines, power lines, etc.  But at least in post you can render it and see how much you want to add.  If you do it in-camera, it's burned in and can't be removed.  Again depends on how much you want to do in post, but at least you get the flexibility to decide what you like.
 
Banding is unavoidable with an 8bit codec.  One thing you can do is to use the 1080P60P (or 50P in PAL) 50mbit codec.  This codec is better in a lot of ways than the 72mbit All-I codec.  1) digital noise characteristics is better 2) you can still do work in post 3) more recording time 4) great for slow-motion.  The only thing you can't do with the 50mbit codec is use EX Tele mode, that only works with 72mbit.  But for everything else the 50mbit codec is best (even in 30P or 24P).
 
As far as Noise Reduction goes, by dialing it down you get more noise in the image, but you also get more detail.  This also helps reduce banding.  I use Neat Video to reduce noise (very reasonable price at around $100).   It does a really good job of reducing noise and bringing in detail.  The penalty is that it requires HUGE processing time.  An hour of footage is easily 40 hours of rendering time (at least for my Mac Tower).  This is a significant consideration.  However it also enables you to shoot ISO 1600-5000, which can help in very low light interiors or night shots.
 
Another way to reduce banding is to use some kind of Film Grain plug-in.  I'm just starting to use this for sky or sunset shots.  You can look at my tests here:  https://vimeo.com/57102903
 
I think your footage is great, and I really like the results from your 12-35mm lens!  The OIS is awesome!  I know you have some jitters and movement, again this can be removed in post but at some cost (processor and $$).  Most common is Adobe After Effects and using the Warp Stabilizer.  But I have another suggestion.  I use a Joby Gorillapod.  This can be used as a mini-tripod as well as for stability.  And since you cycle and need things small and lightweight, I think it's perfect for you.  You can clamp it on your bike, chairs, poles, and you can also form it around your arm for stability.
 
http://www.bhphotovideo.com/c/product/680973-REG/Joby_GP8_BHEN_Gorillapod_Focus_Ballhead_X_Bundle.html
 
Sorry this is a lot to cover, I hope it helps!

What it means is the death of sacrificing video performance just to get the look of full frame.  None of the full frame options currently available, or announced, offer anywhere near the best motion picture performance.  
 
$1000 USD camera (why, when you could buy a GH2, but whatever) + $700 adapter cost = video performance that totally curb-stomps a $1700 full frame camera, or a $2975 full frame camera, or a $6728.99 full frame camera, or a $12,728.99 full frame camera.
 
That's what this means.

Don't forget the fact that every lens you buy suddenly has a double purpose (with and without speedbooster). And you don't have to buy f/1.4 lenses anymore, you can pay less than half for a f/2 and have the same light gathering capabilities. Even a dirt cheap f/2.8 prime makes a nice f/2.0.
 
For example: I'm interested in buying a reasonably fast (at least f/2) standard lens (50mm equivalent) and a wide angle (35mm equivalent) for my GH2.
 
I could buy a Canon FD / Nikkor 24mm f/2 to get the 50mm equivalent, that would cost €300 to €400 on eBay.
To get a 35mm equivalent, I'd need a 17mm. I could buy some crappy, rare, huge 17mm f/3.5, but that's not fast. The only choice is a Voigtlander 17.5mm f/0.95, for €1200!
 
With the Speed Booster, I could buy a Nikkor 35mm f/2.8 for €100 to get 50mm f/2 and a Nikkor 24mm f/2.8 for €200 to get a 35mm f/2.
 
For the price of the Voigtlander 17.5mm, you could buy the Samyang 24mm f/1.4 and the Speed Booster, and get pretty much the same effect: 35mm f/1.0. Plus the added bonus of all your lenses suddenly behaving the same as their much more expensive counterparts.
 
It is not cheap. But if you are in the market for a a few fast standard/wide lenses like me, then the Speed Booster instantly pays itself back. Plus you get options that aren't even possible without the adapter.
 
To me this actually is the most exciting product I've seen in years. Yes, we have seen a lot of technical developments on the camera side. Better sensors, cheaper camera's... Nice, but glass always stays the same. Nothing changes physics. Good glass is expensive. In a way it's nice that this doesn't change, lenses retain their value.
 
But now Metabones manages to come up with something that turns this very fixed part of photography/filmmaking upside down. If I were Sony (CaNikon are too slow anyway..) I'd instantly buy out Metabones and release this technology inside a mirrorless camera (with A-mount, just fit the element in place of the SLT-construction and use Phase Detection pixels) That would be the break trough cheap 'fullframe' camera and stir the market :)

Getting a certain look in-camera always was the DPs job, be it for movies or documentary. Not saying that things shouldn't progress and there are A LOT of situations I`m happy to be able to shoot flat. But for Blanche´s situation, travel documentary, I´d opt for the best in-camera look, expose properly and roll.
And then you can really say "I did that." ;)

According to the MB whitepaper*, the SB will increase image quality (measurable MTF) in the same way an extender reduces image quality. If accurate (in mass production- the math/physics makes sense), that means that not only is FF possible and -1 effective f-stop, but also higher IQ. I ordered the SB and and FS700 to test. If it works as advertised, the FS700+SB looks to be the most affordable/versatile pro camera right now (S35 and FF, very low light, multiple frame rates, super slomo, true 1080p pixels). C300 has better ergonomics and perhaps a more bulletproof internal codec, but has far less features and costs $16k.
 
* http://www.metabones.com/images/metabones/Speed%20Booster%20White%20Paper.pdf

Speed Booster at 35mm F1.0 on the Sony NEX 7

With the Metabones Speed Booster APS-C and Super 35mm sized sensors have taken a barrelling charge up to the door of full frame.

Does it blow the doors off?

personally, i'm not a big fan of the warp stabilizer. it can end up causing some really weird warping and jelloing (which technically i suppose is working correctly, but i don't like it) in more shots than expected; if i need to do some stabilization i like using the motion tracking in AE. but that's kind of moot since OP is using final cut.
 
 
1) it is and it isn't a 'lesser' HD. the difference is in the ipb frames, frame rate and bitrate. the all i hd mode means every frame at 24fps is its own individual 'picture,' whereas other modes there are predicted frames between the keyframes. this allows more compression of the video. for the most part, it's hardly noticeable, especially at 60fps. frame rate, 24fps vs 60fps, pretty obvious. if you take your 60fps footage and conform it to 24fps, you have some really nice ~40% slow motion. and with the lower bitrate as well, as nahua said, it will take up less space on your memory card (helpful for #4).
 
2) it looked like there was a little bit of banding in the opening fade in from inside the tent. though i think it was only exacerbated by the fade in.
 
3) i shoot with a gh2, but i have the noise reduction as low as possible, no matter what i'm shooting. if i get into the higher iso's, i just make sure i use a faster lens (or add lights to the scene if i can) to help lessen the amount of noise in the darker areas of the image. and if i go the whole way to 12800, the only way to make it really usable is to shoot it in b&w; and even then, bumping up the NR does very little.
 
4) the image flatness/color profile should remain near enough the same between the AVCHD and h.264 video

https://vimeo.com/57306424