Why KineRAW MINI could be your next camera

[/p/]

You're clearly not cognizant of how Japanese Industrial Complex works. There are lost of books out there for that.
Basically, they pretend to complete. They actually share a lot and as we already know, fabrication is typically single sourced and multiple 'badged'.
The challenge that all the camera manufacturers have is the same- They have a difficult time parsing the marketplace and therefore of rationalizing thier roll-out roadmaps. That and the fact that the tech is shifting so quickly.

What I believe needs to happen soon in Japan is brand consolidations ( ie- mergers). There's too much duplicationof effort and I know that the Japanese know that duplication is a WASTE. Once we have brand consolidation, then you'll see the opposite of competition as companies will be reduced to two at max within each market segment. Sony/Olympus, Nikon/Canon ( Panasonic?)

BlackMagic is a fringe operation. They remind me of Flip Video ( a brilliant product before they disappeared) Flip was as revolutionary and as quirky as BlackMagic. CISCO bought it, marketed it and in 2011, Killed it outright.

I suspect this is what might happen to BlackMagic, though I doubt anyone will buy it just for Resolve ;)

 

Blackmagic is the company that does the buying out....................................................................... The problem is people think BMD are some new company that decided to bring a little nifty 2013 version of the "flip video" cameras and are willing to be bought out by some other company.. 

[Brellivids]

Kineraw mini at say 2500 dollars .. would be a good price point.

When considering the 4K path .. I understand that you need the 290$ module for that. Then later the ~3300$ box

 

Some calculations .. I could not help my self :D

 

 

BMCC - 1780 euros

Chinese 12v 6800mah LI-ION battery - 25  euros each x 2 = 50

1TB Samsung 840 EVO - 500 euros

 

=  ~  2330 euros.  (get's you 12bit at good enough resolution RAW + Resolve 9 )

Canon mount or passive m3/4..    Only problems in my book are the multiplier and large size along the poor form factor.

 

 

BMPC  880 euros

Chinese Nikon battery 4 euros a piece.  x 6 =  24 euros

Speed booster when available  550 euros

Sigma 17-35mm  1.8   750 euros.

3 x 64GB Sandisk Extreme pro = abouts  360 euros.

 

total

~2564 euros..

[Andrew Reid]

KineRAW MINI now has 1080/60p and 2K at 48p in firmware update.

[mtheory]

Would love to see a short greenscreen test or unkeyed clip from the Kineraw in the upcoming review!

[Dan Hudgins]

The highlights in the dngs look harsh with this camera.  They look clipped even when they aren't.  Even in the wide shot of the park which looks exposed to the left, the highlights on the bike, the table cloth and the guys back looked clipped.  It doesn't even change if you drop the exposure down to minus 4, the brightness and contrast to 0 and recovery all the way up: they are still there beaming away like solid blocks of bright.

 

I will be very interested to see if the Exposure Index Mode you speak of can be utilised to alleviate this.

 

You cannot judge the camera's RAW exposure from graded images on a monitor because the monitor may not comform to ITU601 limits or the images may not have been processed to retain the cameras output range within the monitors limits, in particular if you are viewing on an LCD or HD monitor.  The data values can be measured direct from a linear TIF made from the DNG to see if anything was clipped in the camera's exposure, then its just a matter of processing and grading the DNG's data to retain what was in the recordings to see it on the screen.

 

The camera makes 3D-LUT for monitor matching that relate to the look selected, but there is an output range selection of the look group selected.  The camera is supplied with KineLOG which gives Cineon (tm) standard film log output, in that case super white data above 90% white subject would be over code value 685/1023 so unless you grade that back down to be in the visable range you would not see the highlights captured by the camera, that is normal for all DPX type film log files that have not had softclip applied.  KineCOLOR defaults to full range output so cannot be loaded into Rec.709 workflows or shown on an ITU601 limited monitor without having range reduction applied, that can be done with the range limits override menu to select one of the HD/ITU601 limits options (with midtone at 46% or 40.7%).  Kine709 sets range limits for the 3D-LUT to default to ITU601 by default, but if you do not use the 3D-LUT made by the camera as part of the grading, or you use DNG processing software that has AGC, the software you are using may clip the recorded data and cause loss of lightlights.

 

These are serious issues, and using some software my degrade the results possable because of the way the software clips the data that the camera records, or does not prep the data ranges for the monitor or projector used.

 

You can email me if you want to have me check your DNG to see if they were overexposed, of you can use DNG_validate.exe and IRFANVIEW (tm) to read the highlights and see if they come out over 90%.  If you did not see zebras when shooting and you had them set to 90% raw levels, then the highlights were probably recorded where there were no zebras.  You can also set the waveform monitor to raw levels rather than post to see what the sensor ADC CV were at the time of shooting (the waveform monitor can work during playback in the camera).

[Dan Hudgins]

But the smaller sensor has it's advantages too, more lens options, better DR. But i agree that the main competitor to the Kineraw is probably the 5D3 because of the full frame.

 

Either way, the Kineraw seems like is stuck between a rock and a hard place in this market. If I was desperate for full frame, i really don't see the value of it compared to the 5D3. 

You should do careful testing of the 5D3 when anti-OLPF sharpen is applied for aliasing and moire when good sharp lenses are at prime stol and accurately focused.  Likewise with the BlackMagic 2.5K camera and its lack of an OLPF filter.  The OLPF filter in the KineRAW-S35 (tm) prototype I tested seems to give the needed results using real cinema 35mm lenses in the 2048x1080 mode for cinematography. 

 

Its not just a matter of price of the camera, if you are making a movie you want the MOVIE to have some value, so the tools you use impact the look and feel of the end result, and how well you can do the cinematography, if you have a KineRAW (tm) its similar to using a movie camera so does not maybe get in the way of your working like you would with a film movie camera, and the OLPF is designed for cinemitography with sharp cinema lenses, so you want to avoid being penny wise and pound foolish in selecting a camera based on cost, you are making a movie and will have much larger costs involved so you want to get the images that will help you make the time and money well spent overall...

[Dan Hudgins]

Would love to see a short greenscreen test or unkeyed clip from the Kineraw in the upcoming review!

How you process the DNG will impact the results, as does the exposure, lens, f/ stop and de-noise applied...

[Dan Hudgins]

But the smaller sensor has it's advantages too, more lens options, better DR. But i agree that the main competitor to the Kineraw is probably the 5D3 because of the full frame.

 

Either way, the Kineraw seems like is stuck between a rock and a hard place in this market. If I was desperate for full frame, i really don't see the value of it compared to the 5D3. 

I added support for the 5D3 and Magic Lantern RAW DNG conversion and have looked at test frames shot in the Canon.  I have also added support for the BlackMagic 2.5K and have seen their data.  Processing all three cameras in my free de-Bayer program I can see the impact of the OLPF or lack there of on all three cameras results, and I would prefer the KineRAW (tm)'s results for my own projects based on what I have seen myself so far.  If Canon would improve their OLPF for use in movie mode or Magic Lantern would support un-binned modes in the Canon sensor, then its OLPF might work better, but I would prefer not to have serious aliasing and moire problems you cannot remove chroma moire using software in any way that ends up beter than having a working OLPF to start with.  The lack of a good OLPF in BlackMagic's 2.5K camera is a serious issue for me, I would need to add one for my own use, but the extra glass thickness adds problems with wide stops so its beter to get a camera engineered to work right to start with, maybe.  You should not pass judgment based on things you see converted to H.264 on the internet, rather you sould rent each camera and do your own testing under the conditions you will be shooting under with the lenses you intend to use to see which camera's ergonomics and aliasing, moire and noise in the DNG RAW data levels works for you, as there are tradeoffs with Bayer type cameras and the results or DNG processing will vary widely depending on if the Adobe SDK is used or not or if you convert to Cineform (tm) or use another processing option.

[Dan Hudgins]

Better? Or just different.

 

Super 8, Super 16, S35 and even IMAX have existing alongside each other for decades. No filmmaker considers one 'better' than the other, just a different brush to paint with.

 

With cameras, of course there are ones that do the same thing better than another, but I don't automatically consider a camera like Digital Bolex 'DOA' just because it has a small sensor!

How much jell-o does BM Pocket have?  I asked if they would support 2048x858 in the BM Pocket and they ignored me, its not a 2.39:1 cinema camera even though that is less bandwidth than 1920x1080, so where is the Cinema Part?  At least Digital Bolex and KineRAW (tm) support 2048x1080...

[Dan Hudgins]

@HurtinMinorKey: KineRawmini is NOT fullframe

S35 depends on if you look at projection apture which is what you see on the screen or camera apture which you never saw on the screen.

[Dan Hudgins]

How is the low-light performance of the mini?  Is it the same as the S35?  Will you be able to upload any more tests in the near future?

 

Thanks

It should be, in tests with the KineRAW-S35 (tm) I found speeds up to 5120 (1280! x4 gain) to be usable after de-noise on shots were the camera is static, people at NAB liked the results show at the KineRAW (tm) booth.  Its possable maybe to use speeds up to 20480 with heavy de-noise or if you are going to DCP where noise can be higher than BluRay, but speeds above that would probably best be limited to SD or DVD.  The maximum speed is (2560! x64 = 163840) but that is best limited to documentery or SD news.  For normal shooting 1280 works well with fast lenses, you can pull up the shadows if needed in grading to get higher effective speeds.  Using 2560! x1 gives the most highlight range for high contrast subjects, about 5 stops above 90% white subject.

[Dan Hudgins]

Kineraw mini at say 2500 dollars .. would be a good price point.

When considering the 4K path .. I understand that you need the 290$ module for that. Then later the ~3300$ box

 

Some calculations .. I could not help my self :D

 

 

BMCC - 1780 euros

Chinese 12v 6800mah LI-ION battery - 25  euros each x 2 = 50

1TB Samsung 840 EVO - 500 euros

 

=  ~  2330 euros.  (get's you 12bit at good enough resolution RAW + Resolve 9 )

Canon mount or passive m3/4..    Only problems in my book are the multiplier and large size along the poor form factor.

 

 

BMPC  880 euros

Chinese Nikon battery 4 euros a piece.  x 6 =  24 euros

Speed booster when available  550 euros

Sigma 17-35mm  1.8   750 euros.

3 x 64GB Sandisk Extreme pro = abouts  360 euros.

 

total

~2564 euros..

All Kinefinity.com (sm) needs to do to stay in for the long run is to price their products above actual costs and sell enough to make it worth the bother, they are a division of a successful telescope camera company, so they parent company should stay around as far as I know. They have good prospects for the Asian market, as well as for people that like their image quality.  As far as I can tell they are positive on developing new and aditional products and do not relay on large volume orders.  BM seems to need volume orders to meet its price tragets from my impression.  From what I have seen on the internet BM is spending money on advertising, wereas Kinefinity.com (sm) is not, so in the end, who's bottom line will look better?  Better to not through volumes money out if you can get a smaller but steady flow in?

[Dan Hudgins]

With regard to the media costs, it may be possable in the Mini to have the SSD dump to the USB port, its a Linux computer in the camera as far as I know, so its just a change in file path maybe, I have suggested that to them.  In the larger S35 you can insert a notebook SSD into the second slot to dump or back-up the SSD for re-use.  That way you don't need to pack around a notebook computer to empty the SSD and shoot more.  You can use any compatable SSD you like, its just a matter of it being fast enough no to drop frames as I understand things, the KineMAG (tm) are qualified to work.  With the 60GB ones I have been testing, I did the first testing with their prototypes, in both the KineRAW-S35 (tm) and higher bandwidth KineRAW-S8p (tm) I have not noticed any dropped frames shooting for over a year now.

 

Its even possable to shoot to an HDD if the shots are short, when the buffer over-runs the camera breaks out of the shot.  I use a 500GB HDD that I keep with me so that when the SSDs fill up, I can dump in them the camera  and shoot some more.

[Dan Hudgins]

One additional point that I can see as a real issue is focus aids, the KineRAW (tm) have 800% zoom that makes for sure focus, in addition to 100% and 200% as well as full view, without those features being fast and easy to use with region of interest, and being able to increase the monitoring contrast without impacting the recording, it can be hard to be sure of your focus, so check out the focus aids and how easy they are to use on other cameras.

 

For my own projects I want these things:

 

1) A good OLPF so there is control of aliasing and chroma moire

 

2) Focus aids that are useful for being sure of focus

 

3) Being able to use mount adapters for real cinema glass in the various mounts we have from our 35mm cameras

 

4) True RAW DNG that are uncompressed that fit into my free de-Bayer program that gives full manual control over the color balance

 

In addition to that the KineRAW (tm) cameras have an open monitoring LUT system you can develop your own looks for to use in the camera and those files are user loadable and will show up in the cameras menus, you do not need to depent on anyelses idea of how your footage should look.  The LK5 controls the EI ISO curves and can work with native sensor balance, useing optical filters like a CC50M, or full analog gain for white balance.  The LT5 file currently works as EQUAL.LT5 but they could revise it to work for unbalanced LK5 as well, you can edit the EQUAL.LT5 in any text editor to get any ratio of red, green and blue analog gain, normally though you calibrate the K and light type to equal ADC CV for a netural subject, you give those values a label, and then load it onto a USB thumb drive, put that into the camera it says "nnn user files loaded!" then you take the USB drive out and re-boot the camera, at that point your custom white balance and or additional LK5 look groups show up in the camera's menus just like the factory presets.  Why complain about not liking the way the highlight rolloff is setup, just add your own curves at will...

[mtheory]

5D3 vs Kineraw Mini from Cinema5D

 

http://vimeo.com/74372043

 

Looks like 5D is better for low light scenes, Kineraw better for bright daylight scenes.

[Andrew Reid]

The DR is distributed differently so it isn't a very good test.

 

Rather than just use the same exposure, they should have exposed for the sky so that the highlights were the same on both cameras, and then looked at the shadows.

 

Admirable performance in terms of colour from the 5D Mark III though.

[Dan Hudgins]

There are many factors that impact the end results when shooting a film, and important things to concider are the ergonomics of use and getting shots in focus fast enough during real shooting.  How the cameras focus aids work and their ease of use is an important point and one that overrides any price difference within what is afordable.

 

The aliasing issues with the 5D3 are serious enough from what I have seen processing test frames, it it varies with f/ stop used, that I would not concider the 5D3 usable, as with the BM 2.5K camera, and maybe Digital Bolex, without a suplementry OLPF filter.

 

The KineRAW (tm) cameras' sub-PL mount lets you use lower cost S35 and A35 real cinema menses designed for non-reflex 35mm film cameras, something that is not really an option on the 5D3 because of the mount placement and reflex mirror.

 

Price is less important than if the camera is usable for the kind of filmmaking you want to do and if the camera gets in the way of you expressing yourself.   Chroma moire is better delt with by having a good OLPF filter in the camera to start with, that way you don't get a supprize later when the anti-OLPF sharpen is applied because you were in focus at prime f/ stop and using good lenses that focus sharp to start with.  When all is done, the cost of the camera is a tiny part of a films production budget.  And its false econimy to cut corners that will impact the end results that make your reputation by having aliasing and moire in your footage.

 

You cannot compare cameras shooting CinemaDNG and say this or that was the result of the camera, other than making linear TIF with a program like DNG_validate.exe since the anti-OLPF sharpen and de-noise neeed for each camera's footage needs to be optimized to the camera settings at the time of shooting, the f/ stop and lens focus quality and many other factors.