Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Cinema Camera in your Pocket: Best Android, IOS, Windows and Linux Camcorder software

Wayne Morellini

Well-known member
Joined
Jan 3, 2008
Messages
6,157
Reaction score
0
Points
0
Here is the question , does anybody know of any high quality video camera recording (and stills) apps for portable devices (ubuntu based Linux for mobiles is coming)? Interesting qualities are features, controls and recorded footage (higher bitrate and raw). Treat this as a thread to track these things, and cinema/video camera assistant apps in general are welcome also.


Even though this is something that most professionals could use for personal use, someone doesn't need a big sensor or lens to film something useful outside of cinema. Modern software can correct many things in post. Phones and tablets are getting to a surplus of performance, enough on the high end to shoot good fullhd for video/TV, and on some models with a bit of customization maybe even in ultra HD.


It was said elsewhere:
It is a matter of what professional camera app can be run on it, and I call on Red to sponsor one, ........ on app store $10-$100. The Red Real Camera app for the professionals off time. (not to mention learning young camera people). No hardware manufacture lines needed.[\quote]

The interesting thing is that an Android app could run on an Android camera, even an Epic if it had Android. On phones, a standard generic version might handle HD or fullhd, with customisations for ultra HD for only a few top models that can handle it (to reduce development costs).
 
A few hours off my life, I've found out IOS has many. But Android has a few. I've got some downloads to look through to eee if any has what it takes. It is amazing that even though up you paid for an app people still want to spy on your phone.

Lgcamera Pro is apparently a leader in upgrading your data rate:
https://play.google.com/store/apps/details?id=rubberbigpepper.lgCameraPro&hl=en

CineDroid (not with small d, that is q cinema program) is the king to be. The guy has withdrawn it (the playstore app pages are different) he is concentrating on snapdragon (which is what is in the ultra HD phones, so hopefully). and moving it to github under global public license:
https://play.google.com/store/apps/details?id=in.shrthnd.cinedroid&hl=en
http://mobility.forumsee.com/a/m/s/p12-9287-0304180--cinedroid-pro-video-camera.html
http://forum.xda-developers.com/showthread.php?t=2230563

People have been hacking the camera to get 30mb/s, but filmic on iPhone gets 48mb/s (last year so newer Android maybe faster).

http://forum.xda-developers.com/showthread.php?t=1537606

ICS after market version camera app claims 2x the data rate:
https://play.google.com/store/apps/details?id=com.moblynx.cameraicsplus&hl=en

The interesting thing is that the Android operating system was originally designed for cameras, according to an interview I read.
 
Well, unbelievably, I did get lgcamera pro to record at around 90+mb/s, set to over 720p in mpeg4, which probably requires a lot less processing power. The results were less than desirable though, with often unrecorded 0byte files, jerking, seizures (though not every shot), but a bigger file did play smoothly. This pad is not the greatest for this though, but does have an tegra2. My replacement amoled 7.7 would probably fair better +and give a better image). The other surprising thing is it has p25 and p60, but no p50 on my pad, but that might be a hardware mode restriction .

ICS produced a smooth image in it's 2x bit rate mode, but only around 16mb/s, which is just OK in the video field if it is proper h264, but not much past that.

I can imagine what might be happen with the latest hardware, maybe with 4 times the processing power or more.

A note:

I did try filming a soccer match with my 7.7 inch amoled. I was under no illusions that it would turn out good, but the screen did make it look nice. But I did determine something, that the 7.7 inch from factor was nice size on the eyes, but maybe a bit big. I gave tested 6 inch, and that seems much better, anything under 5 inch will provably get a bit small for comfortable framing. So 8-5 inch tabs/phones might be best, but one big factor I found with my 7.7 was the weight (it is a metal case). The new tab 3 10 inch with twin Intel atom CPU's is light, maybe similar to my 7.7, and a lot less than my old 10 inch, thinner and nicer to hold. It will be interesting to see what happens as far as a good maxinum size to hold filming, as they get thinner and lighter, until the wind starts to steer them. It will also be interesting to see what sort of rig people might make for them. Now, for a large flat lens system to go on a tab (Sony is already starting down this path).
 
Wayne, here is your Ambarella... 4K@30fps

E. :-)
 
He, he. Thanks Emanuel. 30 fps is old school. The problem is the power envelope for a mini action cam, you need something like Intel's latest processes to make it work. Ambarella do custom jobs, so there should be a design for 60p out there, just not as a regular order able line. But can this be programmed to do this anyway. The thing is that on very small cameras ambarella is probably going to offer better performance characteristics (including power and heat) and by the time we get to 8k, a custom circuit like ambarella might be the only viable option until circuit technology deals with a few power/heat issues (looking into a few technologies to do with this). 8k will be a sweet spot for video stills, so desirable for action and phone cams that gpu's might not be able to compete with.

If Intel implemented ambarella circuits in their leading edge low powered chip processes, on their new low end processor series it would be a real coup for them, and enable 8k sooner. They could take the camera market like a storm. With the sort of mobile lens/focus technologies coming, mobile 32mp stills and videos are a viable option. There is a new resistive dram technologies that can replace flash and boost relative cost densities, so as soon as they can design and make the chips they could do this. But a codec like redray that approaches my old ideas, would be very handy.
 
OK, reading that article Emanuel, something very interesting pops up, but are you sure they have an actual leak, or is it just recent speculation putting the A9 together with the next go pro. There has been speculation of a bigger gopro product. By doubling up the A9 chip, you can manage 3d 4k at 30fps, or 4k at 60fps, likely at 160-200+mb/s. If I pass by Didlier on holidays again (didn't know who he was the first time), I'll have to talk with him about future direction. Ambarella is a truly exciting little company.

I don't know what go pros numbers are, but you would need millions of sales to drive a custom chip run big enough to drive down the price. But such a run using a9 technology, or a10, on the lowest available low power chip circuit could make 4k 60fps/3d part possible that could fut in such a device. This is a lit of its planning and timing to make it possible. The a9 technology is designed and so less work on a expanded design is needed (which probably is already on Ambarella's custom design board), and we are due for release of the next chip, maybe speculatively, something like a a9+ or a10. So, that is possible. Also remembering that a single core 2ghz Pentium could do cineform raw fullhd capture 9 years ago, a more modern mobile system might scrape in 4k 60p taking advantage of all the onboard processing circuits, only just speculation here.
 
Traced it, does not seem to be an actual leak, just hype. The original article is going on just discovering the existence of the a9 several months after it was announced and talked about in relation to the gopro. I have also been talking about possible timing of their 4k parts for the last tear or two or three, and about the indication from the announcement of the 4k security/car video chip since last year that could be used. Some of these things are obvious and predictable by Moore's Law with a few shifts and taking into account a few issues that affect it. Which brings to mind, we must be getting closer to 8k Epics, next year or the year after, but maybe they will go to 10k+.
 
If the combination of our Vizzywig and Video Filters Apps isn't already the best mobile app solution, we will add whatever features and settings the Red community thinks is necessary and missing in the near future.

Let me know what you want/need.

Btw, I still own (2) Red One MX's and (2) Red Epic's and am near the front of the line for the Dragon upgrade. Smartphones and GoPros are, and always will be, accessories for the Epic and other Pro Gear not replacements or challengers. The real excitement is in smartphone software and how it will eventually compliment a 6K workflow.

GeekBeat.tv Interview on Vizzywig 4.0 (last week)
 
Michael,
Just saw that demo of Vizzywig 4.0. Looks amazing, and very powerful.

Best of luck, and continue the awesome work.
 
Hi Michael. This may take a number of editions over a few years, but would give great flexibility for users to expand the software many different ways. Basically, as tablets don't have lots of controls professionals would need lots of controls on screen, unfortunately, and high quality recording. I'm occupied (now I've side tracked too much) and have to talk to another camera development team, so I'll have to get back. But quickly: selectable bit depth and format (4:2:0, 4:2:2, 4:4:4 and raw) 48 and 50fps, resolutions above tge official resolution at highest frame rates, hdr hardware support, settable and adjustable gain, shutter, iris, focus, color and pixel matrix, log controls, zebra, focus assistant/lock, I see you have some smooth zoom assistant there, anti shake, anti blur, and as much coprocessing acceleration as possible. A lot of the advanced features on the higher end Sanyo Cameras were good, such as control points for focus and zoom during preplanned movement. Interframe denoise. A user wysiwyg customizable and savable control/menu system layout that users can share with other users, and customizable user added controls (this is where I would prefer to spend time). Another thing is plug in modules interface for the user base to add code and scripts, with open cl support). A new scripting language is not needed, an easy to use standard scripting language would do. An external hardware control interface (wifi and usb).

As for challengers to Epics and the like, not really, but for school students, hobbyist, amature film makers, newspaper reporters and photographers, websites, musicians, marketing agencies, low level documentary and film makers, yes. But also, if it were on Android, it could be added to an Android camera to as a sort of friendly Back magic Pocket, maybe the consumer micro 4/3rds Kodak coming. So my suggestions could be useful. I have been around a long time, and the above suggestions were from my experiences working out design directions in the day cinema camera projects that influenced the formation of Red, they remain right for today (as with recording raw and changes separately). I also suggested to Sanyo multiple levels of user interface modes, where you select between easy to shoot mode, to intermediate mode to advanced mode that gave you all the options, so all the hard stuff was hidden away on default. This can help hide advanced functionality away from beginner users. As for cost, you could charge a bit for an advanced version. Personally, something like above is worth $100 to me.

Will you be doing an Android, windows or linux version (there is an mobile version of Ubuntu coming that sits on Android)?

Thanks again Michael.


Wayne.
 
LgCamera and CineDroid Tests

LgCamera and CineDroid Tests

Well, still consumer like with some nicer selections. Found it hard to get the selected frame, resolution or data rates. It appears that the settings don't get written/read properly, but think it might be also a matter that only modes supported in firmware or hardware modes are supported. Anyway, putting it in 30p mode finally gave me nearly 60p mode on my 7.7 inch tablet I think, at over 50mb/s h264. The codec is extremely variable. Sports was as close as I got to max frame rates. The other thing is that h263/mpeg4 part 2 seemed to drop into a bad software only compression mode on one tablet, were data rates and frame rates were absurdly low, but did well on the other tablet where the h264 maxed out at consumer rates. Unfortunately both tablets have camera setups not so good for this software, but could be much better with fullhd given hardware acceleration software. So, it is worth testing out which settings work the best, and getting a device with good camera performance that is verified to work well with the software.

CineDroid has some fiddly controls on the bottom edge of the screen next to the Samsung popup toolbar and android controls. Popup help and help menu item will probably be very useful when it gets more controls. But it has the same issues as LgCamera with various things like frame rates etc. But this is beta software software, so a lot of things to come. There is a mile of things that could be done, and at low cost. The good thing about Android is that a startup could do an online app and if very good even be picked up by hardware manufactures to include in products/cameras. The only issue is the large money for licensing many patented technologies.
 
Last edited:
Thanks so much Eric!

We just submitted version 4.1 to Apple which adds landscape left mode for people that like to have the volume buttons on top and a really cool new feature that lets you turn any clip from your session or your camera roll into a multi-clip. Then you can edit that multi-clip and record additional clips of same length while watching a video or audio+video preview. Once recorded, you can tap "play all" and live switch between the clips in real time to create the final edit. As far as I know, this is the fastest and easiest tool ever created for creating amateur music videos.

We also added roundtrip-support for our Video Filters 2.0 app (coming this week) so that you can choose any clip in a Vizzywig session and open it in Video Filters, apply true-optical slow motion, fast motion or any one of a 100 filters plus basic grading tools and then export that clip back to the Vizzywig timeline. Video Filters also supports iFrame recording, 24p, stereo audio recording and hardware stabilization on supported devices. It's a powerful combination.

We also have a new App Icon.

Michael Zaletel
Founder, i4software
 
Hi Michael. This may take a number of editions over a few years, but would give great flexibility for users to expand the software many different ways. Basically, as tablets don't have lots of controls professionals would need lots of controls on screen, unfortunately, and high quality recording. I'm occupied (now I've side tracked too much) and have to talk to another camera development team, so I'll have to get back. But quickly: selectable bit depth and format (4:2:0, 4:2:2, 4:4:4 and raw) 48 and 50fps, resolutions above tge official resolution at highest frame rates, hdr hardware support, settable and adjustable gain, shutter, iris, focus, color and pixel matrix, log controls, zebra, focus assistant/lock, I see you have some smooth zoom assistant there, anti shake, anti blur, and as much coprocessing acceleration as possible. A lot of the advanced features on the higher end Sanyo Cameras were good, such as control points for focus and zoom during preplanned movement. Interframe denoise. A user wysiwyg customizable and savable control/menu system layout that users can share with other users, and customizable user added controls (this is where I would prefer to spend time). Another thing is plug in modules interface for the user base to add code and scripts, with open cl support). A new scripting language is not needed, an easy to use standard scripting language would do. An external hardware control interface (wifi and usb).

As for challengers to Epics and the like, not really, but for school students, hobbyist, amature film makers, newspaper reporters and photographers, websites, musicians, marketing agencies, low level documentary and film makers, yes. But also, if it were on Android, it could be added to an Android camera to as a sort of friendly Back magic Pocket, maybe the consumer micro 4/3rds Kodak coming. So my suggestions could be useful. I have been around a long time, and the above suggestions were from my experiences working out design directions in the day cinema camera projects that influenced the formation of Red, they remain right for today (as with recording raw and changes separately). I also suggested to Sanyo multiple levels of user interface modes, where you select between easy to shoot mode, to intermediate mode to advanced mode that gave you all the options, so all the hard stuff was hidden away on default. This can help hide advanced functionality away from beginner users. As for cost, you could charge a bit for an advanced version. Personally, something like above is worth $100 to me.

Will you be doing an Android, windows or linux version (there is an mobile version of Ubuntu coming that sits on Android)?

Thanks again Michael.


Wayne.


Hi Wayne,

We are already working on some of these features within the constraints of Apple hardware. Yes, we are working on an Android version and will be working on Windows Phone and perhaps Tizen versions soon We already have a remote camera client for the Android for Vizzywig so you can use any Android device as a remote camera for an iPad or iPhone in a multi-camera session.

We will likely add some of these really advanced features as In-App Purchases to help cover the R&D and development costs. For example our module that will allow users to control multiple Double Robotics Doubles from the Stage master device while previewing and recording will definitely be an In-App Purchase because of the low quantity and the fact that each Double costs $2,500.
 
Hi Wayne,

We are already working on some of these features within the constraints of Apple hardware. Yes, we are working on an Android version and will be working on Windows Phone and perhaps Tizen versions soon We already have a remote camera client for the Android for Vizzywig so you can use any Android device as a remote camera for an iPad or iPhone in a multi-camera session.

We will likely add some of these really advanced features as In-App Purchases to help cover the R&D and development costs. For example our module that will allow users to control multiple Double Robotics Doubles from the Stage master device while previewing and recording will definitely be an In-App Purchase because of the low quantity and the fact that each Double costs $2,500.

Thanks Michael, sorry for the amount, it is more like a condensed list of cinema camera design thinking over the last years. I can see the market you are going for, that is why I mentioned Sanyo that also added some good scripted changes that could be useful for professionals. I'm more interested in an alternative to the Black Magic Pocket Cinema Camera that lays between a pocket cinema and a pro video camera, with consumer ease features. A hybrid, so to speak, that a cinetographer, pro video or consumer could pick up and use to great potential. I'm also interested in maybe doing external control designs myself with some radically new approaches, and using external cameras (currently there is only one usb3 4k camera available that I know of) that can be put in a rig with the android device as the display, engine and recorder, or it could be attached to the tablet case. So lots of potential there. Wifi direct and the wifi consumer av standards could be a simple basis of many different things.

Michael, it is good you are working on Windows and Android versions. Does that mean that those versions won't be constrained like with Apple hardware?


Thanks

Wayne.
 
Getting 4k p50/60 recording on mobile devices.

Getting 4k p50/60 recording on mobile devices.

I have had an idea that may help record higher resolution or framerate video at higher bitrate or quality if the hardware allows it.

An alternative codec could do this, and the webm vp8 and vp9 codecs are open royalty free web standards and are meant to use less processing to get similar compression ratio's to h264 and h265. V9 supports a higher profile byte looks of it. So, it might be possible to find a carefully crafted multipath codec plugin that takes full advantage of hardware acceleration options to boost recording on a 47mb/s 4k p30 Note 3 to 100mb/s 4k p50 if hardware can support that frame rate. If such a plugin's license requires the user to load it themselves for it to be royality free, so it can be access by other programs, so be it. The only issue is compatibility with editing software to reduce image degradation from transcoding if there is no native editing solution for vp8 and 9. But this maybe the only way to get the highest resolution/framerate on many mobile devices.

http://en.wikipedia.org/wiki/WebM
http://en.wikipedia.org/wiki/VP9

I forgot to mention, the vp series was released into the public domain from Google after they assured the original company.

Another that aims to be better than h265 in performance.
http://en.wikipedia.org/wiki/Daala_(video_codec)

The other option, is lightly inter compressed lossless raw 1.5:1-2:1 put on the latest generation of high speed card/usb3 (if available).
 
A query, is there a codec that could be used for video for jpeg-xr which is similar in performance to wavelets but less processing intensive? This might fit hardware coprocessing easier than wavelets. It was based on the Microsoft picture format and has some public domain aspect to it I think.

Thanks.
 
TI'm also interested in maybe doing external control designs myself with some radically new approaches, and using external cameras (currently there is only one usb3 4k camera available that I know of) that can be put in a rig with the android device as the display, engine and recorder, or it could be attached to the tablet case. So lots of potential there. Wifi direct and the wifi consumer av standards could be a simple basis of many different things.

What camera is that? Very curious?

-Michael
 
http://ww2.ptgrey.com/USB3/Flea3
http://www.ciol.com/ciol/news/19433...tibility-point-greys-flea3-cameras-ni-labview

As you can see it is fairly mediocre and not suitable for cinema unless you can overclock it. There should be many more. A more relevant project us:

https://www.apertus.org/axiom

There is some interesting usb3.0 possibilities: Some tablets have usb host.
http://en.wikipedia.org/wiki/Mobile_High-Definition_Link#MHL_3.0_Features

I personally prefer thunderbolt/2 and there are mobile versions coming.


However, a likely outlier possibility is the use of 7gb/s wigig wireless in items like these (though the silicon imaging standards does over 26gb/s on the same frequencies):

http://sonyalphalab.com/2013/09/new...s-use-your-cellphone-or-tablet-as-the-screen/

There should be others, plus this makes a good opportunity for small sensor manufactures to supply superior parts to Chinese competitors, and they will need a camera app to run them. I have considered doing this many years ago, but only Bluetooth was available back then.

http://www.mirrorlessrumors.com/first-lenscamera-rumors-from-samsung/

At short distance the wireless power consumption should also be acceptable.

There is also a launch of a module phone platform, something else very close to my heart, as it is also something very similar to what I was planning. This means that sensor manufactures and others can offer sensor+processing modules.

http://www.phoneblocs.com/phoneblocks/

The important thing is if it sets up a standard for on board modular components, the blocks can be used on board precertified for electromagnetic interference. Small manufacturers can now more easily "lego" designs together of various hardware, even as hybrids of custom circuits using modules. I've looked at many ways to do this across many scales looking for viable methods for products.

So Michael, there are a number of possibilities, that might come through and offer potential.
 
Back
Top