Welcome to our community

Be a part of something great, join today!

Log in
  • 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 :)

Vizzywig 4K Footage (Anamorphic Lens - Swift M3 Stabilizer)

michael zaletel said:
vizzywig-4k-8.jpg
Click to expand...

Where are you Michael, that sort of tree is native to around here?
 
OK, I wrote this earlier, but could not get it to go through:

------

Where do I start Michael. I thought you had a newer solution for the 6s at higher frame rate, I know what you are talking about now.

I had thought of using jpeg in a cinema camera. I've circulated around a few cinema camera developments, and dropped my (non JPEG) cinema camera proposal when that .. pocket (Scarlet) was announced years ago. Now the pocket BMD is filling out the market.

You could use the native arm code execution feature of android, turn off conflicting services, but you really need root access Replikant OS, or Cyga..mod (however you say that) version of android. But then, as you said, everything has to hum to get the write speed to flow through (have a look at the other thread where I describe a setup to streamline).

But the reality, is it is not about write speed, it is about compression processing support to halve the write speed down.

600mb/s jpeg1 is roughly equivalent (without much motion) to 300mb/s mjpeg2, h264 intra, jpeg2000 or JPEG new technology. One of the mjpeg2, avc intra or newest jpeg intra codecs is likely going to be the least processing intensive and have more processing support. Maybe it will be the newest JPEG. For all others reading, the 600mb/s should be similar to 150mb/s h264 inter, or 75mb/s h265. But with Bayer intra even less required again over the intra codecs. Now stepping back, we see some fairly fairly common standards and codec support, so compression hardware, processor and GPU sections could be employed on different parts of the image, the same parts, and or frames in parallel. Frankly. cineform's raw wavelet codec is the best, getting cinema level performance between 6:1-4:1 (lossless) (but I think 10:1 might also be in the pro range) compared to Bayer data rates (not 4:4:4, but 10:1 on non bayer pro video) designed to work on weaker CPU's without GPU's. Which they have been trying to make into a standard through a standards organisation). But you would have to contact them (David Newman) over at go pro about a very reasonable license. But you can just go with any pro video standard and codec, as BMD did, including, soresson (I forget his exact name) who is working on a new high performance video codec. Also the VP8/9/? codec, open licensed by Google as an official web video standard, were designed to be a processing light alternative to h264/265. Maybe vp could be made to act as a intra codec, but software and processing support might be low for now.

Now taking a step back again, and a leaf out of David's cineform raw codec development. You could compress each raw Bayer color channel as an individual intra frame, then look at difference between them. Going one step further, as Juan, I think, who works at Red, looked at back then, was record the simple difference between frames, without going into all the complex motion compensated differences. It is mainly for reduced storage space while avoiding a lot of other inter intensive compression, but not maximum data rates. I advised BMD of this, and simple data rate throttling.

So, it might be possible to get higher frame rate 300-600mb/s Bayer based 4k on Apple and Android products using the above techniques. Now, mentioning Android products. The old batch were blunt instruments, so let's talk about the new 64 bit processor platforms, particular Nvidia K1 processors with desktop level GPU inside and CUDA support. I think the days when an android could not do what you did on the 5s are rapidly drawing to a close.

Now, occasionally I mention something I'm working out, just to see if anybody is awake and intelligent enough to jump on board business wise. I am wanting to get to a side compression technique project of mine, which aims for 10:1 lossless, and a couple of other techniques aiming for under 36mb/s 8k combined lossless 12 bits, with low processing overheads. I am thinking of doing it for a new camera product, but it could be used for cinema. This is not my old work from the 80's to 00's which is too complex to do easily.

Wayne
 
I'm not trying to tell you that you shouldn't make this app, or that it isn't functional, or that it can't deliver beautiful images, or any of that. What I am saying is that I am confused by your business strategy. Currently, your target market is a demographic that:

- Owns an iPhone 5s (likely only a 64GB version).
- Values 4k image quality over any other metric (filesize, framerate, etc.).
- Has to have another $1k to spend on top of the cost of the phone.
- Can't be interested in any other 4k capable smartphone or small form factor 4k camera, both of which cost significantly less than your app+phone combo.

I won't argue that this demographic doesn't exist--something tells me that you might have already sold a few licenses of this app--but I don't think it's unreasonable to assume that your current target demo is dwarfed in size by the demographic that would spend $10-$20 on the app. I don't think it's insane to argue that the $10-$20 demographic is easily fifty times bigger than your current demographic.

I can't test your app because I don't have a 5s (just bought a 128GB 6) but it does look like it can deliver some stunning images. I know that it takes real R&D to make something like that happen and you should be proud of your achievement. That said, I feel like you're shooting yourself in the foot if you don't meet consumers needs with the price, especially when you have a limited window of opportunity here. All it's going to take for the bottom to fall out of this already small market is for Apple to release a 4k capable iPhone, which could very well happen during the next revision, and the incentive to use your app drops to zero.

Something to think about.
 
I'm not beating thin air.

Long ago in a galaxy not too far away, I advocated that Red or other could do a professional cinema app with redcode/other for $100, even just the codec. The price was accessible for any serious enthusiast, but also the lower data rates also meant cheaper cards. Even $20 is a better price, depends on the market numbers you are going for, as you could make clearer profits at the cheaper price points from more people buying. Let's look at your product. An industry best, useful for pro video, useful for enthusiast, actually useful for any serious person that wants a YouTube or Vimeo account that can pretend to actually film, especially in a simple mode with a word of mouth reputation. So, millions of potentials, if they can be convinced to buy and it is on enough devices. Taking a step back, what does the 4k version of the product offer over the normal version? How much extra is 4k burst photo record worth compared to Bayer codec support?

Does your software allow a codec installed in an IOS or Android device to be used, or a recording back end, in case I might want to implement my codec?


Thanks


Wayne.
 
The lumix cm1 is a pretty slim and sexy looking Android device: "the CM1 is something we haven't seen yet: a smartphone with a real compact camera lens that shoots 4K video."

Still
I'm sure there are enough people out there that will want Michael's, even if they don't exactly know why… etc.. And do the math, a modest 500 sales isn't too shabby.
Good luck with this Michael.
 
Kemalettin Sert said:
why sell few from 999$ instead of selling thousands from 20$ ?
Click to expand...

Agreed. Video quality is absolutely important to me. I have owned the Red One M, Red One MX, Epic MX, and now Epic Dragon. At great expense, I just fought to complete post on my first feature in 5K. You've created a wonderful app and I would love to have high quality 4K recording on my iPhone, but unfortunately $1000 is just way to steep for me.

If you sold this for $100 or less, I'm sure you would sell more than 10 times the amount you'll get at $1K, especially since in six months we'll be able to buy 4K GoPros for $400. Just a thought.
 
Wayne Morellini said:
Now, occasionally I mention something I'm working out, just to see if anybody is awake and intelligent enough to jump on board business wise. I am wanting to get to a side compression technique project of mine, which aims for 10:1 lossless, and a couple of other techniques aiming for under 36mb/s 8k combined lossless 12 bits, with low processing overheads. I am thinking of doing it for a new camera product, but it could be used for cinema. This is not my old work from the 80's to 00's which is too complex to do easily.

Wayne
Click to expand...

The way this type of development usually unfolds is to implement the proposed core methods in MATLAB, which as a veteran imaging scientist, you must have used before.
Then you shop it around seeking interested parties who might take it further in a hardware or regular software version.
Having some ideas that may make sense on paper or in your mind rarely works the way you hoped for as images and noise and motion cause non expected monkey wrenches to upset your hoped for results. In other words you must prototype it before you can even begin to sell it to somebody , even as a proposed idea for a compression strategy.
 
You price point this puppy somewhere around $50-$99 and it will sell like hot cakes and make you LOTS of money.
 
Bill Anderson said:
The lumix cm1 is a pretty slim and sexy looking Android device: "the CM1 is something we haven't seen yet: a smartphone with a real compact camera lens that shoots 4K video."

Still
I'm sure there are enough people out there that will want Michael's, even if they don't exactly know why… etc.. And do the math, a modest 500 sales isn't too shabby.
Good luck with this Michael.
Click to expand...

Thanks Bill. I appreciate that. I am hoping people will realize that they cannot edit in 4K, add titles, lower-thirds, background music and extract full-resolution stills using any of these other devices. That's why Vizzywig 4K is priced at $1,000. It's the workflow and the time savings.
 
Kemalettin Sert said:
why sell few from 999$ instead of selling thousands from 20$ ?
Click to expand...

Hello Kemalettin,

Determining the right price for a product is never easy. Price it too low and people will assume it isn't valuable. Price it too high and people may not be able to afford it. There is no guarantee that we would have sold 50 times more copies at $20 than we will sell at $1,000. Perhaps no one would even know about it if we had priced it at $20. It is currently the most talked about and written about app on the Internet.

But there is much more to this decision. We want to create a sustainable business model for the long term. It's almost impossible to build a successful, profitable, growing business around a $20 piece of software. We want to be here for the long term. I truly believe this app is worth $1,000 to tens of thousands of people worldwide who need to get real work done everyday and need to save time. Most of our users will probably buy a used iPhone 5S for $300 and dedicate that separate device for Vizzywig 4K.

"In fact, the first person here on RedUser that wants to take me up on the offer and buy Vizzywig 4K in the next 24 hours, I will send you a free iPhone 5S 64GB along with a free Moondog Labs Anamorphic Lens plus give you my cell phone number for VIP tech support. You can't make offers like that when you are charging $20. We are about solving people's workflow problems, not driving app downloads. I can't imagine why anyone on here serious about 4K content wouldn't carry around a 4K capture, edit and distribution tool at all times if it fit in their pocket. It doesn't replace Pro 4K cameras, it substitutes for one when you are in a pinch."

It's not so much about the camera, there are other affordable small 4K cameras. It's about the mobile 4K editing and distribution workflow and the still images workflow that allows for dual-purpose video and photo/print work. A reporter can film a 4K interview, edit it and post it in 4K to YouTube in 15 minutes and post high-res stills from the sequence for print in seconds. No other device or app in the world allows that right now... at any price.

Time is money...more than anything else, people value their time.

Sincerely,

Michael Zaletel
 
Les Dittert said:
The way this type of development usually unfolds is to implement the proposed core methods in MATLAB, which as a veteran imaging scientist, you must have used before.
Then you shop it around seeking interested parties who might take it further in a hardware or regular software version.
Having some ideas that may make sense on paper or in your mind rarely works the way you hoped for as images and noise and motion cause non expected monkey wrenches to upset your hoped for results. In other words you must prototype it before you can even begin to sell it to somebody , even as a proposed idea for a compression strategy.
Click to expand...

It is a matter of where you start. Matlab is for pussies. But of course you understand why it is not necessary, and why compression does not require it. Of course you realise I conditionally qualified all my statements, what "aimed" for. Of course I'm looking at doing it and proving it and making it in software and on a leading edge hardware platform I'm waiting for, but the reality is only patent costs. It can be made into a codec plug-in and sell itself. If I can get anywhere above 4:1 lossless it will be great, just because of processing savings. But there is one thing that great compression requires, elimination of noise (as not an authentic part of the image). that is the monkey wrench (motion in an intra codec does not matter so much).

Being able to accurately prototype in your mind is a much better skill than needlessly knocking what one doesn't know about. Absences tell a lot about knowledge.

Now Les, your profile picture makes ME feel like crying, its even more distressing then the puppy dog, but maybe not more than death skulls.

I personally think we could do great work together Les, rather than bickering, with my youth and your experience. I personally hate seeing good things being shitted on for no reason.

Thanks.
 
I just shot this in 5 minutes, edited it, added titles and credits and had it on YouTube 15 minutes later. That has never been possible before.

 
You must log in or register to reply here.
Back
Top