Guessing 1st April for 2019 - Share your thoughts

Started by theBilalFakhouri, March 25, 2019, 09:13:04 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

IDA_ML

I also would like to share a few thoughts on the future of Magic Lantern.  This time, I will do my very best to be as positive as possible.  ;)

As I stated several times on this forum already, in my opinion, ML is the most revolutionary achievement in photo/video technology since the invention of the DSLR.  The first videos published by Trammel Hudson inspired me to start reading on the ML forum, and although I did not understand much of what was written in the threads, I admired the efforts of A1ex and the other developers and followed with great excitement their first implementations of ML functionality into Canon DSLRs.  When the first RAW videos shot with ML on these cameras appeared on the net, I was so amazed by the video quality that I decided to save every penny I earned with my second job as a photographer to buy a ML-capable camera.  Shortly after that, RAW video was successfully ported to the 7D, I bought this camera and couldn't be happier that I did that.  ML on the 7D inspired me to shoot my first amateur videos.   Quality was so good that it exceeded all my expectations.  I remember, friends of mine asking me if these videos were shot on the RED or Arri Alexa.  And when I told them, they were shot on the 7D, they wouldn't believe that.  After I made my first demonstrations with filming RAW video, extracting the DNGs with ML RAW Viewer and processing them in Lightroom, only then they believed that.  At that time, many people were inspired to start experimenting with ML. 

Now I can say that ML changed my life.  Thanks to it, I learned so many new things, tested so many new exciting techniques and also learned to be more careful and precise when shooting stills and videos.  As a result of that, also my photographic results improved and my ML videos brought a lot of joy and fun to my family and friends.  It's hard for me to express my gratitude to all developers for making this miracle happen!

Enough with the history, let's talk about the future.  In my opinion, if ML continues to move forward in the same way as it is moving now, it has a great future for at least the next 10 years to come!  Here is why:

1) According to A1ex's statements above, if I understand properly, soon we will have ML on several other DIGIC 6/7/8 models.  This will keep the interest in ML alive and will attract many new users that own such cameras.  The interest in ML will grow rapidly worldwide.

2) Nikon's engineers are working very hard on implementing RAW video in their latest mirrorless Z6 and Z7 models. 

https://www.diyphotography.net/12-bit-raw-video-is-coming-to-the-z6-z7-and-nikon-dslrs-to-get-cfexpress-support/

Once RAW video is officially working on them, many people worldwide will become aware of its quality and filmic look and will want to shoot their family and amateur videos in RAW.  People are always inclined to want and buy the best!  However, how many of them will be able to afford these models?  The body alone costs about 3.4K US$, add another 1.4k for the Atomos Ninja recorder and another 5k for the Z-lenses and adapters and you easily end up with 10k for a system like that.  It is quite clear that only professional film makers will be able to afford such gear.  Consumers will look for less expensive solutions.  But are these available from other brands?  No and it will take quite some time until other companies develop and release them.  And even if this happens, how many years will these companies take to get where we are with ML now?  Until then, the only reasonable alternative for RAW video will be Magic Lantern.  Nowadays, you can shoot 2.5K RAW video with a cheap Canon camera like the EOS-M that costs in the 100-150 US$ range used and you can use almost any lens with it.   So, I expect prices of old ML-capable Canon cameras to sky rocket and the interest in ML to further increase dramatically!

3)  And what about all the ML science, developed so successfully by A1ex and other developers and implemented and now working in most of the ML-capable cameras?  What about Dual ISO, noise reduction techniques, sensor reading, pixel binning techniques, card overclocking and spanning, ISO research, lossless compression and so much other innovative stuff?  What about MLV Producer and MLVApp? Isn't all this new knowledge and know-how the most important aspect of Magic Lantern?  All this will be available for generations to come and I am sure, many companies will take advantage of this new knowledge and will implement it in their new models.

So, my suggestion to A1ex and the Community is:  Let's keep Magic Lantern going and it will have a great future!  This statement may be a little overoptimistic but I truly believe in it!


nikfreak

In my opinion we should invest into those SONY IMX sensors used in smartphones and team up to do what Google does but use AI (read: a1ex intelligence not artificial intelligence  :D ;D :D). Let's put a dozen of those sensors into a small body and chime in a snapdragon cpu use Android as a base, do some RE on Pixel devices and let's forget about apertus, canon, nikon and whomever and make our own camera. Google will win ofc because they know better and prove that single lens setup beats all those triple / quadrouple camera setups with Google AI but hey we can place 2nd after Google, right?
everyone's after a GCAM port on his smartphone. Why not announce an "A1ex app" trying to bring revolutionary features like dualiso from ML to smartphones etc..?? Millions would buy it later in Play Store. Lots of people would support only the idea - that announcement.
a1ex just your name would bring you enough $$ to do whatever you want to do as a full time job if you would announce something like above. Don't be shy. No risk no fun and thousands of people would spend a whole lot of money onto any kickstarter project you would anounce. I am not talking about an unbricking service for cameras. You got THAT famous name and are freaky enough to spend another thousands sleepless nights to do whatever magic you have in your mind.

This community alone has enough people willing to help in anything you decide to do.
[size=8pt]70D.112 & 100D.101[/size]

ilia3101

Z cam image looks lik e crap. Similar to mlv app at max chroma blur.

nikfreak

okay okay let's use the unbricking service but make it a converting service for existing cameras: replace sdcard / CF card with flash / UFS storage. Some EOSM / 5D3 with ulrafast 512GB or 1TB flash storage would be nice.
[size=8pt]70D.112 & 100D.101[/size]

nikfreak

maybe this one???
Quote from: nikfreak on July 20, 2017, 12:29:56 PM
If all goes wrong you can use Canon sensors together with a1ex's qemu work and a raspbi and call it A1EXIOM  :P
[size=8pt]70D.112 & 100D.101[/size]

a1ex

Quote from: nikfreak on April 03, 2019, 09:26:23 PM
but hey we can place 2nd after Google, right?

That's quite a bold statement, but I don't consider it impossible.

So far, I've chosen to stay away from the smartphone world, mostly for "comfort zone" reasons. I've actually played with an Android phone for a few weeks, installed LineageOS & stuff, even was halfway successful in running a kernel exploit, and... went back to my old Nokia with over a week of battery life :D

My impression is that Android market is quite saturated, and a lot more fragmented than the EOS ecosystem (so many different models to consider), but... I'm an outsider there. No idea at all about iOS. I'm not totally opposing this idea - I actually wanted to check the camera internals of smartphones before (curiosity), but didn't do any step in this direction.

Yes, I'm living under a rock; how else do you think I could achieve all of that with ML? :D

nikfreak

Let's sum it up. If you ever do a D6+ port on real hardware I promise to do another one too (read: I don't give an ETA)
[size=8pt]70D.112 & 100D.101[/size]

nikfreak

Oh btw regarding Android. people just make 'nuff money only by cherry-picking kernel commits and releasing a kernel and additionaly paid kernel app to manage it. Ever heard about Francisco Franco? Check the amount of installations:

https://play.google.com/store/apps/details?id=com.franco.kernel&hl=de
[size=8pt]70D.112 & 100D.101[/size]

jackmoro

Quote from: nikfreak on April 03, 2019, 10:22:37 PM
Let's sum it up. If you ever do a D6+ port on real hardware I promise to do another one too (read: I don't give an ETA)

WORD 8)

scrax

Quote from: a1ex on April 03, 2019, 10:18:13 PM


So far, I've chosen to stay away from the smartphone world, mostly for "comfort zone" reasons. I've actually played with an Android phone for a few weeks, installed LineageOS & stuff, even was halfway successful in running a kernel exploit, and... went back to my old Nokia with over a week of battery life :D

My impression is that Android market is quite saturated, and a lot more fragmented than the EOS ecosystem (so many different models to consider), but... I'm an outsider there. No idea at all about iOS. I'm not totally opposing this idea - I actually wanted to check the camera internals of smartphones before (curiosity), but didn't do any step in this direction.

Yes, I'm living under a rock; how else do you think I could achieve all of that with ML? :D

You are totally right, android is oversaturated and camera users knows only gcam is best. They don't even know how to compare two images from two different camera apps, also if you use Android camera api quality is same across all app basically only post process work can give some differences because the basic files that controls the "microsensors" of those ridiculous camera.

And about diversity and models the communities in android word are changing faster because they are almost all device specific, you don't follow a lot works and feat. of other phones than yours and people change phone more frequently than dslr (supposing) and to make money (like Franco) you have to develop for the biggest userbase of that moment, and usually the bigger the user base the worst the community is

Quotewent back to my old Nokia with over a week of battery life
plus smaller and toughter, I went back to my old Nokia 1208 after playing around with some xiaomi

Attached the camera lens and sensor package from my broke up smartphone
I'm using ML2.3 for photography with:
EOS 600DML | EOS 400Dplus | EOS 5D MLbeta5- EF 100mm f/2.8 USM Macro  - EF-S 17-85mm f4-5.6 IS USM - EF 70-200mm f/4 L USM - 580EXII - OsX, PS, LR, RawTherapee, LightZone -no video experience-

ilia3101

Quote from: Ilia3101 on April 03, 2019, 10:07:47 PM
Z cam image looks lik e crap. Similar to mlv app at max chroma blur.

The post that was a reply to has disappeared??? What happened on this thread ???

dfort

Ha ha -- went into the "Hall of Shame" a.k.a. SPAM.

QuoteRe: Guessing 1st April for 2019 - Share your thoughts by jackmoro (Hall of Shame)

Looks like this topic jumped the shark.

KelvinK

Quote from: dfort on April 04, 2019, 07:00:56 AM
Ha ha -- went into the "Hall of Shame" a.k.a. SPAM.

Looks like this topic jumped the shark.

No, I guess he just pressed "delete" instead of modify  :D Happened to me once.
6D - 5D - NEX - M50!

chris_overseas

Quote from: KelvinK on April 04, 2019, 09:24:35 AM
No, I guess he just pressed "delete" instead of modify  :D Happened to me once.

Yes I can confirm that if you delete your own message it goes to a hidden "Hall of Shame" section, which is the same place all the spam ultimately ends up. It can be confusing initially when moderating to see seemingly good posts mixed in with the spam (apologies to a1ex who had to help clean things up that time I restored a bunch of seemingly good posts that had in fact been intentionally deleted!).

If anyone does accidentally delete something substantial, feel free to drop me a PM and I'll restore it for you.
EOS R5 1.1.0 | Canon 16-35mm f4.0L | Tamron SP 24-70mm f/2.8 Di VC USD G2 | Canon 70-200mm f2.8L IS II | Canon 100-400mm f4.5-5.6L II | Canon 800mm f5.6L | Canon 100mm f2.8L macro | Sigma 14mm f/1.8 DG HSM Art | Yongnuo YN600EX-RT II

12georgiadis


theBilalFakhouri


theBilalFakhouri

Well I am not pretty sure what's a1ex thoughts about paid goals for ML, but If you can make a Patreon account (You because we trust you and it is difficult to make these things in my country Syria e.g. PayPal and VisaCard) for the things beyond our reach like MLVApp with GPU Accelerated by making an offer on freelancer paid by ML community and so on.

Or let's start with very simple task and it doesn't require high paid price, Giving cr2hdr GPU support with CUDA and OpenCL.

That's can bring some more experiences here.

Not sure if someone will feel this is not fair, because the main developers are not getting paid unlike the second developers in freelancer if that happened, if I was the developer of these free open source Apps and tools I will don't mind if the money can push it forward.

Still waiting some thoughts about this subject, it can be big improvement and step forward for ML project .

alpicat

It's possible to do a Patreon account without having goals or tiers. What could be easiest in this case is to set up a page and just have people contribute whatever monthly amount they want - without the pressure of having specific goals.

On the page you could say something along the following lines: pledges won't result in specific Magic Lantern features appearing any faster than is currently the case, but it will help keep the project going.

I think people would be interested in supporting this given the amount of Magic Lantern users. Maybe you can start out this way, see how it goes and then the Patreon page can always evolve and add goals in future if required.

If A1ex isn't comfortable handling donations for Magic Lantern as he mentioned, maybe the focus of the patreon could be more about supporting A1ex as a person? (i.e. the idea being that donations could eventually allow him to spend more time on programming, and he can decide to distribute part of the donations towards other Magic Lantern developers when required).

These are my thoughts, don't know if it makes sense though.

a1ex

Quote from: alpicat on April 06, 2019, 04:45:51 PM
If A1ex isn't comfortable handling donations for Magic Lantern as he mentioned, maybe the focus of the patreon could be more about supporting A1ex as a person?

Yes, that's what I'm thinking. Handling donations for the entire ML project puts too much risk on me, so it's maybe better to let each individual developer (me included) decide what's best for them. Should they go with Patreon or Freelancer or whatever, that would be their decision and they will be on their own (such as taking full responsibility for delivering whatever they promise, figuring out what taxes to pay and so on).

However, there were strong opinions against money, particularly in this thread, which are holding me back. Not sure what to do. I was thinking to accept money for something unrelated to ML (such as the repair services, but only for the kind of errors I already know how to diagnose, and only as remote assistance, to minimize the time spent of these -- but that already makes it a very small niche). Or, for Android development - except... I'd have to learn it from scratch, and it may take a lot of time (for me) to get the knowledge required for a top-notch app. I'm afraid this route would slow me down even further on the ML side.

Quote
On the page you could say something along the following lines: pledges won't result in specific Magic Lantern features appearing any faster than is currently the case, but it will help keep the project going.

In my case, they will - as this kind of support means I'll be able to spend significantly more time on ML. Still, some things are genuinely hard, so I cannot promise I'll do X, Y or Z within a given time frame (such as one week or month or whatever). I can only promise I'm going to work on the things requested by the supporter.

According to the linked thread however, this (i.e. some developers accepting money) may demotivate other developers, and that's exactly what I want to avoid. Of course, me being unable to integrate their contributions in a timely manner... is not going to motivate them either. Here's an example from another open source project.

Another good thread to review - here.

After all the feedback - thanks, folks! - I'm considering a Patreon page; just trying to get used to the idea.

alpicat

Yes this is definitely more complex than I thought!

It probably would make sense to keep your Patreon page quite general in that case - as said it could be more about supporting you as a person so you can spend more time on your ideas, whatever they may be; but mention to your supporters that you cannot promise you'll be able to do any of these things (be it android app development or anything magic lantern related) - simply because some things are too difficult. I think people would understand that.

nikfreak

set up that Patreon page @a1ex. Just do it!

Quote from: a1ex on April 06, 2019, 06:19:55 PM
..may demotivate other developers, and that's exactly what I want to avoid. Of course, me being unable to integrate their contributions in a timely manner... is not going to motivate them either...

Don't think about others. As I said earlier. It's you, your famous name and you doing 99,9% of all reverse enginering at the moment. Think about it like this. If a Patreon page can pay the bills and motivate you to do more for ML then the end result for ML could be better support in Digic 6/7/8 etc. That again will lead to more traffic to the community site and again to more followers and again to more reputation and that again will motivate new developers to join the community.

A win-win situation for all.
[size=8pt]70D.112 & 100D.101[/size]

jackmoro

Quote from: nikfreak on April 06, 2019, 08:36:22 PM
set up that Patreon page @a1ex. Just do it!

A win-win situation for all.

totally agree!

Greg

Alex, if you need energy try this (lime, lemon, blue is a special mixture only for devs) :


IDA_ML

I final thought on my part:

In my opinion, one of the major keys to the success of Magic Lantern has always been the fact that it is a noncommercial project.  Every time, money gets involved, chances that something goes wrong, increase dramatically.  This is because supporters expect and demand and recipients feel the pressure to satisfy their supporters, although very often they do not have the capacity and\or abilities to do so.  It is extremely hard to satisfy everybody when money is involved.  Moreover, even if some of the above suggestions raise some money, I doubt it that it will be enough to substantially help developers increasing their living standard.  Therefore, I fully agree that every developer should decide individually what works best for them.

A1ex, if I were you, I would keep my current job, slow down a bit with the hobby projects, dedicate more time to my family, stay with us and keep going with ML in the way you have been doing this sofar, maybe considering some of the suggestions from this thread.  Read them again, please and give them some additional thought.

There are two more things that come to mind:

A. Every time users decide to download a new built or something else from the ML site, a message should pop up, reminding them that they should donate to ML.  This will intensify the donation campaign.

B.  Every developer and skilled user who has come up with new knowledge and ideas that have been implemented and verified experimentally in ML, should consider filing a patent on it.  This should again be a personal decision but on the long run, it could bring substantial amounts of money to the authors.  Filing a patent is not cheap but if the money collected through donations in #A is used for that purpose, this could help the authors quite a bit.

There is an old saying:

"If something works, don't touch it and don't change it!"

So, let's stick to this saying and keep going with Magic Lantern ...

Danne

This project needs funding for the heavy ongoing groundwork being done by a1ex. Without this I expect this project to slowly fall into sleep.
I guess donations could raise questions and there's some uncertainty to this process but in whole I think it's fully possible to keep this place more or less the same in other aspects.
Imagine upcoming cams coming "alive" :D