Zoom feature? How?

What a truckload of Apple Fanboy horse manure. :rolleyes:

Android runs very solid when not loaded with carrier bloatware. Samsung and Sprint are notorious for running as many services on their devices as possible. The crying baby monitor service for example. C'mon meow...u dont need this to run by default.

Just root and be done with all that nonsense.
 
Yes, it works fine on my P4, I like it. When recording video in 1080 you have up to a 2X zoom capability, and it doesn't lose much clarity, even though it's digital zoom. I think this is because they are cropping the 4K sensor, so the result is a native 1080, it looks pretty darn good. If you're using 2.7K resolution to video, the zoom is less. The only downside is the zoom action is not smooth, it looks really jerky if you leave the zoom action in the video. Most will edit out the zoom in and out action since it looks so cheesy. If you touch and spread your fingers fast, it will go from 1X to 2X instantly, which doesn't look bad, and vice versa, pinch FAST to go from 2X to 1X is OK too. This leaves out the jerky steps from 1.0 to 1.2, to 1.3, etc.
Guys, don't you understand there is ZERO difference than just shooting in 4K and then zooming into the image.

If you take a 4k image and put it into a 1080p sequence, you can (without losing any quality) zoom in 2x, move it right and left and all over the place without losing ANY quality because you are shooting at 4K.

All you are doing by doing it in the app is losing the ability to choose where you want that to happen.

Shoot at 4k and then just zoom into whatever you want and you will have zero degradation if you are in a 1080p sequence. You can even zoom in more if you are at 480p.

This is just marketing. It's the same as "digital lossless zoom". It's a marketing term. There is NO SUCH thing as digital lossless zoom. There is only such a thing as shooting at a larger image than you are delivering and having the ability to zoom in digitally without any quality loss because the image has more resolution than your sequence.

***The ONLY time this would have some benefit is if you aren't bringing your footage into an editor.

If you are bringing it into an editor, than don't rely on zooming in the bird which a) is busy recording and might cause you more artifacts and b) serves no purpose.

If you wanted to have 2x digital lossless zoom at 4k you would need to be shooting at 8k if that helps. If I am missing a reason as to why this is important to someone than I apologize but there is nothing here that you couldn't do before. A rose by any other name is still just taking an image larger than your sequence and having room to move around in. It's also good to shoot at 4k, even if you are going into a 1080p sequence because you can punch in in case props get in the image.

Why not record in 4K and then do any zooming in post? Your final output can be 1080p and you'll have 2X zoom capability at any time. Many video software will do this.

Bruce Webber is absolutely correct. In fact, you don't eve need to do a straight zoom. You can move it left or right, and as long as you are at 200% increase in size, you get zero degradation.

Also, I like flying the phantom 4 for fpv purposes. I could care less about capturing video(I rarely record). I know many people purchase products like the phantom solely as a camera platform. I fly because of the the enjoyment of flight. Anything that makes flying more enjoyable is a plus. Been flying rc for 20+ years. Didn't but this for the recoding ability. It's just all around the best product for hd fpv with video goggles.
Fair enough on this reason. I am a little surprised to find out someone buys the P4 for flying FPV. It is fun to fly so maybe I shouldn't.
 
Last edited:
Guys, don't you understand there is ZERO difference than just shooting in 4K and then zooming into the image.
In the narrow manner you clearly mean this (as indicated in the balance of your post), you're right.

However, there is a very real issue of framing when shooting live, and without this real-time visual feedback it's actually possible to MISS material you wanted in the shot, off one edge of the wider-field view.

It's hard to explain this, but it's a cognitive-perceptual issue having to do with most of what you're looking at being irrelevant (3/4 of it) while still having to process and make sense of it in the visual cortex.

The ideal solution is the ability to zoom the view in the display on the ground to a portion of the full-sensor image, while recording the entire 4K resolution. This allows the operator to properly frame and center the subject, while still retaining full resolution imagery. This also preserves the advantage of being able to change the framing or even the subject in post-processing for a zoomed clip.
 
  • Like
Reactions: John Locke
In the narrow manner you clearly mean this (as indicated in the balance of your post), you're right.

However, there is a very real issue of framing when shooting live, and without this real-time visual feedback it's actually possible to MISS material you wanted in the shot, off one edge of the wider-field view.

It's hard to explain this, but it's a cognitive-perceptual issue having to do with most of what you're looking at being irrelevant (3/4 of it) while still having to process and make sense of it in the visual cortex.

The ideal solution is the ability to zoom the view in the display on the ground to a portion of the full-sensor image, while recording the entire 4K resolution. This allows the operator to properly frame and center the subject, while still retaining full resolution imagery. This also preserves the advantage of being able to change the framing or even the subject in post-processing for a zoomed clip.
Your point is well taken, if not a little snarky, but it doesn't give you any more or less room to reframe, you'll have the same exact amount of resolution.

I've personally never found myself in a position where I wished I could zoom in by 2 in order to see what I want to frame later so I wasn't able to relate to it but if it helps someone then I guess it's a valid reason. Someone else pointed out that they want it for FPV and you would see that my answer was "fair enough, I get your point".

This is a forum, meant for discussion and I am all ears to other people's desires as maybe there are some things that I didn't think about and could use for myself. In my years of using cameras, aerial or not, I never had or needed this desire. If I wanted to get closer, I pan, walk, or optically zoom, or in this case fly forward.

I can see what your point is, just it doesn't ring as something I would ever want. The FPV made sense to me, this, not as much but that doesn't mean it's not a reason. Obviously people wanted it. Feels more like marketing to me though unless the FPV reason which is a good one.

EDIT: Actually, I would feel the opposite of what you're saying as far as framing. This is all moot for me because I am always shooting for 4K but I would feel more likely to MISS the shot if I was zoomed in digitally than if I was at the pulled back framing.

For example, lets say the principle is some building in my forward view and I was shooting for 1080p at 4k. I would just get the building as center as possible and then frame it in the editor. If I am zoomed in, what benefit does that give me for not missing the shot? I put the building in the middle and then I have the full latitude offered to me by 4k in a 1080p sequence whether zoomed in or not. I would feel safer about getting the building in the shot if I was pulled back and knew that I was nice and centered. Preference I guess and also habit.

You reasoning is fine if it works for you. I am a big proponent of "whatever works for you", not "this is what to do because it works for me". You said your piece and that's fine by me. :) No matter what you guys may think, I am always trying to learn and sometimes that means stepping out of your box so thanks for your thoughts on it.
 
Last edited:
Your point is well taken, if not a little snarky, but it doesn't give you any more or less room to reframe, you'll have the same exact amount of resolution.
Really sorry you took it that way... no snark was intended! Just sharing information.

As for reframing, it's not a resolution issue, it's the fact that being unable to properly frame can actually cause you to have the larger FOV positioned so that material you want in the shot is out of view.

This can happen because 3/4 of the material you are seeing is irrelevant to the shot, but your visual cortex has to process it and you have to consciously try and keep track of what is supposed to be in frame and out, without any framing rectangle on the screen or any other indication.

This results in mistakes more often that you might think, and for live footage there's no second chance.
 
Your point is well taken, if not a little snarky, but it doesn't give you any more or less room to reframe, you'll have the same exact amount of resolution.

I've personally never found myself in a position where I wished I could zoom in by 2 in order to see what I want to frame later so I wasn't able to relate to it but if it helps someone then I guess it's a valid reason. Someone else pointed out that they want it for FPV and you would see that my answer was "fair enough, I get your point".
Haven't you edited aerial videos and seen things you didn't see while flying? The 2X zoom could have helped. If you had zoomed in you could have seen it and spent more time featuring that POI that you missed.

Here are some examples of the utility of the 2X zoom in scenarios that post edit zoom and crop will not benefit.
1. Inspecting antenna towers for birds nests, rats nests or any obvious wind or storm damage. Close up zoom view while you're flying helps to know where to spend more time inspecting with better angles for clarity.
2. Looking for animals on the ground, a close up real time view helps to see through the bushes better to locate animals hiding.
3. Staying a good distance away from people you want to record to not be rude, but you need to find the person you're looking to feature in the video. Safer flight too.
4. Recording rock climbers, finding a specific climber on the side of a mountain that you're looking for, zoom helps while not getting too close.
5. When recording a car zooming by on the highway, zooming in to identify the car you're looking to capture before it gets to you, so you track the right car as it goes by.
6. Inspecting bridges for structural flaws to take photos of the point of interest, seeing the problem areas during flight to spend more time taking photos of those problems.
7. Search and rescue. Time can be of the essence to find the lost person, there's no time to review in post edit. You would need a separate crew to review video while you're flying to cover more ground if you want to zoom during post edit.

What you can see while flying influences the flight agenda, the closer you see the more detail you see. The more you see, the more interesting the video can be. Seeing something in post that you didn't see while flying cannot be reshot. If you miss something you didn't see while flying, you've simply missed the opportunity to optimize the shot, spending more time, looking for a better angle. You can't turn back time and capture something you wish you would had seen during the flight.

Besides the utility of 2X zoom, I find it cumbersome to shoot in 4K all the time. I realize I lose the opportunity to post edit zoom and crop, however that's not as important as the SD card space, and 4K takes more time to edit. I just don't need 4K for my purposes, today. It works for me, today, maybe not tomorrow.
 
  • Like
Reactions: Jussaguy and Myetkt
Haven't you edited aerial videos and seen things you didn't see while flying? The 2X zoom could have helped. If you had zoomed in you could have seen it and spent more time featuring that POI that you missed.

Here are some examples of the utility of the 2X zoom in scenarios that post edit zoom and crop will not benefit.
1. Inspecting antenna towers for birds nests, rats nests or any obvious wind or storm damage. Close up zoom view while you're flying helps to know where to spend more time inspecting with better angles for clarity.
2. Looking for animals on the ground, a close up real time view helps to see through the bushes better to locate animals hiding.
3. Staying a good distance away from people you want to record to not be rude, but you need to find the person you're looking to feature in the video. Safer flight too.
4. Recording rock climbers, finding a specific climber on the side of a mountain that you're looking for, zoom helps while not getting too close.
5. When recording a car zooming by on the highway, zooming in to identify the car you're looking to capture before it gets to you, so you track the right car as it goes by.
6. Inspecting bridges for structural flaws to take photos of the point of interest, seeing the problem areas during flight to spend more time taking photos of those problems.
7. Search and rescue. Time can be of the essence to find the lost person, there's no time to review in post edit. You would need a separate crew to review video while you're flying to cover more ground if you want to zoom during post edit.

What you can see while flying influences the flight agenda, the closer you see the more detail you see. The more you see, the more interesting the video can be. Seeing something in post that you didn't see while flying cannot be reshot. If you miss something you didn't see while flying, you've simply missed the opportunity to optimize the shot, spending more time, looking for a better angle. You can't turn back time and capture something you wish you would had seen during the flight.

Besides the utility of 2X zoom, I find it cumbersome to shoot in 4K all the time. I realize I lose the opportunity to post edit zoom and crop, however that's not as important as the SD card space, and 4K takes more time to edit. I just don't need 4K for my purposes, today. It works for me, today, maybe not tomorrow.
All good reasons and not something I thought about.

When I fly my Phantom it's for fun and I am not usually trying to do any one of the things you listed so it didn't cross my mind but I suppose I can understand it. The one that I could see being something that would be useful to me is to find a person climbing a mountain or something without getting too close. I can understand that, otherwise, I would just fly in closer if I wanted to frame differently but I am ALWAYS shooting at 4K. I don't think I have shot at under 4k, or UHD (the other 4K) since owning a 4K Phantom going back to the Professional (and even on the advanced when I was shooting for 1080p, I would shoot at 2.7 when it became available so I could punch in which was the purpose of it as there is no 2.7k standard and I will almost guarantee a 4.7k or something a little larger than 4k on the P5 that will allow for the same reasoning). If I am outputting at 1080p which I also almost never do, I am still happy I shot at 4K so I can re-position or chop off blades that made their way into the shot.

As for the framing of people you don't want to get to close to, I can't completely relate. I shoot surfers all the time with my Phantom and have actually been getting a few offers from other surfers who want my services and when I have extra time, I am happy to do it, and I like to stay as far back as possible, and am usually shooting for 1080p in this situation and if I was zoomed in, it would actually be harder for me to keep them in frame when I can pull back and do orbitals around them and just keep them in the center as much as possible. if I am zoomed in by 2, it would be harder to keep them in frame.

That said, the other question I have for you is that you don't edit 4K because you don't have the resources but when you are shooting for 1080p zoomed in, aren't you in fact shooting at 4K or does the software truncate the footage for you right out of the bird? If it does, I'd be interested to know if that extra operation effects image quality (EDIT: actually it would probably do it after the mission on second thought). I am curious as to what is going on underneath the hood when you are shooting for 1080p at a 4K resolution in that mode. I have no idea.

Your other reasons are all valid and now with that and the other reasons, I can completely see why some would want to zoom in. For my purposes, I can't see a situation where I would use it, I would just fly to the spot I wanted as nothing that I am shooting with my Phantom would fall under the things you listed but I get them and thanks for laying them out for me. Much appreciated John. You are always good about this stuff and intelligent. I can dig it baby. :)

Adam
 
  • Like
Reactions: Myetkt
That said, the other question I have for you is that you don't edit 4K because you don't have the resources but when you are shooting for 1080p zoomed in, aren't you in fact shooting at 4K or does the software truncate the footage for you right out of the bird? If it does, I'd be interested to know if that extra operation effects image quality (EDIT: actually it would probably do it after the mission on second thought). I am curious as to what is going on underneath the hood when you are shooting for 1080p at a 4K resolution in that mode. I have no idea.
The camera actually crops it and converts it to 1080 and records it to the SD as 1080, no matter if you are 1X, 1.5X or 2X with the zoom. The quality is good, I can hardly tell the difference between zoomed and native 1080 when capturing zoomed video in 1080. I think that's why they call it Lossless Zoom, because when shooting 1080 resolution it's always 1080 and clear, zoomed or not. It's definitely not traditional digital zoom. The biggest downside is during the zoom power increase, it's not smooth like an optical zoom. It jumps in steps as you pinch or spread to zoom in or out, so those must be edited out.
 
The camera actually crops it and converts it to 1080 and records it to the SD as 1080, no matter if you are 1X, 1.5X or 2X with the zoom. The quality is good, I can hardly tell the difference between zoomed and native 1080 when capturing zoomed video in 1080. I think that's why they call it Lossless Zoom, because when shooting 1080 resolution it's always 1080 and clear, zoomed or not. It's definitely not traditional digital zoom. The biggest downside is during the zoom power increase, it's not smooth like an optical zoom. It jumps in steps as you pinch or spread to zoom in or out, so those must be edited out.
What I am saying is that I think it is not doing the conversion on the fly. I think it does it after the mission. Otherwise, it would be using valuable computing power while recording to truncate the 4k to 1080. I might be wrong, it's just a guess, but what I think is happening is that it's recording at 4k on the fly and then it writes it at 1080p somehow after the fact. Either that or if it has that much computer power to spare, why can't we get some more megabits! :)
 
The imaging board has a hardware encoder on it, capable of several formats as well as basic scaling and window functions.

The camera sensor delivers a 4K image's worth of pixels. When recording in 1080p the image is reduced in resolution via real-time scaling, and then encoded into H264 in real-time and written to the SD card.

When zooming, the encoder is configured to extract a subset of pixels (a window) and then encode them at the configured format, scaling if necessary. This is why the zoom function isn't smooth, but instead has a few discrete zoom values: That's what the encoder SOC in the P4 supports.

Dedicated SOCs capable of real-time 4K video encoding are pretty common nowadays, and eliminate the need for a beefy general-purpose CPU to do the encoding in software in embedded applications like video recorders. Surely you didn't think 4K camcorders were doing post-processing for hours to encode recorded video -- same situation. Dedicated encoding hardware.
 
  • Like
Reactions: Jussaguy and Myetkt
The imaging board has a hardware encoder on it, capable of several formats as well as basic scaling and window functions.

The camera sensor delivers a 4K image's worth of pixels. When recording in 1080p the image is reduced in resolution via real-time scaling, and then encoded into H264 in real-time and written to the SD card.

When zooming, the encoder is configured to extract a subset of pixels (a window) and then encode them at the configured format, scaling if necessary. This is why the zoom function isn't smooth, but instead has a few discrete zoom values: That's what the encoder SOC in the P4 supports.

Dedicated SOCs capable of real-time 4K video encoding are pretty common nowadays, and eliminate the need for a beefy general-purpose CPU to do the encoding in software in embedded applications like video recorders. Surely you didn't think 4K camcorders were doing post-processing for hours to encode recorded video -- same situation. Dedicated encoding hardware.
Yeah, I know there is an avc H264 encoder on board of course. I am not saying I know what it's doing. I was just saying I would be surprised if it was doing an extra format operation on the fly. I will post more about what I think but am reading bed time stories to my baby girl.

Good post though.
 
Yeah, I know there is an avc H264 encoder on board of course. I am not saying I know what it's doing. I was just saying I would be surprised if it was doing an extra format operation on the fly. I will post more about what I think but am reading bed time stories to my baby girl.

Good post though.
Yes... you may not be aware that hardware video encoders have more processing functionality in them than simply video compression. They all include basic "editing" (i.e. scaling, cropping, etc.) capabilities in real-time as part of the encoding pipeline.

That's all I'm trying to share. You seem a bit hostile to people responding to your speculations with real facts and information. Honestly, I don't have the interest to share this stuff and be called "snarky", or have a current of resentment obvious in every post you make in response.

I'm outta this thread.
 
  • Like
Reactions: Myetkt
Yes... you may not be aware that hardware video encoders have more processing functionality in them than simply video compression. They all include basic "editing" (i.e. scaling, cropping, etc.) capabilities in real-time as part of the encoding pipeline.

That's all I'm trying to share. You seem a bit hostile to people responding to your speculations with real facts and information. Honestly, I don't have the interest to share this stuff and be called "snarky", or have a current of resentment obvious in every post you make in response.

I'm outta this thread.
Huh? Is there something that is showing up after my posts that I don't know about? I said "I will add more later and also 'Good post though'.

So bizarre. Also, it's not my thread.
 
Yes... you may not be aware that hardware video encoders have more processing functionality in them than simply video compression. They all include basic "editing" (i.e. scaling, cropping, etc.) capabilities in real-time as part of the encoding pipeline.

That's all I'm trying to share. You seem a bit hostile to people responding to your speculations with real facts and information. Honestly, I don't have the interest to share this stuff and be called "snarky", or have a current of resentment obvious in every post you make in response.

I'm outta this thread.
Ok, now that I have some time, I did not think your post was snarky at all. I was half kidding on a post about a page or two ago and I was sort of just saying it in jest and I thought, to be honest, that you were being that but it was probably because of the post in my crash thread. All that said, I didn't think anything at all of your post about the AVC encoder doing the leg work to crop the 4K image on the fly.

In fact, I took it as informative and when I said "I would post more later" it was because I wanted to know what you know. Like do you know for a fact that it has the processing power in that little P4 AVC to do that work? It would seem to me that it would be more likely that it would do it after the fact so I was literally just asking as you seem to have knowledge on it when I have none. I would like to know. It seems more logical that they would save easy processing for after the fact rather than try to cram more already bottle necked data down the minimum bandwidth provided by the P4 which has a hard time writing already at 60mbps even with the fastest writing cards so I am sorry if you took my post as anything more than that. It definitely wasn't meant to be.

So now that it is out of the way, I do know about encoders and some encoders indeed have the power to do a ton of extra work. I've worked with dedicated encoder farms so I have seen in action what a dedicated encoder can do. I have been around digital cinematography since it was being introduced. I worked on the machines that were the precursors to Avid Media Composer and have worked extensively with OMFI, XML, MXF and all types of software and hardware media encoders and am just curious as to what is happening underneath the hood. Do you know for a fact that the AVC encoder is scaling it on the fly? I wouldn't be completely surprised or anything, it just seems like there would be a way to do it after the fact and not tax the bird anymore while it's working on recording things. Why not have it process after it has at the very minimum buffered the write?

I am just throwing it out there as a thought. I didn't even know it had the functionality until this thread as I haven't updated my firmware since the last time everything was working wonderfully on my Phantoms and also, I can't ever see myself wanting to use the function as I am always shooting for 4K.

Anyhow, sorry I said snarky on that one post if you weren't being that. I didn't really care either. It was just a poke. I don't get offended easily.
 

Members online

No members online now.

Forum statistics

Threads
143,066
Messages
1,467,359
Members
104,936
Latest member
hirehackers