ฝัง
- เผยแพร่เมื่อ 30 พ.ค. 2024
- You've seen us shoot anamorphic on the iPhone 15. So what does it take to make iPhone footage look cinematic? Is it the lens? The crew? The lights? In this episode, we go in depth into all of these things. Most importantly, we experiment with dynamic range on a XYLA chart - what it is, how to measure it, and why it matters. As always, thanks for your comments and questions - don't forget to subscribe and stay tuned for our next episode!
0:00 Intro
0:15 This is a DSC XYLA Step Chart
0:34 How Good is Apple Log?
1:02 Our Anamorphic Footage on an IPhone 15
1:30 Our First Cinematic iPhone Experiment
2:09 Our Second iPhone Experiment: AI DOF
2:29 Apple Pro Res in Log
2:45 Atlas Anamorphic Lenses x BeastGrip DOF Adaptor
3:50 A Talented Crew Makes A Great Image - Gaffer Jacob Abrams
4:30 Crew + Tech
5:42 Dynamic Range is Key: What is it and how to measure it
5:15 The DSC XYLA Step Chart
6:05 How to Measure Light - Linear vs Logarithmic
7:40 Low Dynamic Range vs. High Dynamic Range
7:55 iPhone 6 Dynamic Range
8:59 iPhone 13 Dynamic Range
9:14 iPhone 14 Dynamic range
9:50 RED One Dynamic Range
10:25 The Social Network
11:04 iPhone 15 Dynamic Range
11:13 Apple Log
11:24 Log Encoding Explained
12:02 iPhone 15 is the first iPhone to introduce Log
13:00 iPhone 6 vs. iPhone 15
13:32 RED Raptor can see 18 Stops
14:15 Gaffer Jacob Abrams talks about iPhone vs. Red
15:30 Talent lives at the heart of any production
15:50 iPhone is the most readily available camera on a global scale
16:50 How can innovation add value to your craft? - ภาพยนตร์และแอนิเมชัน
You guys should try outdoor lighting now!!
We're thinking of our next analysis and are going to do outdoors! Gonna try something new as usual!
iPhone 15 has solid 14 stops … Gerold Undone is rolling his eyes 😂
I saw 15 stops bro. That's mad. I only know it's mad because of how you expressed that the graph should be plotted logarithmically not linearly. Mad respect, these videos are like Masterclasses.
I am astonished on how good this video was, how well and easy you explained everything and made the video feel like half its actual time. You guys are great!
Top tier content, this better get millions views
12:38 if you just want to see the dynamic range charts.
I've been a huge fan of Beastgrip since 2021 with their Pro and DOF 2. The DOF 3 is amazing and it''s the reason I will continue upgrading my phone rather than upgrading my camera.
Please keep on creating these amazing videos!
These RESULTS do not surprise me at all...
Back in 2009 ...
I was in a movie being shot by the NEW Canon 5 D Mark II
I was used to the BIG BULKY REEL FILM...
This was amazing because I have always loved Canon Cameras...
BUT...
What was more amazing at that time is... The person that was doing "B" Roll and Continuity... was using an iPhone 4 ... Yes, you read that right... an iPhone 4.
The film company at that time found out, they could USE the iPhone 4 for some shots and truthfully you could not tell the difference from the Canon 5D Mark II than the iPhone 4!!! (This is only a few seconds of the iPhone 4).
Right then and there, I KNEW one of these day... (waiting on iPhone) to make a Phone that could be used to make a MOVIE... Amazing...
Now... I am an Acting Director/Teacher and want to use an iPhone for a small film project (just kidding... As Big As I Can)...
Thank you so much for sharing this GREAT Video...
This is a Game Changer...
Strada Strada Strada...How many times did i cal you? Why do you keep making videos that set off fireworks in my mind. hehh...
EXACTLY
Great job, guys! Enjoyed watching it
So... it goes without saying ... that y'all (yes I'm a Southern Lady)
Will work with me on my Film when I get ready to shoot.
I am working on the Screenplay right now... I already have the manuscript done.
Waiting on moving back to Albuquerque, New Mexico to RE-Open up my Acting Academy.
I was going to purchase a NEWer iPhone to showcase/Teach my students how they too, can make a Film on an iPhone. (I only have the iPhone 8 right now).
I will keep you posted.
AGAIN... Thank YOU ALL so much for making this GREAT video...
I am so EXCITED NOW... I have been waiting for this day for so long...
loving this iphone 15 camera content. please continue pushing the camera’s boundaries 👍
love this channel, keep pushing the iphone to its limits
I'm starting to think that the best trait of this channel has also led to the stagnant subscriber count on youtube. You guy should be having millions of followers. Perhaps, youtube bundles you together with the avg "camera reviewer youtuber", whereas TH-camrs like PotatoJet who do the same thing has a serious Sub count and I say that because you're professional videos don't have a "personal POV" type of touch. Also the indoor Teal and Orange color pallette.
Brilliant... thank you for posting.
And how about color accuracy at 14 stops on iPhone? Or at least 7 stop? If you have professional lighting you can have good result with any phone
One more thing to consider...
My Screenplay/Movie... will be set in 1959...
This will allow the iPhone to LOOK, like it was shot in 1959.
35mm, grainy look... that way the iPhone will not have to work so hard to look Perfectly Crisp. Which no movie wants.. (smile).
Just great episode. Thanks!
awesome! always great videos from you guys
I could say this video will be great just from the 1st second of watching !
Laughs in LG cine log back in 2016
The LG had log, but not in 10 bit 422.
I love these videos. Professionally done and well thought out. It's amazing how far these cameras have come. Keep it up, guys!
From Smartphone + camera to Camerone.
With this quality of content and production, how do you only have 10,600 subscribers? Is this an alternate universe or something? What am I missing here? Your content and production quality are fantastic!
I love this vid :) It seems the Iphone is cleanest and has more dynamic range around 1280 ISO (In Magic Lantern app) which may be considered the iphone logs Native ISO, and for whatever reason H.265 is often less noisy then Pro Res and can help with the Dynamic Range scores.
I’d be curious how this compares to sensors used in scientific contexts and medical imaging.
Excellent video as always guys!
The time and effort put into this video to get this level of quality is just incredible. Great work, really enjoyed watching this. Once rigged out to match a full frame sensor image, how big is the footprint of the iPhone right now to the RED? I imagine it isn’t much difference?!
Thank you for such a detailed review and sharing you insights 💯
Thank you for making sure me to purchase this iphone 15 pro
We're so close to September...maybe you wait till the iPhone 16???
Thanks for such a detailed and informative review on the dynamic range of the Iphones and progress of the dynamic range in the past few years with respect to innovation in the technology. It really inspires to bring the innovations in my craft of filmmakimg. Please keep making such kind of videos.
Thanks for saying that. Really appreciate it:)
We've got more to come for sure and a TON of great episodes from when we started last year!
Great vid, great advice.
The one good thing that we CAN use the iPhone... is we want our Movies to have that 35mm look... a little grainy... So it doesn't look Perfectly CLEAR..
Michael, your insights are so motivational! Let's also remember apple is a media ecosystem. A huge benefit for camera to cloud and new applications like Strada. I fully agree, if we pay attention to what's happening today, it's a clear path to what's on the horizon. By the way, is it confirmed if the sensors apple are using for the 15 are from Sony? I don't believe apple makes their own sensors. Love the creativity!
Great points! I believe apple has been public on sharing the SONY sensor data and are using the CPU to leverage every ounce of quality out of it (hence 10bit sampling, ProRes, etc)
Thanks for the comment! Be sure to check out old episodes, too! Lots to learn here!
Aren’t those charts symbolizing Log for exponential, not logarithmic?
You're correct at 11:45. These show exponential which is how we can fit exponential light capture into a fixed gamma space.
Is the audio synced? I think its a little faster then the video
Brilliant ❤️
Any idea if the iPhone is using computational videography when it comes to the dynamic range it captures?
(Not that it makes much difference in use, but I’m just curious if this ‘is’ the DR of the sensor, or if it’s using something akin to how the iPhone stacks images to create HDR photographs).
Thanks so much! Loved the video! & mostly the level of professionalism and expertise here! Subbed!
11:50 a logarithmic graph is actually bent to the opposite direction,as the curvature will be downwards
You're right on the logarithmic graph. What I've found is when you map the RESULTS of a camera linear light into logarithmic (video) gamma you end up with an X axis that curves upward. So my example here is showing the how plotting dynamic range results would be measured. However it also depends on what you put on the Y axis (and different tests may plot different values on X vs Y. Great comment! Thank you!
Are you guys forgetting that the iPhone stacks multiple images in order to mimic hdr? That makes this test completely flawed. Also it’s just common sense that an iPhones tiny sensors will never compare to the large and sensitive sensors in dedicated cinema cameras, that’s just physics.
I'm not sure that's accurate in video mode. I suppose you could be right, but I don't think a dual-exposure is happening when recording ProRes in video mode. Please do let me know if you find evidence of this verified anywhere. There are often "signs" that give away the muxing of dual exposure. But I could be wrong! Thanks for the comment!
Not in log. Only in regular mode and hdr mode.
can you guys do this test on 35mm film cameras (celluloid film)?
Quality Content 👏
Outstanding!!! You described the excellence of Apple products as a whole. I have been using the products for over 30 years and if I'm shooting on a budget which means I don't have a cinematographer I use an iPhone. If you know how to light you can basically shoot with anything. As soon as I ask my producers to order something. They laugh and say, "we know, it's got something to do with lights!" Gaffer is God on the set and the best I have ever seen quietly go about their business creating a masterpiece of lighting designs and don't get nearly as much credit today as they should. Good stuff, keep up the good work of explaining the art of filmmaking and more importantly storytelling.
Wow! Thank you! Ava you seen Episodes 23 and 24? They really go deep. Plus episodes 11 and 18. If you like Apple, you'll like those episodes:)
Brain less sheep detected
Brilliant work and well explained!! Could you please do a video like this one the Canon EOS M with Magic Lantern
To demonstrate the dynamic range of standard shooting vs Magic Lantern RAW
12:47 guys i can see 15 stops! I'm watching this video on a macbook pro m1 pro on full brightness. I can see the video compression too but im pretty sure theres the half portion of the 15th stop outside the frame!
Edit: I did see 15 stops on iphone 14 plus too
Excuse my ignorance but how are you putting a cinema lens on top of the iphone lens itself? Thanks
Great question! Check out Episode 23 and 24 and we do a deep-drive on how we configured iPhone for a fully cinema setup!
12:02 The iphone 15 pro is the first phone to do internal prores and log encoding yes, but Lg and xiaomi and whole load of other android smartphone manufactures have been putting in their own log inplementations for a long time, the lg v30 having log and cine modes built in back in 2017. And ofc not to mention amazing work done by motioncam and it's team, allowing for up to 4k120 uncompressed raw video to be recorded on android phones and even built in cineon log encoded in prores, with much less denoising, sharpening, and no hdr unlike the iphone. Not to mention that certain android phones are now equipped with 1 inch sensors and some even having dual gain outputs. Would be lovely to see you guys mess around with something like the one plus 11 that has a big sensor good lens and super fast storage and processing enabling those 4k120 uncompressed raw
I've been following and utilizing motioncam across different android devices for years now. It's an incredible development. However, even that application suffers from overheating and crashes more often than not. Not totally reliable, even on the type 1.0 sensors. While the other devices you mentioned had log, they did not have 10bit 422. Only 420 encoding. The key to the iphone is the stability of the 4k log at 10bit 422.
That's right I was searching for this comment right after I heard the first phone with Log.
@Strada-Tech
One last note, the devices with "one inch" sensors as you say, are not so much bigger than what the iphone 15 is already using. Those sensors aren't actually one inch in size. There's a whole history as to why it's called a one inch sensor, but don't confuse that with that being the actual sensor size. That's why manufacturers like Sony say type 1.0. The iphone 15 series sensor size is 1/1.28". Not significantly smaller than than the other devices you mentioned.
@@magic2253711 I will say also though that apple log in prores still has quite a bit of sharpening and especially denoising built in, I prefer a much less processed look out of the camera and really like dng as a result as i think almost any kind of in camera noise reduction a lot of time produces reduced color information and image detail fidelity, but the filmic pro app on android was allowing people to completely disable any sharpening and even denoising a few years before the iphone 15 with 10bit h265 in newer androids being an option with filmic log
You should try pixels too
Can we get any raw video footage from your test
Не стоит учитывать последние ступени динамического диапазона, так как они потеряются при переводе в rec709. То есть у того же Red One реальный динамический диапазон будет на уровне 11-12 ступеней, так же и у айфона 15. Вообще, использованный в видео метод измерения вызывает множество вопросов. Но в целом тенденция видна :) Было бы интересно сравнить например результаты теста Arri Alexa 35. Спасибо за видео!
Great insight! The Alexa35 and RED Raptor a pretty similar - pushing 18 stops! Thanks for the comment!
@@Strada-Tech Oh, okay! You're welcome!
Really great video! Only thing is it seems you do dynamic range tests similar to how RED does dynamic range tests. Just because it can see detail there doesn't mean it's usable. You have to subtract like 2-3 stops from each of your tests for more accurate results. Thanks for making this video. Very intriguing!
Wished they used Android too with Motioncam pro
Xiaomi 14 pro has 14 bit RAW CinemaDNG video.
great content
Boom. Final Cut Pro Camera.
"It's truly inspiring to witness the eagerness to delve into the world of exploration and discovery, particularly when it comes to uncovering the capabilities of a pocket cinema camera and its potential for empowering low-budget crews."
counting bars isn’t evaluating usable dynamic range, especially with compressed codecs - otherwise the best camera on the planet would be a c70 (see cineD)
Imagine apple’s tech in a full cinema camera with a big sensor and proper lens mount. Would be magical
Very informative video! A couple corrections in the script: "the first phone to have log encoding" (*iPhone) and "... a solid 14" (*slight 14, lol)
The iPhone 15 footage needed so much planning, a professional gaffer to utilize methods to add texture and have the “cinematic” look. Industry rate for a gaffer on set is $500 a day which is almost half the cost of the iPhone itself.and this does not include the cost of all the lights, modifiers, ect needed for the shoot to go down for just one single day on set.
With all the the pro staff, lighting and grip equipment used on this video you could even make a cheap used $89 Canon T2i camera on eBay achieve true theatrical results. That’s barely 10% the cost of the the iPhone 15
A RED camera by itself does not need help from hired gaffers, professional lighting techniques, aerosol products for haze, and ect to achieve a “cinematic” look. It can do it even in broad daylight 12pm outside in the bright sun at 5600 Kelvin.
You make an accurate point about the team and cost required to make an image look great. But I'm afraid I strongly disagree with you on the camera not needing help from hired gaffers. Professionals take as much time and care to make a cinema camera look good as a phone. Even thought cameras continue to improve across the spectrum, talent is required to make top, middle, and low-end cameras look their best. I don't expect that to change anytime soon. But that's just my thought. Thanks for the comment!
@@Michael_Cioni SNR in the real world the Iphone 15 pro Max is only rated at 11stops at ISO 55 and 12stops at ISO1200. upon closer inspection, it is not the raw sensor capturing 11 or 12 stops... i will explain.
November 29th, 2023 CineD performed an unbiased comprehensive dynamic range test on the iPhone 15 Pro via IMATEST.. it is stated that initially the high values of the dynamic range on the Iphone 15 Pro is actually achieved with a really high internal noise reduction baked into the video file. In reality the Iphone 5 Pro without the noise reduction has a latitude results of 5 full stops. Some consumer based Micro four thirds cameras like GH6 have 7 full stops of exposure latitude, The APS-C FUJIFILM X-H2S or the Sony A6700 achieves 8 full stops without the need of noise reduction yet. And the ARRI Alexa 35 has 12 full stops before you even start any noise reduction on the file. An old Red Dragon sensor althought is rated at 16.5 can achieve 13 stops before noise reduction needed.
post production wise, a 2014 red dragon has 16-bit colors to play with and the ihpne 15 pro only has 10-bit using pro res raw. you cannot push an iphone 15 pro's image to the limits of what you can push cameras from a decade ago.
with this being said, the 2023 Iphone 15 Pro is catching up but not surpassing 2014 technology just yet.
I was shooting on hybrid log on androids way before the iphone 15 had log though?
I believe you're right. However I think what is new at this point is the combination of log + intra-frame compression + 10bit + 4K. So I believe other iterations were either not 10bit or were limited to interframe (GOP) compression. But I'll need to do more testing to confirm. If you know of a phone that shoots 10bit log using an intraframe codec, I'd love to know! Thanks for the comment!
@@Michael_Cioni oh no I definitely agree that the iPhone's implementation is definitely the most advanced, I just took issue when you said that it's the first phone with log shooting. I definitely wish more Android phones would have a mode where there's reduced sharpening like the iPhone because that's the main thing that makes phone footage look low quality. Idk why people like the sharpening tbh.
If I get a 15 pro, Strada will be single handedly responsible for it!
please please review SONY XPERIA 1 V
If you are using the DoF adapter already, why not go wilder, and use large format camera back as the adapter? the DoF will be crazy!!!
Here is the trend: Cheaper products yields bigger companies. Nikon bought Red, Apple has faster development cycles than camera only companies.
tbh with this much control of light any flagship phone camera will give fantastic result with manual controls
do not believe?? try with Xiaomi 14 ultra Vivo x100 pro samsung s24 ultra etc.
Sony released a $1700 smartphone called Sony Xperia Pro-i. I wonder if y'all could do it's comparision the same way
What ISO are you using to gauge dynamic range? I’ve seen that (supposedly) the iPhone has higher dynamic range but also higher noise above 600 ISO, and has very low noise, but also lower dynamic range below 600! Claudio Miranda suggests that he only will shoot 500 or below on iPhone, but I’m wondering if that limits dynamic range too much!
Great question. I ended up cycling through a few to see what different results I would get. I settled on 800 which had the best retention. But I'm unaware of an official "native" ISO for iPhone. So I just did trial and error.
WAH
Why does he looks and sounds like the big bang theory guy?
Try testing S24 ultra
And again, if you hire a crew, you light, you add (cumbersome) accessories, you use actual cinema lenses... you can get good looking imsges on a "sub 2000" dollars camera. If you don't see the problem here...
Also, you messed up every step in evaluating DR.
The highest quality movie theater projectors have 12 stops. Just sayin
Okay here is my argument. I acknowledge that the people working here are way more smarter than I am and are way more skilled than i am. And nothing I’m about to say takes away from the purpose/explanation of the video. But the reason why the iPhone can produce such good images is not because of the dynamic range being the “single” factor that was chosen to drive through the explanation. It’s because of the color/image processing or lack there of.
I’d argue that if you didn’t shoot in the log profile which strips the image processing that apple typical applies, it would not look nearly as good. And I know that you know this already.
The issue with iPhone footage as well as anyone other smartphone camera footage has always been the way the images are processed. Not because of its dynamic range. Cause we can make cameras with lower dynamic range still look incredible. I shoot with my Lumix g9 and it arguably has worse dynamic range than the iPhone but it’s a lot easier for me to make it pretty cause of the color information and image processing that I can pull and push in post. But when the image looks heavily processed and overly sharpened that’s always the dead giveaway that it came from a smartphone.
So once other manufacturers start giving us log encoding that strips away the processing. Smartphones start becoming viable cinema camera conversations overall. And as a matter of fact, they already are. With apps like motion cam pro that allow us to shoot raw dng on android. I was so stunned that my Samsung could look just like my hybrid mirrorless. And anytime I want the best image possible I shoot raw off my Samsung and not with the stock camera app.
I think that should have been the focal point to drive home the video not the dynamic range
Fair point. I think I agreed with you when I started this. And when I qualified the DR as the single factor, that's my professional opinion. I've been part of A LOT of post production color correction workflows (check my IMDB) and when I factored in all my experience with more than 150 feature films and then started pushing iPhone, I felt the results I was getting were good because of several reasons, but it's my professional experience that DR is the the main one sitting at the top and allowing the rest to fall into place. An example would be if you had a 21 stop camera and captured it in 8bits per pixel in a 25 megabit container, the file would still look absolutely fantastic - even though the bitrate and bit depth suffered, the DR would give it much more mailability than, say, a 10 stop camera that's 16bit at 336 megabits. But it's a fun argument and testing helps raise these great questions. Thanks for the comment!
@@Michael_Cioni yea no worries.. We certainly have to test that hypothesis you put forward. I still disagree atm, especially when I consider cinmea cameras years ago with the 10 stops that are still way nicer/better images than the iPhone shooting 14 stops with the native camera app. Just can't phantom how that's possible.
I know they're many factors like your pointed out but the dynamic range in my mind is not at the top of that list... It most likely is a number 2 or 3 for me. Especially when the dynamic range of a camera can be controlled. Unless you're shooting in uncontrolled scenarios, you can literally fit your lighting to sit within the bucket of a camera's dynamic range.
At that point everything else about how the camera handles color, bit rate, etc. Become increasingly more important.
Healthy discussion though.
How about using a s24 ultra to see who's beter.
Incredible effort but it’s pretty pointless if you are in a controlled set with lighting and cinema lenses you can make anything looking decent. Bring the iPhone out in the real world point it at the sun, shoot at magic hour, shoot a blue hour. That would be a real test. Your effort is appreciated but missing the point.
Your idea of "the point" seems entirely subjective. Their point is to test the iphone camera capability in on set controlled environments. Plenty of other natural light outdoor tests to see on TH-cam.
which camera did you guys measure? what about the other two cameras that have different focal lengths? This is not intended to take away from the impressive result, but rather to give people information on how they can get it themselves
This was all the 24mm lens. That's the best one from Apple in my testing. There are other episodes on this TH-cam channel that help explain more about that.
the color depth in this video is so bad, even at 2160p there is color banding
Resolution has nothing to do with color depth or posterization. This video is mastered in 10bit and the output is a 10bit HEVC. However as soon as it’s uploaded to TH-cam they re-compress it and we have no control.
The banding bothers me too, but it doesn’t exist in any of my hero files. I’d you have insight as to how to circumvent TH-cam compression, we’d appreciate any help! Thanks for the comment :)
Makes me wonder if someday Apple might try to purchase Blackmagic . . .
Your voice doesn’t seem in sync
Well iphone footage looks good on phone screens. As soon you watch it on a big color corrected ditor screens it looks like shit. After all its just a phone with a tiny sensor. It does not make sense to use a phone if you can get really good real cameras for the same price.
I thought the same thing. I grisly directed these and if you check out Episode 23 you can see the whole analysis. My team and I finish these episodes in 4K 10bit and I personally QC them on an 85” LG C2. To my surprise it really looks phenomenal on my TV.
But for that to happen this was with a DOF 35 adaptor. Check out episode 23 and let us know what you think! (Even with TH-cam compression it looks pretty good!)
Thanks for the comment.
Is the audio delayed just for me?
Yes, very slightly for me too
Premium content folks. 🤌✨💯