kye

3 hours ago, SRV1981 said:

Oh didn’t think about this line. Will check! 

Or the GX850/GX800/GF9 camera, which is discussed heavily in this thread...

The rules in travel shooting are simple:

Rule 1: Get the shot.

To get the shot you must have your camera with you (so it must be small, must not have been taken by security, and must not attract too much attention from others around you).  You must have it handy and ready with the right lenses (so a zoom is often more practical).  You don't know what you will encounter, so you must be able to shoot quickly.  You don't know what will happen and how you will edit so get a lot of varied coverage.

Rule 2: Get a high quality capture of the shot.

This rule is secondary to the first rule.  If anything about the camera decreases your ability to get the shot (rule 1) then it is hurting your efforts, not helping them.

Rule 3: The magic is in the edit.

Once you've returned from shooting, get to work.  Even a casual review of any travel TV series will make it obvious that the individual shots are not what make the show great, but the editing and storytelling and sound design.

If you want a good travel film, shoot a lot then open up your NLE and get to work.

Here is an award winning episode of a travel TV series - I encourage you to look through it and analyse the footage and the edit, as I have done with many examples.  I think you'll find that almost no shots are beyond a modern point-and-shoot.

(NSFW)

I have done frame-by-frame analysis of this episode, and this is one of the finest examples of editing I have ever seen.  It has a complex and nuanced storytelling and cultural narrative, has great music and sound design, uses more editing "tricks" than those nauseating YT travel videos that used every fancy transition they could find, and does it all in a way that is sensitive to the narrative.  

It makes even the best YT editors look like toddlers playing with crayons.

3 hours ago, SRV1981 said:

 

Resolve has an AI depth map plugin, but the edges aren't good enough to use yet, and neither is the above.  

It will get better, but it's got a way to go.  Flyaways in the hair is the giveaway, and unfortunately it's both very difficult to do effectively, and it's also visually quite obvious.  If you want to use this technique then you'd have to sacrifice the more cinematic lighting and compositions to hide the artefacts.

56 minutes ago, PannySVHS said:

Some underwater macro marvel. Not bad for a little pocket cam. Upload is 1080p, so wandering if it was even filmed in 4K. So here the little Lumix LX10 for some underwater footage. I really like the framing and the cinematography as such. What do you think? @Davide DB 🙂

 

 

It plays in 4K for me, and looks pretty good!

There is some optical distortions, but it looks like the normal ones from filming through glass, so not the cameras fault.

17 hours ago, Emanuel said:

Not the best of the news we would like to hear anyway, here is a fresh first review...

 

That's more like the performance I remember from previous models.

Field curvature is a thing, which he doesn't test in the video, so that might be the cause of some edge softness.  For example here is an article with graphs:
https://www.lensrentals.com/blog/2019/11/practical-use-of-field-curvature-graphs-the-50mm-primes/

It is odd that such a setup would have these issues more than the lens itself, considering that I thought the setup was just the lens projecting onto ground glass that the phone focused on.  Or are there additional optics inside the mechanism?  If so, that's probably the cause.

The focus peaking not showing up is just because the setup is very soft, this is a common problem when using vintage lenses.

6 hours ago, QuickHitRecord said:

Thanks largely to Andrew's original post, many people here already know that Shane Carruth shot 2013's excellent indie feature Upstream Color with a pair of hacked GH2s. At the time, aspiring filmmakers everywhere found inspiration that such a masterful film was created with such an accessible consumer camera. It still stands up today as a beautiful, challenging, thought-provoking film:

I just happened across some behind the scenes photos on Reddit posted almost a decade after the film's release. I've always wondered about the technical particulars of how this film was made and even though what was posted is pretty low-res and deteriorated further with early Instagram filters, it does show some pretty bare-bones camera rigs and a lot of natural light.

As I enter my forties and the dream of creating a meaningful narrative film in my lifetime seems to slip further and further out of reach, this is just the kind of inspirational kick in the pants that I needed. If 8-bit, 4:2:0 AVCHD from a Micro Four Thirds sensor was enough to create an immersive story like this in 2013, then the never-ending pursuit of marginally improved gear and software really is just a giant, capitalism-fueled distraction -- one that I know that I use to put off the very real possibility that when I buckle down to make something deeply personal, I might not have anything to say. But I also know that it will be my deepest regret if I never even try.

If there's one thing I've learned, it's that my own thoughts and ideas aren't particularly novel or unique, so maybe others on this board will connect with this too.

The relationship between a cameras image and the emotional experience of the viewer are connected, but a compelling story will completely overwhelm the image quality.  I know there is no discussion on here about what happens to images once they're captured (and much discussion is relevant to only single images) but I think that the image quality is part of the last 10% of polish on top of a finished piece.

• Love is still love, even if it's in SD
• Heartbreak is still devastating even if shot on Alexa
• Betrayal is still ugly even if using Canon colour science
• Emotions are still colourful even if filmed in B&W

If you want to make a feature then go ahead and make one.  You are probably already familiar with Noam Kroll and the principles of writing the film you can make, starting with what you have access to and working backwards.  I think that probably everyone is capable of making a feature film of significance if they were able to look within and tell a story inspired by their own struggles.  

Until recently all of TV was in worse quality than every FHD camera on the planet, and millions or billions of stories were told.  Don't let colour subsampling and DR distract you from the fact that the human experience is universal, and that what separates a good film from a bad one is how much the audience can relate emotionally to what they see.

18 hours ago, Emanuel said:

Same impression going on here... Looks funny the first footage seen over here is not from a studio setup! LOL Looks like they should have trading names between both models instead : D

 

Curious to check it accuracy too... ; ) Seems this one is what the other one has missed to be, size and convenience ;- )

The first footage I saw was in a studio, but everything looks so good in a studio now that it's not a test that really stands for much.  

I've also seen a couple of videos of it on FPV drones, but those were coloured badly or weren't that useful.  I wasn't sure that people would even use it outside a studio at all actually, but thankfully I was wrong 🙂 

One aspect of this camera I'm particularly interested in is the AF.

I use AF-S with back-button on the GX85, which is very good and very fast, but if this has anything like that level of performance then it will be a very interesting thing to add to my setup.  One thing that makes it super interesting is that the Dual ISO means it will have much better low-light than the single ISO GX85, so would make me quite torn between the two.

30 minutes ago, PannySVHS said:

Image looks like from an p4k test video imo. @kye I find the video ok but pretty plain.

Cage, ssd clamp and further rigging are shown in another video of the same youtuber. With the specific cage and external ssd it already looks looks pretty big, almost as big as the m2k with a 3" monitor rigged to the back, look at around 7min 50s. Knowing the great quality the p4k is capable of, this camera is no slouch in the image quality department, of course. It will be great for broadcasting and streaming for sure. I won't trade it in for my m2k though. 🙂

 

I've been chatting with @mercer about it and we think it's the same sensor as the P4K.  The sensor specs that we've looked at all line up between the two cameras.  

So if you think of it that way, it's the small box version of the P4K that everyone was asking for.  It might be larger than the M2K but it sure is smaller than the P4K!

First outside footage...  looks good to me!

Yossy has weighed in:

https://ymcinema.com/2024/01/11/panasonic-strengthens-its-imaging-business-will-the-cinema-lineup-be-renovated/

Lots of discussion of the EVA1, Varicam, etc in there, as well as rumours of a refreshed cine line with S35 and FF/MF options.

14 minutes ago, Jedi Master said:

I learned FORTRAN in high school. It was the first high-level language I learned.

At university, the first CS class that every CS major had to take was a FORTRAN programming class. I was an EE major, but I took it anyway as I could see that's where the industry was headed. The work was done on a CDC 6400 mainframe using punched cards.

The second CS course, which I also took, used Pascal, which was a big improvement over FORTRAN, although even some of the CS majors just couldn't get their heads around the concept of pointers (they seemed natural to me). 

After that, everything was in C, which is the language that I use to this day when working on microcontroller projects (as a hobby). My professional work mostly revolves around working with hardware description languages, primarily SystemVerilog, and the various tools used for simulation and verification.

I started just after The School of Computing had formed and separated from the School of Engineering.

In first semester they taught us to program in Pascal via the "it's like a recipe" -> pseudocode -> Pascal path.  In second semester they kept going with the theory in Pascal and also taught us C and assembler too, so by 9 months in we were coding double linked-list databases in assembler.  Some kids hadn't used a computer before the course....  welcome to computer science!

After that I have vague memories of Java, Javascript, Perl, Fortran, Visual Basic, etc but the main language was always C.

18 hours ago, Emanuel said:

Interesting comparison... to show how good this tiny device is:

 

Definitely getting better.  

Considering it's on a gimbal and is an enthusiasts camera, I think they could spend a bit more on the next iteration (or release a Pro model) and make next model with a collection of cameras like smartphones now do.  Even if it had a 15mm equivalent, a 24mm equivalent and an 85mm equivalent that would be great.  Or make them the 8K like the Insta360 Ace Pro and you could have 4K recording with the 16mm (and 32mm with crop), then 50mm (and 100mm with crop).

14 hours ago, IronFilm said:

I wish I used to code in COBOL! (or heck, Fortran)

But sadly it's something far worse... FoxPro (both FoxPro 2.6 and Visual FoxPro). 

Well, FoxPro is definitely a much "better" language than COBOL. But I'd say its market value in terms of employability is drastically lower. 

In my Comp Sci degree, we reached a point where we realised that after learning 3 or 4 new programming languages in first year there were no more programming units after that.  Unfortunately we got a shock when we arrived in second year at the start of a unit called something like Numerical and Computational Methods and they told us we'd spend the first 8 weeks learning Fortran, then we'd be using it to program all sorts of horrific mathematics-related things, like converging and diverging algorithms, L-U matrix decomposition, path-finding, etc.  The lecturer had written a small book on Fortran that was very well written and absolutely essential because the lectures weren't enough to learn it and so we had to keep the book on hand to refer to while doing assignments etc.

Apparently in subsequent years they worked out that the lectures weren't enough and that students were teaching themselves from the book, so they shortened the Fortran lectures from 8 weeks to 6, and eventually to 4 weeks.  

After that we arrived at the start of each unit thinking "what they hell are they going to sneak into this unit that wasn't included in the unit outline".  Sadly, that wasn't misplaced paranoia as subsequent units involved more additional programming languages than I can remember, and even open-ended assignments where you had to accomplish some goal in a programming language of your choosing.

After all that, if I had to pick a programming language to work in, Fortran wouldn't be it!

13 hours ago, IronFilm said:

On the topic of 2.5K, I saw a crazy cheap BMCC 2.5K MFT for just US$300 on eBay today but the shipping costs are obscene 😞 

Yeah, familiar with that one.

When I was into hifi I once paid almost $100 for a pair of resistors.  They were custom made high-precision ones made from exotic materials, so were expensive at USD$10 each, but it was the USD$60 shipping fee (which at the time was the minimum cost for whatever courier they used exclusively for shipping their products).

Combined with import duties, we call it the Australia Tax.  It's also one of the reasons I don't tend to sell things I'm done with it.

6 hours ago, PannySVHS said:

Exited. Don't know what to film yet. Filming with F8 to 11 allowed me to use the single point AF without worries on the P2K and not have to worry about the screen for focus control in the bright sun. I got a small vari nd now for my panny 14mm, so i will give F4 a try. Exiting! Maybe some nature. Or 14-42 instead at 5.6. Didn't you suggest that vari fstop lenses from 14 to 42 give same dof as aperture closes down through the zoom range @kye? OIS will eat more battery though:) Or i go with the tevidon 10mm. But would need to fix the aperature with tape as its very loose. No VariND though for this one, meaning f11. Think it only goes till f8. Have to look. Tripod and a 90mm might be another thing of beauty. No Nds for that one as screw thread is damaged big time. Blue hour and freezin my ass off.:)

Anyone remembering that komodo test footage in LA from a female dop? I think our friend Don aka @webrunner5 posted that two years ago.

The variable aperture zoom lenses aren't fixed DoF, but they're certainly close..

...and I think you're talking about Mina Rhodes who did a bunch of Urban Symphony style videos like this one?

1 hour ago, MarkAm said:

Thanks, yeah it seems no matter what picture profile I choose for shooting in camera I'm going to have to do a bit of experimenting to use a CST node.  The N-Log gamma in the input settings for DR are probably directly mapped to the N-Log picture profile in the Z9's and the earlier cameras outputting to an external recorder.  I might try it but I'm guessing the mapping will result in some color shifts somewhere.

I think my only other option to start off with a "what the camera saw" first-step approach that Mostyn and to some degree Kelley use is to use my X-rite color checker in lieu of a CST node and use DR's color chart matching feature.  I'm kind of surprised someone out there hasn't come up with a picture profile that maps directly with one of the default gamma settings in DR to make this process much quicker for folks shooting with pre-Z9 nikons.  (Heck, even finding tutorials videos with an example of a CST node used on Nikon non-N Log footage is hard to find.)  

Anyway, I'm a noob and the math/technical aspects of creating a pp that maps directly with a gamma in DR is way above my head so perhaps it's a bit harder and there may be some proprietary issues as well.  

If you shoot a test shot of the X-Rite then you should be able to just pull it into Resolve and have a bunch of guesses on CST settings and just watch the vector scope and waveform to see which are closest.  Just remember to shoot the chart with high CRI lights and expose it well.  I don't know the exact theoretical way to shoot them, but I shoot my chart by just exposing normally and it comes out fine in the charts with the right CST.  

Even if you only find a combination that is close, it's like I said before "if it looks good, it is good" and unless you're doing scientific work or commercial shoots then the colours don't have to be exact, and if you're doing narrative work then the colours are best when being true to the emotion of the scene, rather than the specific objects that were filmed in the scene.

2 hours ago, PannySVHS said:

It just seems that i don't wanna take a chance on the port nor rig the beauty up. I really enyoyed the restrictions and freedoms of using it barebones as I wrote above. 

I got two 64GB cards plus a 32one, all three working well but none for 50p raw 3:1. Read on this forum about more recent cards working without troubles iirc. Maybe we could do a og p2k or m2k cinema verité parade, you, @kye , @mercer , me and anyone else.

My P2K and M2K are both ready to shoot!

3 hours ago, BTM_Pix said:

At this particular moment, though, I'd be far more concerned about trying to get hold of media to use in the OG Pocket than its HDMI port.

The word in the OG FB groups is Kingston SD cards work, with a number of accounts suggesting the file format used matters (the Mac one is better apparently - I think they mean ExFat) and some people suggest the green ones aren't so good and to get the blue or red ones.  I can't confirm though.

2 hours ago, SRV1981 said:

So no phone specially? Just dislike iPhone log? 
 

and are you staying you’d prefer the image of the RX100 over be iPhone 15 with log ?

I'm also curious to hear from @BTM_Pix on why this doesn't move the needle for him, however I can imagine a number of situations why it wouldn't be a revelation for folks.  For me it's a revolution in phone footage, but it's not a revolution in how I shoot.  Here's an outline of my thoughts:

• The iPhone 15 main camera creates a nice image, but is an 8K sensor with a fixed 3mm F1.78 lens, which with the crop factor of 8.0 is equivalent to a 24mm F14.2 lens.  This lens can employ a 2x crop and still be 4K, but beyond that is upsampling.
• The iPhone 15 other cameras have shown that unless provided enough light (basically they need direct daylight) they exhibit significant noise
• An RX100 has a 9-72mm F2.8-4.5 lens (with a crop factor of 2.667 is 24-192mm), which means there is no cropping below 4k when zooming beyond 48mm equivalent, and compared to the iPhones tele camera (5x 120mm f2.8) will keep its 4k resolution when zoomed beyond the 120mm iPhone camera
• Other fixed-zoom cameras like the LX100 will be similar to the RX100
• For me, when comparing the iPhone to the GX85, there is no competition..
  • The iPhone 15 has real log, is small, has PDAF and can instantly change between various focal lengths
    however..
  • The GX85 can be fit with F0.95 primes for low-light
  • ..has a tilt screen for overhead and low-angle shots
  • ..can use super-zooms like the 14-140mm
  • ..can be used with extreme telephotos like my Tamron 400mm F5.6 prime
  • etc

However, let's do something radical and assume we want to make a film, and assess things from the perspective of that end goal.

The iPhone allows you to shoot very fast, and provides good image quality at one or two focal lengths, and good image quality in a couple of others in direct daylight conditions.  

The main camera has zero background defocus when focusing further than 7ft away.   The iPhone 5x tele is equivalent to a 120mm F22 lens with a DoF of 1.76ft at a 10ft focus distance, but the RX100 has a DoF of 0.38ft at 10ft focus distance when it's at 192mm full zoom, and the iPhone 5x camera will be cropped to 2.4K from its 4K sensor at that focal length.

The iPhone lacks the ability to optically zoom to the right composition without cropping past 4K to lower resolutions.

The iPhone is a great camera if you want small size, to work fast, and have a "I get what I get" type of shooting style, but not many films would start with the final product in mind and work backwards to arrive at the conclusion that the iPhone, and all it's limitations, is the perfect tool for the job with the least compromises compared to other options.

My current setup that I will trial for my next holiday is this:

• iPhone 12 Mini
  for very low-key shots and for use when moving fast like going through airports etc where I need to think about other things but want to grab a shot here or there
• GX85 with 12-35mm F2.8 zoom (and iPhone wide camera for ultra-wides)
  for times when a small camera is ok and I need to shoot very quickly to get spontaneous moments
• GX85 with 7.5/2 + 17.5/0.95 + 50/1.2 primes
  for low-light times, such as sightseeing at night, going into caves / mines / etc
• P2K or M2K with 12-35mm F2.8 zoom
  for times when I can work slowly to expose and focus manually / slowly, there is good light, etc

If I bought an iPhone 15 Pro my iPhone shots would be improved, but it's not replacing the GX85 because it can't zoom or operate in low-light, and doesn't come close to the IQ of the P2K or M2K cameras which are closer to the IQ of an Alexa than the iPhone.

A bit of googling found this - I would imagine this answers your questions?

2 hours ago, horshack said:

I have a project where I need to precisely match the timing of video to audio. In the course of setting that up I noticed what appeared to be a fixed audio lag/skew in my S5 video. It's worst as 60p and occurs in both 1080 and 4K. I created a video to demonstrate what I'm seeing:

Setup
Arduino UNO R3 board running code I wrote that turns on the LED at the same time I turn on a buzzer. The LED+sound stays on for 100ms, then is turned off for 100ms. Repeats continuously. This creates 6 frames of the LED on+buzzer, then 6 frames of the LED off with no sound (for 60p recording, which is 16.66ms/frame, so 6 frames in ~100ms).

S5 is configured for 4K 60p 1/250. That fast shutter speed was selected to always catch the LED turning on within a single frame. There is a lav mic into the S5 that is positioned right next to the buzzer on the Arduino. 

Observed Behavior
I expect the video to show the  LED turning on at the same time the audio waveform shows sound from the buzzer, or within 1 frame of each other, to account for any timing skew between the start of an S5 frame and the start of the LED+buzzer. Instead the S5 shows 2-3 frames of the LED on before the waveform shows sound from the buzzer, indicating that the audio is lagged/skewed by 30ms to 50ms. The lag is slightly less if I record the same setup over HDMI instead of internally. I also compare the S5 to the Sony ZV-1 recording the same setup, which shows the expected behavior of <= 1 frame lag between LED and audio.

 

 

I see a delay in the audio from my GX85, and where it matters I just re-align them manually.

If it matters to you, just include an event at the beginning of each clip that allows you to align it in post.  This is why the clapperboard is used in making movies - it allows the picture and sound to be aligned.

If it matters then you should also ensure that the timing is stable over time, so there are no drifts over long takes.

1 hour ago, Emanuel said:

You need to click that link up there to find my "big arms" if any ; ) to hold that thing... : D Now go back there and take a look on the " t-e-r-r-i-b-l-e " stuff they are able to extract from there... Or better call it terrific?! Well, I end confused! LMAO ; ) But you need to watch it to begin with, as usual, buddy ;- ) The toys these guys invent nowadays... And so lovely to think a hard case is still a hard case, isn't it? LOL : )

That said, you know what ideas that Beastgrip (BTW as disclaimer, I am customer of them) outfit gives to me?

The 37mm thread to couple to the AcePro before or as alternative to a Back-Bone release and extra glass use from either lens adapters, filters, suchlike :- )

 

A mobile phone is still a mobile phone. Maybe iPhones are less phones than the leftover but personally I've found them to provide a good fun for a ride but when overheat they are just a brick... In any case, when not, you can obviously forget them as communications device if you're using them for something else other than its primary use.

 

Because a camera is not a phone. There are many times you won't have any other device with you and the best camera is the one you hold with you in your hands when needed.

 

It doesn't exist that, only a camera : ) You buy a system, not a camera. Full of many lenses and so on.

A camera looks a camera.

 

I give you an example, no, two.

June 29, 2023. Public event. People don't like to be shot. Law is at shooter's disposal. Public place in that jurisdiction means consentment for acquisition. Its usage is something different, in any case no one can impede you to shoot in any way other than when start to threat the camera crew.

January 1, 2024. Identical episode. In the beginning, only because of the capture devices used, people in a public event had no idea they were filmed. So, spontaneous footage with people dancing and so on was possible to gather and gather. Until the moment, people realize where the cameras are. End of fun. End of good footage.

 

Here are two examples where the PRO device is the one capable to still produce footage in the memory card or any other storage location. Get the idea? This means a lot to those involved in the field.

 

The point is the technology is there. These people of Back-Bone, Beastgrip and so one, they are solid players to really help the independent film communities.

Playing the part of the film crews twenty years ago doesn't help at all, just for the sake of the art and this craft, if not anything else :- )

 

Speaking of devil...

 

 

 

Stealth can be one motivator, that's definitely true.  The MagSafe quick-attached ND / filter holders would be suitable for such a thing, if you wanted a 180 shutter and manual experience.  The fact the back of the phone doesn't look like a phone might even help to disguise the filming activity.  

If you're pointing the camera at someone the other strategy is to pose, like you're taking selfies instead of using the rear-cameras to shoot other people.  That's probably the mobile phone equivalent of Wingrand constantly fumbling with his SLR  so people think he's an incompetent old man who doesn't know how to hold a camera, rather than someone taking photos.

I stand corrected on the BeastGrip adapter - the MK3 DOF adapter seems to look just fine, unlike the previous ones which had the optical performance of a lens from the 1800's.  Still, it's not a good option if you also want to be stealthy!

1 hour ago, PannySVHS said:

Without a cage these things are hardly usuable.

Do you mean they're not that usable without having a cage for the Shogun Flame?

Why is that?  Surely it's ok if you just mount it to the rig you have?

I must admit I have an attraction/repulsion reaction to external monitors.  Attraction because of their potential brightness, size, and resolution, but repulsion about the extra size, weight, need for a rig, extra batteries (and additional chargers etc), etc.  I keep having the following:

1. I should use the P2K
2. P2K screen doesn't articulate and isn't visible outdoors
3. I could use an external monitor for it
4. What a PITA - I may as well use the M2K (BMMCC)
5. M2K has rubbish buttons and having a rig is too big / cumbersome
6. If only there was a way to get that image without a rig
7. I should use the P2K.  (Go to step 1, loop......  forever!)

1 hour ago, PannySVHS said:

Guess you won't be able to answer phone calls with that constuction around it and with another construction around it.:)

This was always the problem I had with using a phone on manual settings - you are forced to choose one of the following:

• Buy the latest phone and use it as a phone and camera, and use a rig for the camera use
  Crap option because you now don't have good access to your phone, plus you risk interruptions from calls / notifications or have to use aeroplane mode and miss potentially important things
   
• Buy the latest phone and use it as a phone and camera, with some sort of rig that easily removed for the camera use
  Almost any clip-on/off rig (even just an ND) is normally incompatible with proper cases that protect the phone, either requiring you use their protective case (which is invariably a POS), or you take the proper case on and off (which is hard to do and damages them quickly), or you have to use the phone without much drop/scratch protection
   
• Buy the latest phone as camera and use your old phone as phone
  If I just paid a bunch of money for a phone that is only a camera, why didn't I just buy a proper camera?
• Buy two of the latest phones, one for phone, one for camera
  Same as above.
   
• Buy an older phone as camera
  Same as above, with the possible exception of buying an android phone that can shoot RAW

I also did a deep-dive on the setups that use ground glass and a real lens and the image had huge problems - if any camera looked like that then it would be laughed off the internet.

One potential exception is that newer designs seem to use the MagSafe attachment to hold a filter / ND, which might actually be an acceptable solution for filters, but not likely strong enough for lenses.

14 minutes ago, SRV1981 said:

what does that mean, exactly? 

Our host isn't the biggest fan of that particular person and has requested previously that his content not be shared.