DJI RONIN - independent review and in-depth test

[Peter Majtan]

Why is that, have I missed something?

Over at the main GB thread...:

As posted above..Today was supposed to be the last day to get full payment in for the AMP DJI Ronin Group Buy but
just spoke with Patrick at AMP and because neither of the 2 tests going have posted any footage or results yet..
he has graciously agreed to extend the AMP group buy deadline for another week till July 7.

So July 7 will be last day to get in the full payment for AMP DJI Ronin Group Buy.

Here are links to the 2 independent tests...

One of these of course being mine, the other is Ketch.
I am editing the first videos as we speak and should be able to post in few hours...

Note that the above only applies to those who went with AMP. Other dealers might have their own deadlines, rules, etc...

:sifone: Peter

 

[Demetri Zuev]

Over at the main GB thread...:



One of these of course being mine, the other is Ketch.
I am editing the first videos as we speak and should be able to post in few hours...

Note that the above only applies to those who went with AMP. Other dealers might have their own deadlines, rules, etc...

:sifone: Peter

Holy shit, Peter, I should hurry up then) Thanks, missed it

 

[Note Suwanchote]

Wait is AMP extending the deadline until the 7th?

 

[Steve Das]

Wait is AMP extending the deadline until the 7th?

yes...

But this was applicable only for people who had already signed up and had the email from AMP.

Last Monday...was supposed to be the last day to get full payment in for the AMP DJI Ronin Group Buy but
spoke with Patrick at AMP and because neither of the 2 tests going had posted any footage or results yet..
he graciously agreed to extend the AMP group buy deadline for another week till July 7.

So July 7 will be last day to get in the full payment for AMP DJI Ronin Group Buy.

Here are links to the 2 independent tests...

Hi folks! I know everyone has been frustrated with the lack of proper independent review and in-depth testing of the new DJI RONIN 3-axis gimbal...
I am pleased to say that the wait is over! I have just received the RONIN and I am ready to "dive in" into the testing and review...

I am cooperating with a local rental house in order to put the Ronin through its limits.
We intend to try all sorts of configurations - from the lightest, most compact Scarlet with affordable still lenses and accessories - all the way up to fully kitted CF Dragon Epic with proper cine glass, matte-box, filters, FIZ and wireless HD...

Hi every one...
As one which has always Tested and put specialty Gear to the test, I checked and used any Image Stabilization piece of equipment I could get my hands on, as I get motion seek easily when viewing Non-stabilized imagery or imagery that for what ever Story Telling decision has too much shaking going on...

 

[Peter Majtan]

Ok fellas, here are some first bites...

Few things to keep in mind while you watch:

- this is BETA unit still on the first firmware (that has been already updated twice since I have shot this)
- this is my first shoot with the unit - I literarily had no practice at all - this first footage is that practice...
- this is also shot without me adjusting any of the functions / features in the user-friendly smart-phone app
- 90% of the shots have no stabilisation what so ever applied in post and the few that do was focused on removing the bounce of the bungee cords
- only exception to the above are the two 105 mm shots near the opening which needed a little help - but hey, its 105 mm and given all the above - just look at what you can accomplish...
- no tricks such as high-speed or quick cuts - ALL shot at 24P and I have kept the cuts as long as possible to keep the video interesting
- the second to last shot with Morgan Fisher has been slowed down to fill the edit gap to match the music - but it was still shot at 24P...
- used two lenses - the EF 17-40mm f/4.0L and FD 35-105mm f/3.5 - all respective focal lengths are marked in bottom left corner along with the mode of operating the Ronin...
- I didn't have my Paralinx gear with me at that time, nor a thin LCD cable - so was forced to use the 5" TOUCH on camera folded forward - half the time I can't see what I am shooting (let alone being able to check focus)

Is it perfect? Nope...
The purpose of these beta tests is to use the gear in real-world scenarios and provide feedback to the manufacturer to fine-tune the product. This is not unlike what RED is doing...
Never the less - especially given all the above - I was amazed with the results and would have been proud to have shot any of these on a steadicam...
Bottom line - if the product was "only" as good as I have received it - I would have bought it without a hesitation...

There are two concurrent routes to improve the RONIN:

A. Improve myself. 95% of the bumps in the footage is my fault for not adjusting properly the follow-mode settings... The most important thing to do is to practice with the RONIN to fully understand its capabilities and to take advantage of the various controls, settings and assists as I learn to understand them and how to use them (which I haven't done for this first test). Just like any professional equipment you need to master it. Ronin is no exception. It will only be as good as its operator...

B. Improve and fine-tune the control SW. As I have said - DJI has already done two new releases since I gave shot this video exactly one week ago. Just like with RED - things will get a lot better before they ship the units and will continue to improve and add features as the feedback comes back from end users...

Hardware-wise the unit is perfect. They are working on an optional accessories to allow fitting of longer lenses and cameras, but for shooting RED Epic / Scarlet (and all the other similar shaped or smaller cameras) with proper cine glass and accessories - this unit is ready for showtime.
Just don't try to mount stuff you wouldn't mount to a steadicam (like Optimos LOL)...
Again - this was me getting to know the unit. I will be shooting an actual demo next week in proper cine configs with MB, FIX, Paralinx, etc...
But I know that people are desperate for some demo footage - and this is more then good enough for that...

Original music composed and performed by our own Tiko Lasola - thanks Tiko!!!

https://soundcloud.com/colasola

Also big thanks to Andrew Thomas - producer / creative director friend of mine - who not only allowed me to test the Ronin on his live production, but he also helped me the next day and drove the car... Many thanks Andy!


Enjoy!

https://www.youtube.com/watch?v=mJBqyjtLpdQ

:sifone: Peter

 

[Zeb B]

Nice work. Love the Deer park!

Did I miss it? What camera etc was used?

 

[Peter Majtan]

Nice work. Love the Deer park!

Did I miss it? What camera etc was used?

Of course my Scarlet as pictured on page 8 of this thread with the preview snaps from this first test. This is REDUSER Zeb... :cheers2:
Canon EF 17-40mm f/4.0L and FD 35-105mm f/3.5 with screw-on variable ND...
All footage 24P @ 800 ISO.
Sunrise in Japan is 4:45 AM at this time of the year - so we started shooting at 4:30 AM...
Speaking of which - even thou the battery is rated for 4 hrs - having started the RONIN at 4:15AM (setting it up and balancing prior departure at 4:30 AM) by the time we finished at around 1PM there was still some power left - that's more then 6 hrs of continuos operation!

:sifone: Peter

PS: That ain't a deer park. These are wild animals in general public areas. Nara is famous for this (just look it up online) and the residents (and tourists) have developed a mutual harmony and coexistence with these animals. In my country if I was this close to a deer - I would be getting it ready for lunch... ;o)

 

[Joel Arvidsson]

Thanks for keeping the cuts long. Any chance to se it in vimeo instead?

 

[Peter Majtan]

Thanks for keeping the cuts long. Any chance to se it in vimeo instead?

Vimeo won't allow me to upload anything over 500MB on a standard account and even the 720P version is 1.5x that...
But here is a direct download links from my server - these are also pre YouTube re-compress...

(right-click and download linked files):

1080P: http://www.derylgroup.com/downloads/DJI/RONIN/FOOTAGE_VIDEO/1stTEST_notes_1080p.mov (1.6 GB)
720P: http://www.derylgroup.com/downloads/DJI/RONIN/FOOTAGE_VIDEO/1stTEST_notes_720p.mov (0.8 GB)

I hope that helps...

:sifone: Peter

 

[Bryan Koss]

That was great and very helpful, Peter. Beautiful stuff. The lower left descriptions were great.

 

[Peter Majtan]

That was great and very helpful, Peter. Beautiful stuff. The lower left descriptions were great.

Thanks for the kind words Bryan! The actual final review video will have even more info and also an accompanying footage from GoPro to show the difference and actual shooting conditions...

:sifone: Peter

 

[Joel Arvidsson]

Thanks it was exactly the download option I wanted from vimeo so this is perfect. Thanks!

Vimeo won't allow me to upload anything over 500MB on a standard account and even the 720P version is 1.5x that...
But here is a direct download links from my server - these are also pre YouTube re-compress...

(right-click and download linked files):



1080P: http://www.derylgroup.com/downloads/DJI/RONIN/FOOTAGE_VIDEO/1stTEST_notes_1080p.mov (1.6 GB)
720P: http://www.derylgroup.com/downloads/DJI/RONIN/FOOTAGE_VIDEO/1stTEST_notes_720p.mov (0.8 GB)

I hope that helps...

:sifone: Peter

 

[Demetri Zuev]

Thanks for the video Peter, those deers are super cute

After watching your video and some other tests I can see that the single biggest thing you have to fight with Ronin (or any other gimbal I suppose) is vertical movement. While out of the car shots are almost perfect I can actually see almost every step you take mirrored in this vertical movement of a camera when you operate it on ground. I suppose that you cannot completely fight it just using your arms, so I'm trying to think of some solution that will allow clean shots without any vertical movement on camera's part. EasyRig perhaps?

P.S.: Thanks to you I managed to order my Ronin in time with a GB discount, thanks very much for that, man!

 

[Scott Pultz]

Thanks for the video Peter, those deers are super cute

After watching your video and some other tests I can see that the single biggest thing you have to fight with Ronin (or any other gimbal I suppose) is vertical movement. While out of the car shots are almost perfect I can actually see almost every step you take mirrored in this vertical movement of a camera when you operate it on ground. I suppose that you cannot completely fight it just using your arms, so I'm trying to think of some solution that will allow clean shots without any vertical movement on camera's part. EasyRig perhaps?

P.S.: Thanks to you I managed to order my Ronin in time with a GB discount, thanks very much for that, man!

There are other gimbals that allow mounting on a steadicam arm. Not sure about the Ronin.

 

[Demetri Zuev]

There are other gimbals that allow mounting on a steadicam arm. Not sure about the Ronin.

Scott, the whole point of Ronin for me is to eliminate the need for an expensive Steadicam op on set, when I don't have a budget for one.

I know that these are two different technologies and they both shine in certain situations more suitable for them, but I've seen some MoVi shorts and footage and to be honest I couldn't tell the difference, for my eye it was like a good steadicam shot footage. One of the most impressive examples is this — https://vimeo.com/81111195

 

[constantine Tirintzis]

hello , here is some hints about hand held and other adaptations versus vertical footstep stabilization.

some facts

1) the slower the walking move , the harder to isolate the vertical walking step errors, no matter what steadicam arm you have and its more or less learning a walking technique
refer to steadicam workshops. However that applies to very slow moves and any good arm should also isolate them in a good degree in conjunction with factor 2. Speed and accel, decel are one of the most important factors that will effect the way your steadicam arm reacts on the walking vertical vibrations.

2) the weight of the rig and camera is very important for No1 the heavier the rig the better the isolation , the lighter the rig the harder it gets for the arms to respond in slow moves and that together with the quality of the arm and gravity is factor No2

3) your natural arms are also a good stabilizer for vertical walking errors, and because both factors 1 and 2 apply to this as well, sometimes you wll discover that hand holding with 1 arm isolates better that hand holding with both. That is because your rig is probably very lightweight for the strength of both your hands that can easily handle the weght, than one hand that has half the strength . Every error your 2 arms can pass is because your muscles are overpowered to pass any accidental reaction, 1 arm has considerable less and with gravity its just cannot pass too much of those errors other that try hold the rig in position. Eventually its like doing steadicam with an arm that is tuned to hold 70 lbs while your sled is 16 lbs, it will never work.Problem is of course that one arm means low mode as you cant do eye lever shots but thats the catch. Most low mode shoots are better in isolation than high mode and that is the reason why.

4) if you are to do a very short move around a subject and very slow try doing it just by using your arm reach, dont walki, if the move is for example a short dolly like take around a subject and short perform it swinging your around hand holding the rig, and that is the best for slow surrounding short moves around a subject.

5) steadicam arms and gimbals in direct mode no matter what gimbal you use of the known cases, will give you some more vibrations that you will have to deal with. Direct adaptation on the arm post that is. The publicly stated 4 axis ( which is not really an extra axis) will not compensate for that , it will only give you the easiness of using the single follow mode and specifically with tilt axis, as the direct adaptation of the ronin as is doesnt apply, and you wont be able to do other than boom the arm up and down to compensate. The reason of all the problems that arise is the fact that steadicam arms dont stabilise all but one axis, and the steadicam gimbal stabilise all but the not vertical .
axis. the results from direct connection and the way the gimbal will be placed , over or under , together with the hard connection will produce those jerks.Depending on the rigity of your gimbal they will be many or less,but they will be there.

Conclusion

if you are into hand hold , and since you cant do much about the total load, learn how to walk to minimise the vertical errors , do the moon walk if you have to lol ( that is a joke)
If you are to place it on steadicam, your best adaptation is to place it on the sled and not directly mounted.
if you dont want to use a sled, then use a gore link .

if you look however for good support of the weight for longer takes, high to low mode and vertical walking vibrations elimination in a very good degree, use the easy rig and the upcoming Serene arm that will give solution to your high mode (more height with the easy rig) , eliminate vertical errors from walking( very slow moves you still will need to learn walking technique just like any steadicam arm) and also be able to go from high to low mode on the fly.we are also designing a nice tool that added will make tilting in single follow mode and easy rig much easier.

sorry for the long post ...thats the minimum I can write to describe your cases. its much more to be consider but...

Kind regards

 

[constantine Tirintzis]

Peter has done a good job , dont forget that some of the shots that the footsteps as more visible, are a longer lens shots, and not open wide and its natural to show up the footsteps more. Also its not such an easy terrain to walk in some cases.. It also happens to the steadicam operators and the more experience with longer lenses and better equipment the better the results.

Edit - very long shots, one man can only start getting tired and then everything starts to show up more, walking , your arms try to fight to keep the rig up etc. Ths is a tool for good planed short takes if one man is to perform , not long lasting shots. Peter has done a considerable good job here..

well done Peter.

 

[Peter Majtan]

Demetri - you are little contradicting yourself when you are aware of human motion strangulating into gimbal shots and yet you think gimbals could replace steadicam. Not at all...

Here is the deal...

YES - both Steadicams and gimbals are used for very similar shots (when we need to translate the camera freely in three-dimensional space) - so, when well executed, the final shots might end up looking almost indistinguishable...
But you need to realise that there are 6 axis of stabilisation to consider when moving camera in 3D space - 3 axis of translation (moving camera in XYZ) and 3 axis of rotation (PTR, or pan, tilt and roll).
Steadicams and Gimbals do the exact opposite - yet complimentary action.

Steadicams stabilise the translation, while the rotation is under direct control of the operator...
Gimbals stabilise the rotation, while the translation is under direct control of the operator...

So - in the ideal world - you will want both for an ultimate full 6-axis stabilisation (and let us for the time bing ignore the gimmicks of N-th axis of some gimbals, where in reality they are simply doubling one and the same axis)...
It has been demonstrated that you can achieve amazingly stabilised shots when you mount 3-axis gimbal on a Steadicam arm...
But you can achieve the same shots just with a Steadicam if you have an experienced operator who can stabilise the rotation...
And vice-versa you can get the same shots just with a gimbal if you have an experienced operator who can stabilise the translation...
Bottom line - you need to stabilise both (2x 3-axis)...

You have spotted this in my test video and this is simply because I am far from being an experienced gimbal operator (there are only very few in the world as this is reasonably new technology).
This was my first time with the gimbal even just getting to know it (let alone start properly testing). With every extra day I will have operating the Ronin I will improve and eventually reach a level where you won't be able to see those issues...
I know as I have been on the other side of the equation - I am an experienced Steadicam operator...
The learning / mastering curve on a Steadicam was rather long and it took me about 2 years before I was confident I can achieve 90% of shots demanded by my clients. After 7 years flying various Steadicam rigs I am still improving... ;o)
It remains to be seen how long is the learning / mastering curve on gimbals...

Both Steadicams and gimbals can compensate for the "other" 3 axis...
On a Steadicams for years we have been adding brute gyros to help stabilise the rotation and now you can also use assists like the genius Helixes' 1-axis solution...
Gimbals are still new and we are yet to see some successful and proven pneumatic / hydraulic systems to assist with the stabilisation of the translation - like the Serene arm on Easy Rig.
As of now - using supports like Easy Rig will only exaggerate the walking motion - these rigs are not meant for walking (people who think otherwise are kidding themselves).

So to answer Scott's question - yes, you can mount Ronin on a Steadicam and there are several ways of doing so.
You can mount it using a custom bracket directly to the arm (no need for a sled).
You can also mount it to a sled - both as a "3-axis head" and as a kind of a "mini-crane" extension.
And I am sure there are other ways yet to be discovered / imagined...

I am myself working on the custom bracket that will allow the Ronin to be mounted directly to a Steadicam arm - more on this later...

:sifone: Peter

 

[Demetri Zuev]

hello , here is some hints about hand held and other adaptations versus vertical footstep stabilization.

Constantine, thanks for your very valuable input. I'm very interested in Serene, though there's one thing that I don't quite get — each easyrig spring has it's own payload limit, does Serene increase that limit?

 

[constantine Tirintzis]

Constantine, thanks for your very valuable input. I'm very interested in Serene, though there's one thing that I don't quite get — each easyrig spring has it's own payload limit, does Serene increase that limit?

The Serene is fully adjustable for loads from 5-6kg to 20kg. If you need the total range you will need to change springs. generally you tune th Serene to the easyrig payload, and you would have to calculate your easyrig that is inside your gimbal and camera payload. The serene is fully tunable to what ever you like inside those loads. Lets not take over this thread , if you want details please contact us at info@flowcine.com. Thanks.