FS5: Comparison of Internal Recording on FS5 vs. RAW output to Odyssey 7Q+

Toby

I reckon the uncropped frame looks 'ok'

I guess you need expectation management? 50fps, 10 bit Slog, that is a lot to cram in to 50mbs, it is the same as 25mbs 25p which is less than a 5dmk2 data rate, which was only trying to store 4.2.0 colour

It would seem 'obvious' that such a demanding spec 50/10/Slog is going to run into compression artifacts?

Ive use the Fs5 once and it was pretty odd to be honest - with compression artifacts - at 25p

Ive not tried 'offboard' recording as the joy of the FS5 is (to me) the tiny form factor

I dont think your camera is broken

S
 
Hey Toby,
i agree with everything Morgan said.
I don't think an external recorder would have that level of compression artifact. It also probably would have looked better if you'd shot in UHD and downscaled.
Thanks for sharing. I hope this shot didn't ruin your project.
 
Back to the subject of RAW. I suspected that the test that is the subject of this thread was a bit of a softball for the internal codec because most of the clips had shallow depth of field. Large out of focus areas of the frame are easy to compress because there is little difference between adjacent pixels.
I bet that if the test were repeated with greater depth of field, the reaction would be different. By presenting the test compressed on the web even more of the differences are masked.
Toby: was the issue you had on several frames, or was it a frame here and there in an otherwise clean clip?
 
SamEd said:
Back to the subject of RAW. I suspected that the test that is the subject of this thread was a bit of a softball for the internal codec because most of the clips had shallow depth of field.
Click to expand...

I'm just going to ignore what you might be implying there -- that I have somehow purposely chosen subjects that would favor the internal codec. That is not the case. This is the style of footage I shoot on a daily basis and if I wanted the full frame in focus I wouldn't even be using these cameras. But anyway, just to prove your theory is false anyway, may I direct you to the other round of testing that contains very little shallow DoF. The internal still looks just as good or better than the external on nearly every shot. http://www.dvxuser.com/V6/showthread.php?349769-FS5-Internal-Recording-vs-ROdyssey-7Q-PART-2
 
I know this is much information in not much bitrate. But in the FS5 footage I watched before I bought it I didn’t see something like this. Hence the doubts if my camera eventually is broken.

Does anybody have similar artifacts in his/her FS5 footage?

@SamEd: It’s in the whole clip, not just in a few frames.
 
Tobias Brummer said:
I know this is much information in not much bitrate. But in the FS5 footage I watched before I bought it I didn’t see something like this. Hence the doubts if my camera eventually is broken.

Does anybody have similar artifacts in his/her FS5 footage?

@SamEd: It’s in the whole clip, not just in a few frames.
Click to expand...

Did you mean to post a link to the footage you are talking about? And, are you taking about the tearing issue the FS5 had when it first came out? That was fixed in a firmware update. Make sure you have the latest firmware installed.
 
Chadfish said:
We can't play mxf files outside of the folder structure. Can you export it and upload it somewhere yourself so we can see what you're saying without having to work for it? We can't help if we can't see. Thanks. The FS5 image looks pretty nice IMO.
Click to expand...

I rewrapped the original file, I hope this works: https://drive.google.com/file/d/0BxSh6YKgd8ZAQjMxSFZVdkNpeXc/view?usp=sharing

I also transcoded the file to DNxHD in case the rewrapped one doesn’t work. But it’s still uploading.

Thanks in advance for taking the time to have a look at the clip.
 
Hey Doug,
i didn't mean to imply that you were trying to mislead anybody with the tests. As a matter of fact I think it was very representative of the way the camera is generally used. These tests are very valuable to the community and I'm grateful that you took the time to create them and share.
I'll check out part 2.


Doug Jensen said:
I'm just going to ignore what you might be implying there -- that I have somehow purposely chosen subjects that would favor the internal codec. That is not the case. This is the style of footage I shoot on a daily basis and if I wanted the full frame in focus I wouldn't even be using these cameras. But anyway, just to prove your theory is false anyway, may I direct you to the other round of testing that contains very little shallow DoF. The internal still looks just as good or better than the external on nearly every shot. http://www.dvxuser.com/V6/showthread.php?349769-FS5-Internal-Recording-vs-ROdyssey-7Q-PART-2
Click to expand...
 
SamEd said:
Hey Doug,
i didn't mean to imply that you were trying to mislead anybody with the tests. As a matter of fact I think it was very representative of the way the camera is generally used. These tests are very valuable to the community and I'm grateful that you took the time to create them and share.
I'll check out part 2.
Click to expand...

Sam, I'm sorry for reading that the wrong way. And with that said, I agree with you that shallow DoF does favor the internal codec, I just wanted to be clear that I didn't set out to tip the playing field in it's favor by doing those kinds of shots.
 
Hi Doug, first of all thanks very much for doing your tests. Really interesting vid(s).

I got my FS5 less than a week ago, & have mainly been testing the 240fps function under different lighting conditions, with the aim of shooting music videos when i'm comfortable with it. (previously been using BMCC, BMPCC & GH4)

My question, is does the Odyssey/raw combo make any difference to regular 1080p 240fps quality?

I like the idea of the 4k burst, but its the continuous 1080p S&Q i'll be leaning on.

Also, does the 7q provide a histogram when shooting S&Q? It's something that disappears on the FS5 screen when in this mode.

I would just hire a 7q for the day & test myself, but it means spending £450 on the FS5 upgrade first which could prove to be a huge waste of money :) Cheers for any feedback ! Ru. U.K

p.s these are are couple of test shots I did under natural daylight conditions, but in shade due to trees overhead...i'm *fairly* happy with how these turned out, but if you look at the mud either side of the leaves in the first shot, there is a fair amount of noise...is this something that raw/prores will help with in any way in your opinion?

p.p.s this was my first day using the camera, be gentle.. :)

https://www.youtube.com/watch?v=PT1A8xkjEeY

https://www.youtube.com/watch?v=8MebZz9BlRw
 
Last edited:
Hi Ru,

First of all, really nice shots. The focus is a little off on the running shot, but other than that I think they both look really good for early test shots as you get to know the camera. I love shooting slo-mo and would do it all day if I had the time. In fact, I'll be shooting some new 240 fps RAW stock footage with my F55 this weekend.

Unfortunately, I don't know the answers to your questions about the image quality of shooting 240 fps with the FS5 & 7Q+ because I have not had the time to experiment with that mode. But I have no reason to think it wouldn't look basically the same as the other RAW test shots I've been doing with the FS5 (except in HD instead of 4k), but I am reluctant to say that because I haven't seen it with my own eyes. One thing to consider is that you're probably going to have to add some noise reduction in post to the 7Q+ footage, which I find to be a huge hassle and a pain in the ass to get right. I'm actually very happy with the 240 fps internal mode of the FS5 and know it can produce excellent results in short bursts -- as you have also demonstrated. Drop a LUT on it in post and your're practically done grading. But the advantage of the 7Q+ is that you can shoot continuously without being limited by a burst mode. Personally that has no appleal to me whatsoever because of the huge files sizes and clip lengths that would result from it, but I'm sure that will be a big selling point to some people who want to roll for longer than a few seconds.

I'm not going to say that I won't do some 240 fps test shots and post them, but it is unlikely because of my limited time and lack of personal interest in it.
 
The Odyssey7Q+ monitoring tools such as histogram and waveform are available continuously whether recording or not.

It is NOT necessary to add noise reduction to Odyssey ProRes files if shot properly. The camera is a bit noisy in the shadows, so I generally recommend overexposing by one or two stops and then darkening in post to compensate. The Odyssey includes Exposure Compensation LUTs so that the corrected image can be previewed while shooting. These same LUTs are available for download on the Convergent Design website so that they can be applied in post to get the exact same image that was seen on set. The website also has tutorial videos to explain how to apply these LUTs in Premiere and Resolve quickly and easily.

Shooting at 2K/HD at 240p can be rolled continuously on the Odyssey7Q+. This allows great freedom for shooting instead of being limited to a short burst on the camera. By recording in ProRes you can avoid enormous files.
 
Mitch Gross said:
It is NOT necessary to add noise reduction to Odyssey ProRes files if shot properly.
Click to expand...

I'll have to respectfully disagree with that statement. My experience certainly does not bear that out, but then again, maybe I just don't know how to shoot it properly. :-)
 
Last edited:
I woke up this morning to find that my shoot was cancelled due to weather, so I decided I could spend an hour shooting some 240 fps test footage with the FS5 and 7Q+. Nothing fancy, just some stuff around the kitchen and garage. I think the quality looks very good and almost on par with slow-mo footage from my F55. Grading was done in Resolve and included a healthy dose of noise reduction to clean it up.

 
Nirv said:
You let us down, that's all I'm going to say...
Click to expand...

...am sitting on the naughty step & thinking about what I've done ;)

Thank you all for responses..

Mitch...great to see interaction with the manufacturer..a far cry from dealing with Blackmagic ! Good news re: LUT & Histogram...there seems to be a rumour that internal 240/200 fps on the FS5 uses some kind of pixel-skipping, rather than the 'downsampling' from the 4k image it does in normal speed 1080p...could this be another reason why 4k burst apparently looks better? any thoughts on that?

Doug..practical (LUT/histogram/prores/not-having-to-wait-for-buffer) advantages of using the 7q aside...does it *improve* the internally recorded footage in your opinion (at high frame rates) ?
 
Last edited:
You must log in or register to reply here.
Back
Top