The SONY FS-700 & The Odyssey7 - Let´s hope to see some affordable HQ-Output!

Wait... Am I getting this correctly:You pay all that money for a sony recorder + update + interface and they have crippled the Raw in high framerates? wow..
looks like the train keeps on going downwards..
 
Dan Keaton said:
Dear Friends,

We are wanting to shoot some more footage, at different frame rates, and under different conditions.

We may (or may not) have a solution to what Adam Wilt found.

But, we have discussed this internally, and we hope to have a great solution for 2K up to 60 fps.

This is something that we are going to implement and test, to prove that our theory and implementation is correct, prior to promising this.

There is another possible solution that we will be testing for footage from 60 to 240 fps.

Please rest assured that we will do everything in our power to resolve this issue, if it is in our means to do so, with the very technical tools and processing power that we have built into the Odyssey7Q.

Respectfully,


Dan Keaton
Convergent Design
Click to expand...

thank you dan, you guys have done some great work in the past. god speed to you and your team
 
Ofcourse there will be degradation. That's how the superslowmo works. It starts skipping fields. The more you skip, the faster you go. The sensor is fast enough for 60p at 1080p at 4k but for it to get to 240fps it will need to start skipping fields. That's why theres moire and aliasing at those speeds.

Same thing with every other camera. Red just crops the sensor, no moire and aliasing but it crops the FOV. Slower sensors do it to get to 60p. The C300 starts to alias when it goes to 720p60. etc.

What it shouldn't have is jpeg-artifacts so apparently the image does go through some sort off buffer.
 
maarek said:
Ofcourse there will be degradation. That's how the superslowmo works. It starts skipping fields. The more you skip, the faster you go. The sensor is fast enough for 60p at 1080p at 4k but for it to get to 240fps it will need to start skipping fields. That's why theres moire and aliasing at those speeds.

Same thing with every other camera. Red just crops the sensor, no moire and aliasing but it crops the FOV. Slower sensors do it to get to 60p. The C300 starts to alias when it goes to 720p60. etc.

What it shouldn't have is jpeg-artifacts so apparently the image does go through some sort off buffer.
Click to expand...

Thank you for the knowledge :) good to know!
 
brazzam said:
Also, the degradation only occurs with the Super Slo-Mo. From what I've read, the general 24-60fps 2K recording modes will not have these issues.
Click to expand...

that's not exactly what Adam appears to be saying. he writes,

"I surmised (and Sony confirmed) that the same pixel-binning used for 240fps is used in ALL 2k raw capture: that’s how Sony gets 2K out of a 4K sensor yet maintains the same sensor active area."
definitely disappointing, if true. sounds like something is getting lost in the scaling to 2K. perhaps CDs possible solution will be to take the pristine 4K and scale it in the Odyssey. let's hope for that. while workable for high speed, that loss in resolution will be very discouraging in 24fps 2K. the plot thickens.
 
olgagroza said:
May be Odyssey have enough powers to reduce the size 4K-to-2K "on fly"?
Click to expand...

I bet that is the answer, though it will be limited to 30fps since it sounds like that may be all Sony is giving them access to. Also, I'd much rather just have a 4x more compressed 4k RAW recording than a slightly better scaled 2k RAW one. Remember 2k RAW will always only store one color per pixel, or only 1/3 of the color information needed for real 2k RGB. You would probably be better off with a 2k RGB/YUV recording from 4k RAW debayer for that reason, and presumably that will be available on the 7Q as well...

I am very skeptical about the image quality of any frame derived from only 2k RAW bayer data. RAW really needs to be oversampled to deliver high quality/resolution without bad aliasing.
 
and in fairness to Dan and CD, let's all collectively temper our expectations since we're now effectively asking CD to fix what is likely a cheat by the Sony engineers to derive a 2K image (something the camera was never really designed to do) by skipping line reads on the sensor side. that said, it will be interesting to see what the boundaries of the agreement between Sony & CD are... I suspect they're limited to how they record 4K but not necessarily what they do with that bitstream after it enters the Odyssey. hell, I'd settle for a really clean 1080 raw, down sampled from the 4K if such a thing could be done.
 
txdp said:
Dan, sort of off topic question: I have had trouble calibrating my OLED smallhd monitor. Any word on calibration accuracy of your panel/software? Thanks!
Click to expand...

Dear Txdp,

We have been comparing our panels accuracy to that of a calibrated Sony 25" OLED monitor and we find it is very close.

We are also working on a way to calibrate it.

Respectfully,


Dan Keaton
Convergent Design
 
blackdogdylan said:
and in fairness to Dan and CD, let's all collectively temper our expectations since we're now effectively asking CD to fix what is likely a cheat by the Sony engineers to derive a 2K image (something the camera was never really designed to do) by skipping line reads on the sensor side. that said, it will be interesting to see what the boundaries of the agreement between Sony & CD are... I suspect they're limited to how they record 4K but not necessarily what they do with that bitstream after it enters the Odyssey. hell, I'd settle for a really clean 1080 raw, down sampled from the 4K if such a thing could be done.
Click to expand...

Dear Friends,

Please rest assured that we will do our best to create HD, 2K and 4K from the signals that we get out of the FS700.

And we will do our best to ensure that these images will be as good as we can make them.

The amount of processing power that we have in the Odyssey7Q is immense.

But, for competitive reasons, it would be inapproraite for me to go into all of the details.

And, our engineering team wants to implement these features and modes, then test them, then thoroughly analyze them for quality, prior to our company making hard promises.

So, another way to put this is: "Stay Tuned".

Respectfully,


Dan Keaton
Convergent Design
 
thanks, Dan. your participation in this community is invaluable. and however it all works out, I'm sure I'll love my 7Q as much (or more than!) my nanoFlash. thanks for listening and responding to those of us in the trenches in a way that Sony never would (or even could.)
 
Dear BlackDogDylan,

Thank you for you thoughts.

I am very happy to answer questions.

We are very excited about our new products.

Our engineers are making great progress also, but I feel that generally I should only share the major milestones, not when we are "close to finished" on certain parts of this project.

Respectfully,


Dan Keaton
Convergent Design
 
This fs700 upgrade is like a rollercoster.. You get your hopes up only to get smashed down again, then comes CD and gets it going up and then comes something else that brings it down.

Well at least now I wont be waisting my money doing an upgrade until one has some proof from CD on how much you will benefit from all this.
 
blackdogdylan said:
that's not exactly what Adam appears to be saying. he writes,

"I surmised (and Sony confirmed) that the same pixel-binning used for 240fps is used in ALL 2k raw capture: that’s how Sony gets 2K out of a 4K sensor yet maintains the same sensor active area."
definitely disappointing, if true. sounds like something is getting lost in the scaling to 2K. perhaps CDs possible solution will be to take the pristine 4K and scale it in the Odyssey. let's hope for that. while workable for high speed, that loss in resolution will be very discouraging in 24fps 2K. the plot thickens.
Click to expand...

Ah, I apologize for being incorrect - I missed that piece of info. That truly is disappointing.
 
Hi Dan,

I'm not sure if this has been covered prior but when using the HDMI or SDI output there is a noticable lag in the picture on an external monitor. Sony has confirmed this: "Yes the SDI output will also have a small frame difference compared to real time. There are no firmware updates planned to change this at the moment."

I've found this frame difference have a negative impact on the footage when trying to follow a person with the camera. Have you noticed this on your upgraded Sony FS700? Any chance there would be a fix for this along with that?
 
I've been hoping that the new raw output of the camera would have a much reduced delay because there's much less processing being done at that stage. Whether the Odyssey7Q can then process and display the image with less than the current delay remains to be seen.
 
Dan, an article on another forum presented the findings of a tester of the FS700 Sony 4K solution. The article noted that when they got home, a lot of the footage was not in focus. This is a red flag to me, as if a tester with credentials worthy to be loaned this camera has problems focusing in the field, what will lesser skilled people (or with lesser equipment) be able to readily achieve. Zooming into a 4K image to grab a regular 1080 HD image is not going to be worth much if things are not in focus. Can you comment on focusing?
 
You must log in or register to reply here.
Back
Top