View Full Version : NanoFlash and slowmo


Arnt Mollan
July 21st, 2012, 07:37 AM
Hi. I'm trying to get the fs700 to work with slowmo recording to my NanoFlash. As soon as I touch the SQ or SuperSlow button the signal to NanoFlash is lost. I see others have managed to record to external recorders. What do you do to get this to work?

Cees van Kempen
July 21st, 2012, 12:59 PM
you have to set output to 1080i/480i, not to 1080p/480i

Arnt Mollan
July 21st, 2012, 01:58 PM
Thank you Cees. That worked. No record with autostart trough TC then, as far as I can see, but I guess we cant have it all.

Arnt Mollan
July 22nd, 2012, 12:46 AM
Interesting. I can record a movie with NanoFlash in real-time, while the camera buffer to slowmo. Hit the record button on the camera and record a few seconds of slow motion to NanoFlash, abort the slow motion from the camera and directly continue to record real time. In one take. And the camera starts to buffer slowmo again... :)

Cees van Kempen
July 22nd, 2012, 01:20 AM
However, my recording of the slomo to the nanoflash is not good. The camera writes the 240 fps footage away @60fps, but the nano can only handle 30fps. So it is supposed that the nano takes every other frame, creating a slomo based on 120fps instead of 240 fps. So far so good. But...... It seems the nano does not take every other frame, but the even lines from one frame and the odd lines from the next. So it makes a kind of interlaced writing in a progressive frame, clearly showing the even and odd lines at different moments in time in the same frame and thus creating an useless image.
I have not heard other people reporting this. Am I doing something wrong?

Chris Medico
July 22nd, 2012, 06:55 AM
You are not doing anything wrong. That is how it is. You are getting an interlaced output with half of each of the 60 frames.

To get the best external recording of the slo-mo from the fs700 you need a 60p recorder.

Arnt Mollan
July 22nd, 2012, 02:23 PM
Hi Cees. Just looked at my footage on a big screen, direct from NanoFlash, 100 fps 25p, wildly moving camera, and the video looks fine. Recorded at 100mps MXF. Not tested in post yet. But if there was some field problems I should have seen them.

Cees van Kempen
July 24th, 2012, 01:26 PM
Herewith a part of a frame. One shot in camera with the FS700 @240fps. The other is the same shot but than recorded to the nanoflash. clearly see how the dragonfly is build up interlaced. Am I doing something wrong? Maybe Dan Keaton can shine a light on this matter, or Alister Chapman who reported that slomo footage can be shot at the nanoflash at half of the framerate of the camerasetting. Though I highly value my nanoflash it is virtually useless in combination with the slomo recordings of the FS700, though knowledgeable people have reported that the footage can be recorded to extermal devices that can handle 30fps (at half the framerate).

Cees van Kempen
July 26th, 2012, 09:19 AM
Is Chris right? This means that the statements in other threads that the nanoflash can be used for slomo is not correct. I mean, a progressive frame with two interlaced halves at a different moment in time is in my opinion not usefull. Alister? Dan? Can you give a clue?

Chris Medico
July 26th, 2012, 11:24 AM
My testing with a PIX240 and the FS700 indicates that SDI and HDMI streams contain a normal interlaced signal with 2 fields per frame separated by 1/60th of a second. It is not PsF or true Progressive.

Since that is what the camera is outputting I don't see why anyone will get a different result with a NanoFlash or any other SDI/HDMI recorder.

As soon as I can get my hands on a friends NanoFlash I'll test it.

Cees van Kempen
July 26th, 2012, 02:33 PM
What would that mean for the FS700 with a Gemini? Will that result in PsF frames, or will it also result in progressive frames with an interlaced structure?

By the way, I believe the nano would switch automatically to interlaced recording if the output of the camera is interlaced. But the nano records in PsF, which I believe means the output signal of the FS100 is PsF. I think the FS700 outputs P or PsF at 60 fps. The nano records at 30fps PsF. The device 'thinks' it takes the odd lines and than the even lines from the same frame, but it actually takes the odd from one frame and than the even from the next, because the input does not come at 30fps but at 60 fps.

Chris Medico
July 26th, 2012, 02:52 PM
When you have the camera output set to progressive (at 60p) it outputs a normal progressive signal. There is no PsF at 60p.

The Gemini can record 60p so with the FS700 set to 60p and its output set to 60p you can get the full quality image externally with it or any true 60p recorder.

I suspect the ability to output 60i when the camera is recording 60p is a compatibility thing. My SmallHD monitor doesn't like a 60p signal but works fine with 60i.

Alister Chapman
July 29th, 2012, 03:58 PM
Cees was the camera set to progressive or interlace? I need to check this again as I have not seen such interlace artefacts. Could just possibly be one of the issues with only having LCD monitors these days. If the 700 is outputting interlace just un check the PsF box on the NanoFlash and it should play back as an interlaced video with the separate fields played back one offer the other.

Most of my testing was done with a Gemini at 60p, I'll try the NanoFlash again at 30p.

Chris, an external recorder can't tell the difference between PsF and interlace as electronically both signals are the same.

Cees van Kempen
July 30th, 2012, 02:09 AM
Alister, The camera was on progressive. The nano was PsF and the output of the camera 1080i/480i, because otherwise there is no signal to the nanoflash. Capturing interlaced on the nano might be an option (I did not check if it works), but all my work is progressive, so I do not see much benefit in capturing interlaced. I think I have to live with the fact that the nano does not do what I thought it would, capturing half of the frames in progressive (PsF).

Probably it could be changed in the software. Not taking the second field from the next frame, but from the same frame again. Ignoring every even frame. Dan...?

Cees van Kempen
August 6th, 2012, 08:29 AM
Most of my testing was done with a Gemini at 60p, I'll try the NanoFlash again at 30p.

Alister,

Have you maneged to test the nanoflash again and can you confirm that it is not possible to record PsF because of the interlaced signal?

Cees

Alister Chapman
August 8th, 2012, 01:15 PM
Yes I did test it and sadly the output is interlaced when the camera is set to output 25/30p.

It would be interesting to take that interlace signal and see what happens when you slow it down by 50% in a progressive project in different NLE's. If they split out the two fields and show them one after the other then although there will be a vertical resolution drop it shouldn't look that bad as alternate frames will have slightly different vertical picture information. So the perceived resolution drop won't be half it will probably appear to be around a 1/3rd loss.

Chris Medico
August 8th, 2012, 07:34 PM
Exactly the same result as I found Alister. Sad indeed.

It would be nice if Sony could do a firmware update that allowed for a 30p write to the memory card. I would be willing to wait twice as long for those times if I needed what an external recorder offered.

Cees van Kempen
August 11th, 2012, 01:49 PM
And be aware that this is the case whenever you have the camera in Quick & Slomo mode, regardless if you are actually recording in slomo or not. So if you record real time on your nanoflash while not recording in the camera to capture the slomo footage (thus not pressing the start button to actually start the slomo capture), than the captured footage is interlaced in PsF as well and thus useless. So you can not record on your nanoflash while waiting for the slomo shot to capture in camera.

Chris, your right. I would fully support the idea of writing it to the memory card at 30p and would be wailling to wait twice as long if that solves the problem. Ideally it would be an menu option with the choice between 30p and 60p as write speed. And maybe the solution could be found in an adjustment of the nanoflash as well, though I wouldn't know how. Would like to see Dan keaton come in on this subject.

Dan Keaton
September 4th, 2012, 06:53 AM
Dear Friends,

I am very sorry that I have not seen this thread before, thus I am very late to the conversation.

It appears that we need to get an FS700 into our lab for testing with the nanoFlash.
(We may already have one, I can check in a few hours. We have an FS100, but it will not substitute for an FS700 for this testing.)

I always trust what Alister reports and he says that the output is interlaced when the camera is set to 25p/30p.

This is a condition that I did not expect.

While manufacturers sometimes call (for simplicity sake) PSF (Progressive Segmented Frames) interlaced, it is not.

With PSF, there is zero time difference between the two fields in a frame, and thus we can convert back to true progressive.

With true Interlaced, there is a time difference between the two fields in one frame and we have no way to convert this back to true progressive.

If the output is PSF, then one can set, in the nanoFlash, Video|Record PSF > Prog(ressive) to checked.

But, if the output is actually interlaced, then setting the above is generally not a good idea.


What I find strange, is that many modern cameras, such as the FS700, have progressive sensors and creating true interlaced is a lot of work.

I do not understand what they are doing to get interlaced, as opposed to PSF, as this is a lot of extra work. If it is truely interlaced, then I assume that this was done on purpose (with great effort) so that fast motion would appear smoother on playback.

My recommendation, if it is interlaced, would be to record it as interlaced, with our Video|Record PSF>Prog(ressive) Unchecked.

While I have a huge fan of recording progressive, interlaced does have a place, and is appropriate for certain fast moving objects.

While viewing interlaced footage at normaly speed is ok, freezing a frame and examining it will show that it is interlaced.

I hope this helps.

Cees van Kempen
September 5th, 2012, 12:45 PM
Thanks for coming in, Dan.

Please let us know when you have tested the FS700 if there is any possible solution to this matter. I am sure many nanoflash users would be very grateful.

Cees

Dan Keaton
September 5th, 2012, 12:49 PM
Dear Cees,

I have determined that we do not have a FS700 in our lab, just a FS100.

However, one of our friends has both the nanoFlash and FS700 and has offered to run some tests.

Cees van Kempen
September 5th, 2012, 12:54 PM
Thanks Dan. We will await your findings.

I see that CD is present at IBC and that there even is a chance to win a gemini 4:4:4. You bet I will visit your boot !!

Dan Keaton
September 5th, 2012, 12:56 PM
Dear Cees,

That is great. Mike Schell and Amber Cowles will be there.

This is a very nice price. Just be certain to register for the prize.