1200 fps is NOT True

Started May 5, 2012 | Discussions
ItalianEngineer
New MemberPosts: 5
Like?
1200 fps is NOT True
May 5, 2012

Video at 1200fps on Nikon J1, is not TRUE!
Real video recording is at 600fps, the camera duplicate every frame!!!!!

I shooted an hi-speed video (1200fps) to my analog oscilloscope,
connect to function generator.

I put on input of oscilloscope a tringolar wave at 1200Hz (exactly at 1198.8Hz=29.97x40),

I shoot that wave, and I note that the effective frame rate is 600fps NOT 1200fps.
( I get 2 periods of wave instead of 1, on every frame recorded)

The camera duplicate all frame inside, and all appear to be at 1200fps, but is not true!
It's simple to check using a digital video editor, like premiere or similar,

or with the camera: in play mode, frame by frame, you can see that every frames are duplicate.
Why Nikon said that this camera can record at 1200fps?
I bought this camera for this features only, and it doesn't works.

All Hi speed camera is very important for industrial video, and for sports analysis,
I have another Camera: Casio Hi-Speed, that works perfectly.

Nikon 1 J1
If you believe there are incorrect tags, please send us this post using our feedback form.
normsmith
Veteran MemberPosts: 3,372
Like?
Re: 1200 fps is NOT True
In reply to ItalianEngineer, May 5, 2012

Interesting - have you tested the slower 400 fps setting? this is the one I use, simply because it plays back at a higher resolution.

I am guessing that for the general shooter, the double framing will not matter as play back remains smooth (i.e. your oscillator picked this up, not your eye. so for most people it is a technical issue that they won't even see rather than a practical one).
--
my blog http://pinkfootstudio.blogspot.com
never more than 3 posts per page so you will not be overloaded with images.

Reply   Reply with quote   Complain
Paul Pasco
Veteran MemberPosts: 4,375Gear list
Like?
Re: 1200 fps is NOT True
In reply to normsmith, May 5, 2012

normsmith wrote:

Interesting - have you tested the slower 400 fps setting? this is the one I use, simply because it plays back at a higher resolution.

I am guessing that for the general shooter, the double framing will not matter as play back remains smooth (i.e. your oscillator picked this up, not your eye. so for most people it is a technical issue that they won't even see rather than a practical one).

Yeah but isn't this a form of fraud? I don't care myself but what if somebody bought the camera for this feature?
--
Regards, Paul

Lili's Dad

 Paul Pasco's gear list:Paul Pasco's gear list
Panasonic Lumix DMC-FZ28 Nikon D50 Nikon D300 Nikon 1 V1 Nikon 1 J1 +14 more
Reply   Reply with quote   Complain
clide
Junior MemberPosts: 28
Like?
Re: 1200 fps is NOT True
In reply to ItalianEngineer, May 6, 2012

Wow, I hadn't done much with 1200 fps, but I just checked some test shots that I did and you are absolutely right.

That is disappointing for several reasons. Not only are they lying about 1200 fps but they are also making the file size unnecessarily bigger by duplicating each frame. They could have done 600 fps at a slightly higher resolution if they weren't going to really do 1200 fps.

Have you tried to contact Nikon about this?

Reply   Reply with quote   Complain
normsmith
Veteran MemberPosts: 3,372
Like?
Re: 1200 fps is NOT True
In reply to clide, May 6, 2012

I am guessing that they need the higher frame rate to provide a relatively useful period of slow motion footage that is sufficient to tell a story - the demands on the camera and memory must be quite staggering. It works in principle and seems to be a balancing act between frames taken and playback speed to get an acceptable result.

I think this has to be seen within the context of a £400 - £600 consumer product that is delivering something already quite advanced (with more development needed to improve it in the future to get higher resolution and longer footage) and that is designed to give the 'man in the street' a powerful feature - it is what it is. I have used it and enjoyed the results.
--
my blog http://pinkfootstudio.blogspot.com
never more than 3 posts per page so you will not be overloaded with images.

Reply   Reply with quote   Complain
clide
Junior MemberPosts: 28
Like?
Re: 1200 fps is NOT True
In reply to normsmith, May 6, 2012

Yes, but it is still entirely dishonest and misleading to advertise it as 1200 fps when it is actually only recording 600 unique frames per second. That's like taking a 1080 30p file duplicating every frame and calling it 60p. Yes it may technically be encoded as 60p, but it's giving you the exact same information as the 30p file.

I am aware that bandwidth and high speed memory limitations that make this kind of high speed video difficult, but the nearly 5 year old Casio EX-F1 could do 1200 fps and was actually able to deliver.

Reply   Reply with quote   Complain
Bruce McL
Senior MemberPosts: 1,099
Like?
Re: 1200 fps is NOT True
In reply to clide, May 7, 2012

clide wrote:

... That's like taking a 1080 30p file duplicating every frame and calling it 60p.

Or it's like taking an image from a 2MP sensor, enlarging it in the camera to 4MP, and advertising the camera as having 4 megapixels. I expect to see the word "interpolated" added if that is done.

Reply   Reply with quote   Complain
myzel
Senior MemberPosts: 2,073Gear list
Like?
Re: problems with that test
In reply to ItalianEngineer, May 7, 2012

ItalianEngineer wrote:

I put on input of oscilloscope a tringolar wave at 1200Hz (exactly at 1198.8Hz=29.97x40),

I shoot that wave, and I note that the effective frame rate is 600fps NOT 1200fps.
( I get 2 periods of wave instead of 1, on every frame recorded)

To me this sounds a lot like a problem with the Nyquist frequency. With a 1200 fps sampling rate it should only be possible to completely capture a frequency of 600 hz.

And btw, does the display of your oscilloscope really have a refresh rate of 1200 hz?

A better test, in my opinion, would be capturing a slower wave (for example a pendulum, and not a wave form on a monitor) and then measuring the angle between the frames. This angle can be used to calculate the time between the frames and therefore the fps.

 myzel's gear list:myzel's gear list
Nikon 1 V1 Nikon 1 Nikkor 10mm f/2.8 Nikon 1 Nikkor 18.5mm f/1.8 Nikon 1 Nikkor 32mm f/1.2
Reply   Reply with quote   Complain
olyflyer
Forum ProPosts: 23,257
Like?
Re: 1200 fps is NOT True
In reply to ItalianEngineer, May 7, 2012

ItalianEngineer wrote:

Video at 1200fps on Nikon J1, is not TRUE!
Real video recording is at 600fps, the camera duplicate every frame!!!!!

I shooted an hi-speed video (1200fps) to my analog oscilloscope,
connect to function generator.

I put on input of oscilloscope a tringolar wave at 1200Hz (exactly at 1198.8Hz=29.97x40),

I shoot that wave, and I note that the effective frame rate is 600fps NOT 1200fps.
( I get 2 periods of wave instead of 1, on every frame recorded)

The camera duplicate all frame inside, and all appear to be at 1200fps, but is not true!
It's simple to check using a digital video editor, like premiere or similar,

or with the camera: in play mode, frame by frame, you can see that every frames are duplicate.
Why Nikon said that this camera can record at 1200fps?
I bought this camera for this features only, and it doesn't works.

All Hi speed camera is very important for industrial video, and for sports analysis,
I have another Camera: Casio Hi-Speed, that works perfectly.

If you are right about this, then this is a scandal and a fraud. I'll think about how to test this properly, I am not sure your test is the right way, but it definitely needs to be double checked before considering it a fraud by calling the camera 1200fps.

Reply   Reply with quote   Complain
clide
Junior MemberPosts: 28
Like?
Re: problems with that test
In reply to myzel, May 7, 2012

myzel wrote:

To me this sounds a lot like a problem with the Nyquist frequency. With a 1200 fps sampling rate it should only be possible to completely capture a frequency of 600 hz.

And btw, does the display of your oscilloscope really have a refresh rate of 1200 hz?

That only applies if he was capturing the value of the wave at 1200hz, in this case he is capturing the entire wave by taking a picture of it, so he could even capture higher frequencies.

The oscilloscope refresh rate is a concern, but you can take a video of absolutely anything moving quickly at 1200fps and see for your self that each frame is duplicated.

The only other concern would be if they were adding the frames to slow down the play-back rate but still recording 1200fps. This can easily be tested by filming a clock with a second hand. You'll find that the resulting video which is being played back at 30 fps takes 40 seconds between each tick of the clock. 30x40=1200. So they are indeed counting those duplicated frames in the 1200 fps figure.

Reply   Reply with quote   Complain
ItalianEngineer
New MemberPosts: 5
Like?
Re: 1200 fps is NOT True
In reply to normsmith, May 8, 2012

400fps Works Properly.

I did a lots of tests, the last test with analogue oscilloscope confirm this problem.

I did another test, I put a disc with black and white radial stripes on an electric motor,
and I take a video ad 1200fps.
Result: Every Frame is duplicate!

I suppose 2 things:

1) It was a bug in the firmware (the J1 have 2 processor; may be the firmware copy the same frame from one processor only). A solution is possible.

2) For some reasons it's impossible to recording at 1200fps, maybe bandwith limit, or time to compress each frame; and this is a "workaround" waiting for next model, because in this case, there is NO solution.

I did contact Nikon about 2 months ago,
they wrote me, that problem exists!!, And they will work on new firmware.
I post this message, because I don't want to wait more.

Reply   Reply with quote   Complain
Jimmy jang Boo
Senior MemberPosts: 1,314Gear list
Like?
Re: 1200 fps is NOT True
In reply to ItalianEngineer, May 8, 2012

ItalianEngineer wrote:

I did contact Nikon about 2 months ago,
they wrote me, that problem exists!!, And they will work on new firmware .

That's good news. Hopefully the new FW fixes many (if not all) the previously addressed issues.

 Jimmy jang Boo's gear list:Jimmy jang Boo's gear list
Ricoh GR Nikon 1 V1
Reply   Reply with quote   Complain
Johannes Zander
Regular MemberPosts: 205Gear list
Like?
Re: 1200 fps is NOT True
In reply to Jimmy jang Boo, May 8, 2012

Nikon 1 V1-Firmware-Update 1.12 available. Fixes 1200 fps frame duplication.

 Johannes Zander's gear list:Johannes Zander's gear list
Nikon 1 V1 Olympus OM-D E-M5 Nikon D800 Nikon AF-S Nikkor 14-24mm f/2.8G ED Nikon AF-S Nikkor 70-200mm f/2.8G ED VR II +16 more
Reply   Reply with quote   Complain
Seppi Evans
Junior MemberPosts: 32Gear list
Like?
New Firmware 1.12 released!
In reply to olyflyer, May 8, 2012

An issue that caused some frames to be duplicated in slow-motion movies recorded at a Frame rate setting of 1200 fps has been resolved.

 Seppi Evans's gear list:Seppi Evans's gear list
Canon EOS 100D Canon EF-S 18-55mm f/3.5-5.6 IS STM
Reply   Reply with quote   Complain
wlad
Senior MemberPosts: 1,483
Like?
Re: New Firmware 1.12 released!
In reply to Seppi Evans, May 8, 2012

LOL, so much for the conspiracy about fake features and stuff

Reply   Reply with quote   Complain
ItalianEngineer
New MemberPosts: 5
Like?
Re: New Firmware 1.12 released!
In reply to wlad, May 8, 2012

Where we can find the 1.2 firmware version?
On Nikon support site the latest firmware version is A:1.11 / B:1.10

Reply   Reply with quote   Complain
olyflyer
Forum ProPosts: 23,257
Like?
Good news!
In reply to Seppi Evans, May 8, 2012

Seppi Evans wrote:

An issue that caused some frames to be duplicated in slow-motion movies recorded at a Frame rate setting of 1200 fps has been resolved.

Excellent news. Hopefully the new firmware includes some other documented and undocumented improvements. I’ll be downloading it tonight.

What this confirms in my opinion is that Nikon did not abandon the N1 and takes it seriously. I don't care much about the 1200fps thing, for me it is just a technical challenge to check that it is rally 1200 individual frames, but what is important that Nikon did something about the N1 and not just leave it as it is.

Reply   Reply with quote   Complain
olyflyer
Forum ProPosts: 23,257
Like?
It was not fake, but not working as expected...
In reply to wlad, May 8, 2012

wlad wrote:

LOL, so much for the conspiracy about fake features and stuff

Well, it confirmed that the feature was not working as expected. Good that Nikon took care of the problem, I never thought there was any conspiracy and was quite sure it was an error. Nikon is not know to have deliberately committing fraud, claiming something which is fake, but perhaps threads like these and the mail conversations one can have with them is not really wasted time after all. If an error is found we must let them know otherwise it might not be considered as an error and never detected/corrected.

Reply   Reply with quote   Complain
olyflyer
Forum ProPosts: 23,257
Like?
Re: New Firmware 1.12 released!
In reply to ItalianEngineer, May 8, 2012

ItalianEngineer wrote:

Where we can find the 1.2 firmware version?
On Nikon support site the latest firmware version is A:1.11 / B:1.10

You don't mean it was faked news...?

Reply   Reply with quote   Complain
clide
Junior MemberPosts: 28
Like?
Re: Good news!
In reply to olyflyer, May 8, 2012

Just tested it out and it is fixed. Makes 1200fps much smoother (and more usable from a technical standpoint).

That's good news that it was just a firmware bug. I suspect that this update was pushed out on its own so it could be taken care of quickly. Many countries take false advertising claims very seriously and this one was easily demonstrated to be 'not as advertised'. Interesting how they worded it as "An issue that caused some frames to be duplicated" when in fact it was all frames being duplicated.

Thanks for pointing it out ItalianEngineer, and thanks for fixing it Nikon.

Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads