320Mbps and beyond

  • Thread starter Thread starter Kino Seed
  • Start date Start date
K

Kino Seed

Guest
Don't get too excited... yes it can be done, but there is a catch : )
(I'm sure you will figure it out)

330Mbps: SAM_0112.MP4

here are a couple of frames from above 320Mbps+

ISO 200
ISO 200

ISO 3200
ISO 3200
 
Last edited:
Don't get too excited... yes it can be done, but there is a catch : )
(I'm sure you will figure it out)

330Mbps: SAM_0112.MP4

here are a couple of frames from above 320Mbps+

ISO 200
ISO 200

ISO 3200
ISO 3200
It would also be nice if you told us how you calculated the 330Mbps patch value...
 
Did you get lower quality with higher bitrate as Guru expected?

No, seriously, congrats, man! Fine job. Is it timelapse?
 
Did you get lower quality with higher bitrate as Guru expected?

No, seriously, congrats, man! Fine job. Is it timelapse?
sure... the higher the bitrate, the lower the quality hahahaha : ))
not sure what Guru is smoking, but I did add raw from camera clip, so you decide.
time-laps - yep :)
set x5, and you use any bitrate (almost) you want
 
Last edited:
iso 3200 still unusable?
I wanted to explicitly test the highbitrate with high ISO (to see if it will handle it, and it did). Personally I think it is quite usable, and the noise looks good (not "noise-reduced").
 
Did you get lower quality with higher bitrate as Guru expected?

No, seriously, congrats, man! Fine job. Is it timelapse?
sure... the higher the bitrate, the lower the quality hahahaha : ))
not sure what Guru is smoking, but I did add raw from camera clip, so you decide.
time-laps - yep :)
set x5, and you use any bitrate (almost) you want
I did 180mbps one hour shoot with x20 timelapse but I didn't try 320 )) It's need to be done. As for guru - well it's good to have in mind that any theory need a test that tries to disprove this theory, this is only for the better :)
 
Last edited:
Here is my 2c with NX1

Video xxxx56 - is 80 mbps UHD@30

Video xxxx57 - is 320 mbps UHD@30

Both have same settings, at first one sec it's 3200 ISO and after 6400 ISO. Here is the screengrabs:

ISO3200 - 80 mbps
ISO3200 - 80 mbps

ISO3200 - 320 mbps
ISO3200 - 320 mbps
 
Last edited:
Here is my 2c with NX1

Video xxxx56 - is 80 mbps UHD@30

Video xxxx57 - is 320 mbps UHD@30

Both have same settings, at first one sec it's 3200 ISO and after 6400 ISO. Here is the screengrabs:

ISO3200 - 80 mbps
ISO3200 - 80 mbps

ISO3200 - 320 mbps
ISO3200 - 320 mbps
you probably need something mode dynamic to see the difference : )

however looking at your iso3200, makes me think I may have indeed turned my nr off (by accident) : ))
 
you probably need something mode dynamic to see the difference : )
Of course, but for more dynamic I need to finish my test-stand ))
however looking at your iso3200, makes me think I may have indeed turned my nr off (by accident) : ))
Well, that's how NX1 iso3200 and iso6400 looks with NR-off... or maybe there is some telnet function to do this without NX1 GUI?
 
Last edited:
Last edited:
10000th shutter speed, how u got that?
 
you probably need something mode dynamic to see the difference : )
Of course, but for more dynamic I need to finish my test-stand ))
yeah.. the stand looks suspiciously full : ))
however looking at your iso3200, makes me think I may have indeed turned my nr off (by accident) : ))
Well, that's how NX1 iso3200 and iso6400 looks with NR-off... or maybe there is some telnet function to do this without NX1 GUI?
I've run so many tests and code modifications, including registry modifications, than I care to admint.
however only now when I'm comparing my iso3200 to yours I notice how much different it looks.

When it comes to the effects of the bitrate on quality, I did make that test today, here are 100% crops (used curves at the bottom, to expose the quality).
I hope this will close the question about - are higher bitrates bad for quality, and maybe we can try to prove that the Earth is not flat next.

bdc9bd1c3f2c40609b87003b87a175fb.jpg
 
Last edited:
It would also be nice if you told us how you calculated the 330Mbps patch value...
does it matter? :)
It seems to me that you are under the impression that I make a habit of wasting my time on things that are not important to me.
Considering you calculated the bitrates for the initial mods (on which I based the modified values used later), I can only assume you know how to calculate them. So, why does it matter how were calculated the ones used in the test here?
 
you probably need something mode dynamic to see the difference : )
Of course, but for more dynamic I need to finish my test-stand ))
yeah.. the stand looks suspiciously full : ))
Cheers, man ))
however looking at your iso3200, makes me think I may have indeed turned my nr off (by accident) : ))
Well, that's how NX1 iso3200 and iso6400 looks with NR-off... or maybe there is some telnet function to do this without NX1 GUI?
I've run so many tests and code modifications, including registry modifications, than I care to admint.
however only now when I'm comparing my iso3200 to yours I notice how much different it looks.
Yes! Actually I try to explain... well not explain, just mention this in old vasiles thread about NR, it seems that NX1 and NX500 are using different methods at least to setup NR in GUI, or there is a lack of proper tests on both cameras with same conditions? I see this difference time to time and can't be sure if this a real diff or my imagination
When it comes to the effects of the bitrate on quality, I did make that test today, here are 100% crops (used curves at the bottom, to expose the quality).
I hope this will close the question about - are higher bitrates bad for quality, and maybe we can try to prove that the Earth is not flat next.

bdc9bd1c3f2c40609b87003b87a175fb.jpg
You know, there is always question about quality, if you (or me or anyone) see some quality step in footage CTU, someone just need fixing bug with turn-off time of the camera or rolling shutter, and nothing else matters. It's cool, tho.

Also about bitrate - is there anywhere to go after 320?
 
you probably need something mode dynamic to see the difference : )
Of course, but for more dynamic I need to finish my test-stand ))
yeah.. the stand looks suspiciously full : ))
Cheers, man ))
however looking at your iso3200, makes me think I may have indeed turned my nr off (by accident) : ))
Well, that's how NX1 iso3200 and iso6400 looks with NR-off... or maybe there is some telnet function to do this without NX1 GUI?
I've run so many tests and code modifications, including registry modifications, than I care to admint.
however only now when I'm comparing my iso3200 to yours I notice how much different it looks.
Yes! Actually I try to explain... well not explain, just mention this in old vasiles thread about NR, it seems that NX1 and NX500 are using different methods at least to setup NR in GUI, or there is a lack of proper tests on both cameras with same conditions? I see this difference time to time and can't be sure if this a real diff or my imagination
When it comes to the effects of the bitrate on quality, I did make that test today, here are 100% crops (used curves at the bottom, to expose the quality).
I hope this will close the question about - are higher bitrates bad for quality, and maybe we can try to prove that the Earth is not flat next.

bdc9bd1c3f2c40609b87003b87a175fb.jpg
You know, there is always question about quality, if you (or me or anyone) see some quality step in footage CTU, someone just need fixing bug with turn-off time of the camera or rolling shutter, and nothing else matters. It's cool, tho.
Comparing the detail in highlights I think is crystal clear the effects of bitrate, and it does not stop at 80 nor 180.
If this is general situation when it comes to NR for NX500 I don't know, but here is some 180Mbps 3200 direct from camera in case you want to compare:
180Mbps - ISO3200
Also about bitrate - is there anywhere to go after 320?
Doubt it, even 320 is not really needed ihmo.
Remember how recording black field and only changing the ISO ( noise ) at first produced lower than the set bitrates (because there was not enough visual information to fill the bitrate), and only later (at high noise) it started maxing out? something like this is happening here too.
The weather was not very good, so I ran a few quick tests, at different high bitrates, however in 50% of the cases the effective bitrates did not go over 300Mbps, and from 20-something videos, only 4 went over 330, and of those 4, only 2 were about 340.

So for any practical reasons I really don't see the need for anything over 300Mbps, and this is only possible when timelapsing x5.
 

Keyboard shortcuts

Back
Top