Jump to content

Petition for Samsung NX1 hack


kidzrevil
 Share

Recommended Posts

6 hours ago, Arizona Sunset said:

Some folks have mentioned a desire for a crop mode for NX1 to match the NX500's 4k crop and downsampling.  Out of curiosity, how does the NX1 implement digital / clear zoom during video OOB? Is the request for the feature because there is ineffective or poor quality digital zoom, or because there is none at all?  

Nx1 only shoots with a full sensor readout (1.5 crop factor from 35mm fl). Nx500 only shoots with cropped resolution, i.e. a total of about 2.53 crop factor from ff fl).

 

Hence the 150mm has an effective fl of 225mm on nx1 and of 380mm on nx500. 

 

The cool thing would ve being able to use both on the nx1, so to use a digital zoom with the same lens, as Mattias already explained.

Link to comment
Share on other sites

6 hours ago, Marco Tecno said:

Nx1 only shoots with a full sensor readout (1.5 crop factor from 35mm fl). Nx500 only shoots with cropped resolution, i.e. a total of about 2.53 crop factor from ff fl).

 

Hence the 150mm has an effective fl of 225mm on nx1 and of 380mm on nx500. 

 

The cool thing would ve being able to use both on the nx1, so to use a digital zoom with the same lens, as Mattias already explained.

Agreed, And with cropping the 4K in post would allow some decent reach. Probably somewhere around 450mm(35 equivalent) before the IQ got too bad. The IQ even at 380 would be quite bad but would definitely be usable in a pinch.

Link to comment
Share on other sites

20 minutes ago, Marco Tecno said:

Why do you say IQ would be bad? It would have a 1:1 pixel readout w/o downscale, but no line skipping or else. Hence should be fairly good with a good lens, like 50-150s is...

The NX1 and NX500 have the same sensor so this is not really a theoretical question and is something someone can test right now. I do not own an NX500 but from what I have seen the IQ is not as good as the NX1.

It would be interesting to compare the NX1 cropped to 1080 and the NX500 with same lens and FL cropped to 1080 and well.

My money is the IQ on the NX500 is noticeably worse.

 

Granted there would be less moiré but that has never been an issue for me on the NX1 at 4K.

Link to comment
Share on other sites

8 hours ago, Marco Tecno said:

Nx1 only shoots with a full sensor readout (1.5 crop factor from 35mm fl). Nx500 only shoots with cropped resolution, i.e. a total of about 2.53 crop factor from ff fl).

 

Hence the 150mm has an effective fl of 225mm on nx1 and of 380mm on nx500. 

 

The cool thing would ve being able to use both on the nx1, so to use a digital zoom with the same lens, as Mattias already explained.

Then for the NX1, perhaps the technique would be to either a) grab and crop the readout stream prior to downsampling, or b) constrain the readout area to match the NX500's.

This does not sound like low hanging fruit either way, but I'll root for you!

You could take the approach of digging into out how other cameras implement "digital zoom" in video, e.g., I use 1.5x digital zoom on top of either FF (4K and 1080P) or APS-C (1080p) mode on the S II, and it is almost lossless in the final form.  Could be a model for the NX1 digital zoom hack.

Link to comment
Share on other sites

Arizona, I guess the best (and simplest) approach, for a coder/hacker, would be to see how this is done in NX500, which has the same hw of NX1 (or a lower one), and transpose those routines/algorithms to NX1 in a custom fw. The most "difficult" thing would probably implement a switch to select one or the other, since this is not already present in the menu. Perhaps one could "kill" the (imo useless) real 4k mode on NX1, leaving the spaceholder for the cropped UHD one.

Link to comment
Share on other sites

On 3/1/2016 at 1:25 PM, kidzrevil said:

Which causes more problems for you in post : low contrast or low saturation ?

i am currently experimenting with low saturation , sharpness completely off and contrast at default with the16-235 luma range. The camera is problematic in post at 0-255

as I've found lower contrast to be more beneficial to a compressed image, that's what I've tested the most, and has really been the source of most of my gripes with the camera. The ability to lower saturation hasn't seem to yield many benefits in regards to post work, so I've only tried lowering it maybe to -5 at the most extreme, and I didn't see much benefit so I usually leave saturation at 0 now.

With -5 contrast, 0-255, -10 sharpness and 0 saturation I've found the least amount of negative effects on the image, while gaining the most extra latitude. I have not found gammaDR to yield particularly better results (in regards to grading latitude).

Link to comment
Share on other sites

14 hours ago, caseywilsondp said:

as I've found lower contrast to be more beneficial to a compressed image, that's what I've tested the most, and has really been the source of most of my gripes with the camera. The ability to lower saturation hasn't seem to yield many benefits in regards to post work, so I've only tried lowering it maybe to -5 at the most extreme, and I didn't see much benefit so I usually leave saturation at 0 now.

With -5 contrast, 0-255, -10 sharpness and 0 saturation I've found the least amount of negative effects on the image, while gaining the most extra latitude. I have not found gammaDR to yield particularly better results (in regards to grading latitude).

Casey, if you did any scientific tests, could you share your results? Image examples for downloads and detailed overview of your findings. Details Details Details pleeease, because we are keen to find out also!

Link to comment
Share on other sites

43 minutes ago, Pavel D Prichystal said:

Casey, if you did any scientific tests, could you share your results? Image examples for downloads and detailed overview of your findings. Details Details Details pleeease, because we are keen to find out also!

I think it depends on everyone's taste - someone prefers dynamic range (and more filmic?), someone likes richer and more accurate colours. I have almost same settings as Casey... I did not make any scientific tests but my opinion is that (with this limited bitrate and 8bit codec) it is best solution to get best picture straight out of camera. But please - let's focus on hacking of NX1 in this thread... :-)

Link to comment
Share on other sites

If raw video gets unleashed through hacking, then all these questions will become 'the past'. Consider this: nx1 does 1) full sensor readout up to 60fps, then 2) scales it to 4k or 2k, then 3) encodes it in h.265 in real time.

Now...think about jumping steps 2 and 3 and just saving the raw sampled 6.5k image to sd card. Even if that were 12bit instead of 14, even if 24/25fps instead of 30/60, even if only for few minutes, this would be the greatest achievement ever in a sub 1000$ camera.

 

I would buy a couple of spare bodies if this ever happens.

 

 

Link to comment
Share on other sites

12 minutes ago, Marco Tecno said:

If raw video gets unleashed through hacking, then all these questions will become 'the past'. Consider this: nx1 does 1) full sensor readout up to 60fps, then 2) scales it to 4k or 2k, then 3) encodes it in h.265 in real time.

Now...think about jumping steps 2 and 3 and just saving the raw sampled 6.5k image to sd card. Even if that were 12bit instead of 14, even if 24/25fps instead of 30/60, even if only for few minutes, this would be the greatest achievement ever in a sub 1000$ camera.

 

I would buy a couple of spare bodies if this ever happens.

 

 

But RAW image has to be scale down because bitrate would be insane in this case (24x around 30MB = 720MB/s). Maximum write speed of NX1 is 60MB/s (see http://www.cameramemoryspeed.com/samsung-nx1/sd-card-comparison/ )

Link to comment
Share on other sites

It's magic!

No, its the Megabyte (MB) and a Megabit (Mb) confusion:

https://opensignal.com/knowledgebase/the-difference-between-megabyte-and-megabit.php

https://www.google.de/search?q=megabyte+megabit+converter&gws_rd=cr&ei=463ZVumTOMb36ASKwp6wBg

The NX1 writes with 80 Mb/s and thats 10 MB/s

so possible is 60 MB and that's a 480 Mb Bitrate for the movie size.

 

Edit: But the Size of one 6.5k RAW-Picture is round about:

3:2-Format: 6.840 x 4.320 = 31,1 MB

6.840 x 3.648 = 26,2 MB * 24 Pics/s = holy s... that's impossibel!

 

But UHD with 400 to 480 Mbit/s vs. the original NX1 80 Mbit/s is a really heavy upgrade ... with 10 bit instead the 8 bit, i think it's a big thing.

 

 

NX1_Manual_Page 132.JPG

Link to comment
Share on other sites

What about decoding?

I feel like decoding 400Mbits/s HEVC  would be ridiculous both in time and finale file size.

I also think that most computers wouldn't be able to edit 400 MBits/s HEVC natively using adobe cc (or any other program that can handle it natively like edit ready or wondershare).

Link to comment
Share on other sites

2 minutes ago, MountneerMan said:

What about decoding?

I feel like decoding 400Mbits/s HEVC  would be ridiculous both in time and finale file size.

I also think that most computers wouldn't be able to edit 400 MBits/s HEVC natively using adobe cc (or any other program that can handle it natively like edit ready or wondershare).

I think "just" 160/200 Mbits H265 would improve image a lot. But question now is not "how much" but "how"...:-)

 

Link to comment
Share on other sites

6 minutes ago, MountneerMan said:

What about decoding?

I feel like decoding 400Mbits/s HEVC  would be ridiculous both in time and finale file size.

I also think that most computers wouldn't be able to edit 400 MBits/s HEVC natively using adobe cc (or any other program that can handle it natively like edit ready or wondershare).

I definitely wont be able to work with compressed footage at that data rate. If we can get it to 100 or even the highest the uhs ii cards can handle would be great. 100 mbs with noise reduction and in cam sharpening off would be easier on the codec anyway

Link to comment
Share on other sites

6 hours ago, Pavel D Prichystal said:

Casey, if you did any scientific tests, could you share your results? Image examples for downloads and detailed overview of your findings. Details Details Details pleeease, because we are keen to find out also!

I have made posts and posts about it in the "your ideal nx1 settings" thread. scientific depends on your definition of what a proper scientific approach would be, but i felt i was pretty methodical in trying to get the best image out of the camera possible. the problem is because I didn't apply a crazy filmy looking lut the end result, I dont think anyone cared

Link to comment
Share on other sites

7 minutes ago, caseywilsondp said:

I have made posts and posts about it in the "your ideal nx1 settings" thread. scientific depends on your definition of what a proper scientific approach would be, but i felt i was pretty methodical in trying to get the best image out of the camera possible. the problem is because I didn't apply a crazy filmy looking lut the end result, I dont think anyone cared

Lmao I appreciated your tests bro ! Actually going to try playing with the -5 contrast just with 16-235 instead as I feel h.265 works better with it. No macroblocking

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...