-
Posts
1,014 -
Joined
-
Last visited
Content Type
Profiles
Forums
Articles
Everything posted by Marco Tecno
-
I think this is impossible. There must be, somewhere, a 'step' where nr is introduced somehow. It could be done via a hw function, yes, but it has to be invoked by a sw call.
-
Did you use the high bitrate hack for nx1? Which settings (apart gammaDR) did you use for your test? Did you try andrew's log profile for nx1? Thx.
-
call me crazy, but I always use auto wb...
-
As imagined. It would be nice to port onto nx1 the two features of nx500 (compressed raw as an alternative and cropped 4k, as a sort of losless digital zoom for video).
-
I'll never understand if it's better 0-255 or 16-235. In the profile that Andrew made for nx1, 16-235 needs to be used.
-
110 is already a nice jump from the standard 80. Almost 40% more, actually.
-
Pavel, very interesting. I still can't understand the reason for the reduced framerate when a higher bitrate is chosen. The higher the bitrate, the less work cpu has to do. Btw, is there a way to create a shortcuts for video like this? : Ev+fn= 4k, pro, pal, 25fps, 1x, dis Ev+mobile= 1080, high, ntsc, 30, 0.25x, no dis
-
Did somebody solve the 106fps issue (instead of 120fps) when using higher bitrates?
-
If the macroblocking is still somewhat present at 160mb/sec, could it be the case where it's not due to compressed bitrate but rather to the way h.265 works?
-
otto is working on a gui for the scripts he, vasile and hirsti produced. This is not very far from a magic lantern sw. as to the nr, there is the option to enable or disable it, but I never understood if it affect also video or only jpg stills. It should be investigated and seen if it stores the variable of 'nr on' somewhere, like it happens for bitrate.
-
I agree that a very big thing for nx would be turning off the terrible nr. And when I say 'off' I mean really off, completely. Both in jpg pics and - much more important - video.
-
as said in the other thread, you should definitely take a look here: http://***URL removed***/forums/post/57572799
-
this is important, imo: http://***URL removed***/forums/post/57572799
-
Fine details are much better with 160mb. I guess that's a sweet spot for 8bit 420. Would be really great to see 10bit and 422, but I doubt it's feasible.
-
Samsung hack forum for discussion of all cameras not just NX1
Marco Tecno replied to Andrew Reid's topic in Cameras
Andrew, aren't you going to test the high bitrate hack on nx1? -
One thing I don't understand. 1frame at 4k is = 4 frames at 2k. If I can encode 4k@30 at 160mb, why not 2k@120?
-
120fps sorta requires iso to be max at 400. What happens with the frame rate? Why 109?
-
could be interesting!!
-
Too bad. I have most nx lenses and some of those I was able to get for decent/good prices. But it's so sad to see such a great system slowly disappearing w/o a formal announcement.
-
SD-Card Benchmark (with Samsung tool in DEV mode)
Marco Tecno replied to derderimmermuedeist's topic in Cameras
I think I already made my point three times about that test. With fw 1.4, the writing speed to sd increased considerably. Based on that test, I had bought the samsung pro card. After fw 1.4 came, I bought the lexar 2000x and wow, such a huge difference. I sent several emails to that site asking to retest, but they replied they hadn't got the nx1 anymore. So no new test. But they should at least add a disclaimer about reports from users about 4x higher speeds with fw 1.4. -
Tommix, as far as I can tell, the process is all but simple. As pavel wrote, patience is a keyword here. Otto created some nice scripts. Vasile seems to have upped the bitrate. Chant is delving into the main code. Three must be the conditions joined by a logical AND: - Interested in nx system - higly knowledgeable in sw development and Linux system - have time Not that easy to return a TRUE value ;-)
-
SD-Card Benchmark (with Samsung tool in DEV mode)
Marco Tecno replied to derderimmermuedeist's topic in Cameras
Actually you didn't get my point. I own both samsung pro and lexar 2000x cards. The real world difference (not benchmark) is of the 4-5x magnitude. Hence there is something wrong with that benchmark. -
SD-Card Benchmark (with Samsung tool in DEV mode)
Marco Tecno replied to derderimmermuedeist's topic in Cameras
Tommix, no way. That's why I say there is something wrong in the bench. Going from a samsung pro to a lexar 2000x is a huge jump on my camera. You go from about 1fps when buffer full to 5/6fps. And the clearing time is about 3-4x better. -
I asked him not to disappear, since in the past we had some users who managed hacking the nx system and shortly after went MIA.