sandro
-
Posts
693 -
Joined
-
Last visited
Content Type
Profiles
Forums
Articles
Posts posted by sandro
-
-
33 minutes ago, MountneerMan said:
Ok this is my theory and it is jut a theory at this point because I cannot find any information on the internet to confirm OR disprove it.
Because f-stop = focal length / diameter of the entrance pupil as you zoom the lens the lens needs to change the iris to keep the same f-stop.
Quote from the Wikipedia page on f-number https://en.wikipedia.org/wiki/F-number
So assuming you have a 100-200 f4 zoom lens (fully electronic) and you were at 100mm and f/4 then you quickly zoomed to 200mm. the lens would need to detect the new focal length then calculate the diameter of the entrance pupil then adjust accordingly. The few milisec delay is just the camera calculating the then adjusting according.
Can someone please confirm or disprove this theory because I am starting to get frustrated with the lack of information on the internet about this.
That would explain it.
-
Still doesn't explain why on manual the shift is for 1 second only.
-
18 hours ago, tugela said:
If everything is on manual, including gain, then zooming out will increase the amount of light on the sensor and increase overall exposure. Converse for zooming in. The video clip apparently has gain on auto, but the response to changes in exposure is not instantaneous, which is why the exposure changes like that. If he zoomed more slowly then you would not notice it.
But shouldn't that increase stay the same? Instead with the Samsung lens I used it goes back to "normal" after a moment.
-
Does.it really prevent current hacking?
-
-
3 minutes ago, kidzrevil said:
@sandro nx1 video is terrible in low light ! I hate going above 800
yeah but that it's supposed to be without NR... I'm starting to wonder if Samsung used that aggressive NR for a good reason...
-
50 minutes ago, SMGJohn said:
From what I have read early on was that 4k is upscaled 2160p and cropped in-camera, this was made from the conclusion that it lacks sharpness and had more aliasing than UHD.
Whether this really is the case is debatable because I have no idea how it really works, it could be true line skipping for all I know.
When I shoot 2160p or 4k in NX500 I get fine noise grain like in 1440p mode, its just smaller grain which is expected duo to higher resolution. Why is this? I use Kino's Mod Pack 1.62m, was this a standard feature on the NX500 before the hack? I am too lazy to uninstall and reinstall on my vacation just to check it, can any NX500 owner confirm with me if this was the case? 1080p mode has the ugly NR as we know it, hurts me everytime i look at that blue swirling death atoms on high ISO videos shot from NX1.
LEFT - 1440p mode from NX500 (140mbps) | RIGHT - 2160p mode from NX500 (160mbps)
What ISO? But seriously if this is below 6400 the NX1 really sucks in low light... nothing NR off can solve...
-
1 hour ago, gsenroc said:
Electronic aperture is not the cause, optical design is. Try EF lenses with constant aperture.
So its the lenses.
-
3 minutes ago, tugela said:
Kidzrevil mentioned seeing similar things with his Sigma zoom, which presumably would not be operated electronically.
I think it is just a function of the rapid change in the field of view (with consequent exposure change) and the time it takes for metering to adjust to gain to a different level to compensate.
IMO you will see it with any camera if gain correction is not instantaneous.
how can it be if it happens even when everything is all manual? If doesn't have to adjust to anything.
-
I believe it's a design of these electronically controlled lenses. It's disappointing it happens with a S lens...
-
I don't think hdds configuration makes that much of difference. Yes you can get less jumps between different clips if positioned distant on the HDD because it needs to seek on the platters. SSD removes that problem. BUT I don't think you will humanly notice any difference. For these applications where the CPU is always the limit hdd speed is negligible considerand that most reach 100MB/s at least.
-
My 18-200 does the same, if I set the aperture to 6.3 while zooming out it sightly opens and then closes down to 6.3. Annoying and not acceptable for "video" lens.
-
http://***URL removed***/forums/thread/4001686 330Mbit hack. I don't see any difference for 3200 though... (NR wise)
-
@kidzrevil I'm confused, if I stay at 150-160Mbit do these problem occur also? I mean freezes and stuff.
-
11 hours ago, shanebrutal said:
Yeah try DIS with OIS on, it's very smooth on the 16-50 S. Tripod like if you hold still. I personally wouldnt use DIS without ois turned on if the lens supports it.
But I'm a fan of DIS turned on with vintage adapted glass too.
How much is the focal length multiplier with DIS on?
I wish the higher bitrate would also allow a better low light performance. At least a ISO3200 without those horrible lines.
-
DIS can't be used on non Samsung lenses right?
-
Does DIS work OK with higher bitrate or it doesn't affect it in any way?
-
16 minutes ago, outerbeat said:
Well, IDK yet dependences between real record algorithms in NX1 and any of tested parameters, maybe I'm totally wrong, but for me it seems that random 4kb write tests shows far more consistent results to actual allowed bitrate, than vendors "write speed ABC Mb\s". It is just theory proof test, I might be totally wrong on that.
But again, vendors write speeds is total BS for NX1 4K record imho. Maybe it is a NX1 thing, yes, but Blackmagic's document actually provides some proof of that not only NX1 have this troubles.I see, you could try with this tool https://www.attotech.com/disk-benchmark/ and check if you see a pattern. There must be a block size used while writing files that could explain the "slow" speeds.
-
SD card speed depends a lot on the controlled used by the reader/host. And of course the kind of test you're doing.
@outerbeat when you record video the camera will use sequential writes not random 4K why are you interested in that?
-
Now I have another question. Is there a way to develop smartphone apps or web apps for the NX1? Or directly "apps" on the NX1?
I want to develop one. -
6 minutes ago, Pavel Mašek said:
We do not now how it it would look like in 4K (according some comparsion 2.5K look a little worse than 1080p). But I still think that NX1 has similar low light performance as A6300 in photo mode (or even better), so problem is in processsing of video (NR, codec?, I do not know...) which is not so good.
Not sure what to say at this point.
-
19 hours ago, Pavel Mašek said:
Kino Seed just posted video of 2.5K video without noise reduction - http://***URL removed***/forums/thread/3990893?page=2#forum-post-57632926
Noise grain is very visible here, but we should realize that it is quite high ISO (6400). I think it would be more usable and subtle in 4K.
It is proof that noise reduction is not built-in the codec - really great news I think.
Vasile/Otto - is there any chance if 4K could be inserted into slot for VGA? Maybe then 4K would be without NR.
I am wondering how it would affect 1080p modes which suffer with high NR above low ISO.
I'm not sure how he turned NR off on the nx500? I'm downloading the video file now...
UPDATE: I've seen the video, well the noise is definitely there but it's there in a way that makes me think it's the sensor design that it's not good in low light. Agreed that it's 6400 but isn't the a6300 clear at that ISO? Even the Nikons at 6400 don't look that bad.
-
how do you use them with the nx1 without slog?
-
Nx1 16-50 2-2.8 exposure change while zooming.
In: Cameras
Posted
I never tried zooming with a non Samsung lens since I don't have one but I don't see why it would do it...