Sony 4K Handycam FDR-AX100 thread - Page 72 - AVS Forum

AVS Forum > Other Areas of Interest > Camcorders > Sony 4K Handycam FDR-AX100 thread

Camcorders

bsprague's Avatar bsprague
10:10 AM Liked: 108
post #2131 of 2816
07-14-2014 | Posts: 3,063
Joined: Dec 2008
Quote:
Originally Posted by Ken Ross View Post
....Instead we shoot videos of our travels as well as other 'boring' things ......
I think it would be more accurate to write that "We post videos and video clips from the newest of video capable cameras."

Unless you have a camera that is one of the newest releases, there is little reason to post anything one shoots. I don't mean that to be a bad thing. There are short threads on miscellaneous topics. But, the long threads that are heavily read are about newly released equipment.
Ken Ross's Avatar Ken Ross
10:22 AM Liked: 1663
post #2132 of 2816
07-14-2014 | Posts: 24,862
Joined: Nov 2000
Bill, I think what happens is that for the early life of a thread, we do talk about equipment and do post videos of the camera's capabilities. But as the thread matures, we do tend to post more videos of our travels as well as other things we think others may find interesting. I find zoos fascinating and have posted a few of those videos. Others may find them less fascinating. Mark, due to his business travels, posts the most (I'd love to have his frequent flyer points!) of his travel videos. I always enjoy watching those. Yes, they're not "Indie films", and yes they're not scripted and yes, they're not graded, but who cares? I find them every bit as interesting and every bit as technically competent as a graded video (which more often than not are hard to watch since most people have no clue about grading).

We'll be headed to London & Paris in September and I'll probably post a 'boring' video of that too. I guess we sometimes make the assumption that some of these video might interest some people, but that's obviously not always the case. But at least the videos offer some the opportunity to watch or not watch.

What Mark and I have commented on from time to time, are the few posters that simply criticize videos without offering their own examples of their work. By posting a video, you're laying it all out there. It's very easy to criticize any video, but it's harder to offer constructive suggestions and still harder to offer your own work for others to critique (Bill, you know I'm not speaking of you).
markr041's Avatar markr041
10:55 AM Liked: 145
post #2133 of 2816
07-14-2014 | Posts: 3,333
Joined: Jun 2005
Quote:
Originally Posted by avdmem View Post
I have to agree with you actually, but not really for the poor "content" (I couldn't care less of the "content" in a forum where we're discussing the gear itself). Why I agree with you? Because I can't find pretty much anything showing the real weekness of the AX100 in 4K which is the degradation (massive) of the footage the moment you start panning. In fact all I see (and that goes for AX1 as well) is a static take, mostly flowers and gardens (and here you are right again) but the "panning" thing? No one pans anymore? It looks like we (all) try to hide something, and we (all) let our enthusiasm for a splendid new toy prevail over the facts. Myself included. So your post was indeed a wake-up call.

Panning has limited usefulness, even in cinema. There was a baseball video posted, with plenty of panning:
https://vimeo.com/91928654
Download the 4K original, and if you do and you comment, tell us exactly how you viewed the video - in 4K? what viewing device? Turns out it is viewing devices that are having trouble with motion in 4K, lowering the resolution whenever there is movement.
And if you have evidence of this "massive" degradation, why are you not showing us instead of complaining others are not?
aholck
11:19 AM Liked: 0
post #2134 of 2816
07-14-2014 | Posts: 23
Joined: Jul 2014
Quote:
Originally Posted by avdmem View Post
I have to agree with you actually, but not really for the poor "content" (I couldn't care less of the "content" in a forum where we're discussing the gear itself). Why I agree with you? Because I can't find pretty much anything showing the real weekness of the AX100 in 4K which is the degradation (massive) of the footage the moment you start panning. In fact all I see (and that goes for AX1 as well) is a static take, mostly flowers and gardens (and here you are right again) but the "panning" thing? No one pans anymore? It looks like we (all) try to hide something, and we (all) let our enthusiasm for a splendid new toy prevail over the facts. Myself included. So your post was indeed a wake-up call.
I am happy you take your time trying to understand my point. I'm known for expressing in an aggressive way w/o actually being aggressive. I believe I'm a bit too much "to the point" if you understand.
I strongly disagree w/ 100% accuracy of this device even if it's indeed nifty as it is. I bought it because of the content I saw but in mind I could make it better.
I see it needs as much work as previously mentioned BMPCC out of a grading perspective. Or, rather, correction. The BMPCC indeed needs a alot of changes from it's flat default image while this camera
rather needs to be slightly flatted.

There are absolutely NO intention to be negative regarding editing skills etc etc etc. I understand - we all have different levels and all need to practice. All of us.
But static shots, always beautiful nature doesn't show this camera. The level of how this camera is set from factory is indeed represented in a garden because this is what attracts people.
But it's not really good by default in representation of environment apart from gardens. Maybe my tolerance level is lower than the average. Doesn't mean I'm right but allergy is allergy (=.
And there are, like I already mentioned a lot more information given from the codec than the default grade actually shows.

Yes, this camera indeed has issues such as the rolling shutter problems. I don't suffer that much of it. I always run in a 180 degree shutter angle and the motion blur takes really good care of it. Avoiding too much panning while in tele range.
Auto exposures sets this camera a sunny day to f/8 - -3 gain and 1/200 shutter + 1ND f.ex. This is not a video/film exposure. The camera has it's native ASA sensitivity where the latitude of the dynamic range is at its best.
To achieve the perfect exposure, you need to bring the camera as bright as possible w/o clipping and in post turn it down. This method is as old as film and is actually the way to work with film/video. It doesn't work for people who just want clips and doesn't care.

You can read more here: http://en.wikipedia.org/wiki/Exposing_to_the_right

It's important to point out the weak parts of the camera, which is - according to me, color representation, sharpness (it's over sharp - mind the difference between detail and sharpness), rolling shutter and so on and figure out ways to make it better.
I gave one tip of how I deal with the rolling shutter. I also mentioned earlier to bring more details out of the image which is not even seen default. Also, to put just a few minutes or maybe make a preset of a slight grade (which I mentioned one)
would make it look different and, according to me, more natural and accurate.

The problem in this thread is that once anyone has a view a war almost starts and fights because of "I always do right". I'm happy to know if I do wrong and want to get better me too.

Oh well, I can write for hours...
leamas's Avatar leamas
11:32 AM Liked: 15
post #2135 of 2816
07-14-2014 | Posts: 113
Joined: Dec 2006
Guys, one observation, this may be helpful. When doing various tests in photo mode, I observed one fact. The camera behaves quite bad as apertures bigger than 5.6 (F/8 or F/11 by ex). Pictures done is this mode have a lot worse quality. This happens on any lens I understand, when having bigger apertures than some value (depends on the lens type - the phenomenon is called diffraction if I am not wrong). Ok, then I found out that the camera tries to always use <=5.6, and if not possible because of big light, it warns you to use ND1 or ND2. I think that is the reason, that it knows that the result will be worse otherwise. In video mode, this can be harder to observe since the camera has a much bigger sensor than necessary (20 instead of 8 MP) so when scaling down the problem also scales down.
I wait also for your observations / feedback on this.
markr041's Avatar markr041
11:39 AM Liked: 145
post #2136 of 2816
07-14-2014 | Posts: 3,333
Joined: Jun 2005
Quote:
Originally Posted by avdmem View Post
so you say one thing (pans are not important) and then the exact opposite (plenty of panning in your video) and all in the same sentence?
OK , I see that panning is either useless or fine with your AX100. It's not with mine, and I shoot sports too but where panning is practically continuous. I'm leaving this forum now.

Just to spell it out to you what was said: panning has its limited place. And sports is certainly one. You say you shoot sports, yet you complain that no one posts video showing poor panning and then you do not post your results either. You did not accept the invitation to look at the video that was linked in response to your uninformed complaint that no one posts panning videos, nor did you tell us how you view videos so we could perhaps evaluate what the problem is, even though you were asked politely.


Just one last point (for you): in sports you pan to follow the action. In most cases this means the subject is actually less blurred than the background and less subject to rs; it is the background that gets blurred and perhaps distorted. But this little detracts from the video.
markr041's Avatar markr041
11:43 AM Liked: 145
post #2137 of 2816
07-14-2014 | Posts: 3,333
Joined: Jun 2005
Quote:
Originally Posted by leamas View Post
Guys, one observation, this may be helpful. When doing various tests in photo mode, I observed one fact. The camera behaves quite bad as apertures bigger than 5.6 (F/8 or F/11 by ex). Pictures done is this mode have a lot worse quality. This happens on any lens I understand, when having bigger apertures than some value (depends on the lens type - the phenomenon is called diffraction if I am not wrong). Ok, then I found out that the camera tries to always use <=5.6, and if not possible because of big light, it warns you to use ND1 or ND2. I think that is the reason, that it knows that the result will be worse otherwise. In video mode, this can be harder to observe since the camera has a much bigger sensor than necessary (20 instead of 8 MP) so when scaling down the problem also scales down.
I wait also for your observations / feedback on this.

Yes, I noticed too that the "suggestions" for ND filters seems to want to keep apertures wider than f8. I follow the advice! Thank goodness for those built-in ND filters. This is another reason not to use ND auto mode, as ND filters above 1 are not used (not 2 or 3), and in manual mode, the camera instructs you what to use.
markr041's Avatar markr041
11:59 AM Liked: 145
post #2138 of 2816
07-14-2014 | Posts: 3,333
Joined: Jun 2005
Quote:
Originally Posted by aholck View Post
I am happy you take your time trying to understand my point. I'm known for expressing in an aggressive way w/o actually being aggressive. I believe I'm a bit too much "to the point" if you understand.
I strongly disagree w/ 100% accuracy of this device even if it's indeed nifty as it is. I bought it because of the content I saw but in mind I could make it better.
I see it needs as much work as previously mentioned BMPCC out of a grading perspective. Or, rather, correction. The BMPCC indeed needs a alot of changes from it's flat default image while this camera
rather needs to be slightly flatted.

There are absolutely NO intention to be negative regarding editing skills etc etc etc. I understand - we all have different levels and all need to practice. All of us.
But static shots, always beautiful nature doesn't show this camera. The level of how this camera is set from factory is indeed represented in a garden because this is what attracts people.
But it's not really good by default in representation of environment apart from gardens. Maybe my tolerance level is lower than the average. Doesn't mean I'm right but allergy is allergy (=.
And there are, like I already mentioned a lot more information given from the codec than the default grade actually shows.

Yes, this camera indeed has issues such as the rolling shutter problems. I don't suffer that much of it. I always run in a 180 degree shutter angle and the motion blur takes really good care of it. Avoiding too much panning while in tele range.
Auto exposures sets this camera a sunny day to f/8 - -3 gain and 1/200 shutter + 1ND f.ex. This is not a video/film exposure. The camera has it's native ASA sensitivity where the latitude of the dynamic range is at its best.
To achieve the perfect exposure, you need to bring the camera as bright as possible w/o clipping and in post turn it down. This method is as old as film and is actually the way to work with film/video. It doesn't work for people who just want clips and doesn't care.

You can read more here: http://en.wikipedia.org/wiki/Exposing_to_the_right

It's important to point out the weak parts of the camera, which is - according to me, color representation, sharpness (it's over sharp - mind the difference between detail and sharpness), rolling shutter and so on and figure out ways to make it better.
I gave one tip of how I deal with the rolling shutter. I also mentioned earlier to bring more details out of the image which is not even seen default. Also, to put just a few minutes or maybe make a preset of a slight grade (which I mentioned one)
would make it look different and, according to me, more natural and accurate.

The problem in this thread is that once anyone has a view a war almost starts and fights because of "I always do right". I'm happy to know if I do wrong and want to get better me too.

Oh well, I can write for hours...

Rather than writing "for hours," why not just make and post a video of yours demonstrating how much better your editing tweaks and ETR technique make the video? Armchair comments become ineffective, fast.


Indeed, I welcome you to take any of my AX100 videos (the original 4K versions are all available for download) and show us how much better they can be (most of the sunlit clips were exposed up to clipping, so it's all what you do in the editor). This has happened before, where posters re-graded some of my videos with my permission. We are all eager, I am sure, to see what you can do.
Ken Ross's Avatar Ken Ross
02:46 PM Liked: 1663
post #2139 of 2816
07-14-2014 | Posts: 24,862
Joined: Nov 2000
Quote:
Originally Posted by avdmem View Post
so you say one thing (pans are not important) and then the exact opposite (plenty of panning in your video) and all in the same sentence?
OK , I see that panning is either useless or fine with your AX100. It's not with mine, and I shoot sports too but where panning is practically continuous. I'm leaving this forum now.
In the last 2 pages of this thread, we have a few really amazing posts.

So here you say owners are 'hiding' something by not panning and when Mark shows you a video where there was extensive panning, you run and hide. The fact is that people over use panning as they do zooming and that's why many choose to use both tools in a limited fashion. Of course you can pan with the AX100 and of course it looks fine. Would I do a whip pan with the AX100? No. Would I do a whip pan with any video camera? No. Unless there was some specialized need for it, I just wouldn't do it.

So your intent seemed to 'prove' that you couldn't pan effectively with the AX100 and when it was shown to you that was not the case and nobody was 'hiding' anything, you chose to abandon ship.

As I said, we've had a few odd posts in the last couple of pages. Interestingly it seems to come from the brand new posters. Interesting.
James Jimmy's Avatar James Jimmy
05:03 PM Liked: 23
post #2140 of 2816
07-14-2014 | Posts: 47
Joined: Mar 2014
Quote:
Originally Posted by leamas View Post
Guys, one observation, this may be helpful. When doing various tests in photo mode, I observed one fact. The camera behaves quite bad as apertures bigger than 5.6 (F/8 or F/11 by ex). Pictures done is this mode have a lot worse quality. This happens on any lens I understand, when having bigger apertures than some value (depends on the lens type - the phenomenon is called diffraction if I am not wrong). Ok, then I found out that the camera tries to always use <=5.6, and if not possible because of big light, it warns you to use ND1 or ND2. I think that is the reason, that it knows that the result will be worse otherwise. In video mode, this can be harder to observe since the camera has a much bigger sensor than necessary (20 instead of 8 MP) so when scaling down the problem also scales down.
I wait also for your observations / feedback on this.
I have noticed that as well. I think that this is caused by the lens and sensor system being diffraction-limited: http://en.wikipedia.org/wiki/Diffraction-limited_system. My videos are best when I set the aperture between 4 and 5.6. I always set my shutter to 1/60 of a second. Then set the ND manually and tune the gain.

Here's a recent video from an Alaskan cruise (AX100 in 4k):

--
AX-100 gallery: http://gg.gg/ax100
aholck
05:16 PM Liked: 0
post #2141 of 2816
07-14-2014 | Posts: 23
Joined: Jul 2014
Quote:
Originally Posted by markr041 View Post
Rather than writing "for hours," why not just make and post a video of yours demonstrating how much better your editing tweaks and ETR technique make the video? Armchair comments become ineffective, fast.


Indeed, I welcome you to take any of my AX100 videos (the original 4K versions are all available for download) and show us how much better they can be (most of the sunlit clips were exposed up to clipping, so it's all what you do in the editor). This has happened before, where posters re-graded some of my videos with my permission. We are all eager, I am sure, to see what you can do.
Hi!

I don't see any problems in this. But, as stated earlier and since I talk about delivery from camera, I need to have the files out of camera in order to do anything since the delivery codec might be stripping out all other stuff. A delivery codec such as H.264 strips out all stuff in dark areas since the idea is to attract the eyes which is mainly in contrasty and brighter areas thus saving space by removing stuff where it's not needed.

So, if you can provide me with a clip straight out from camera, I would be happy to make a video to show what I believe gives a richer picture.

Note. It is impossible for me to enhance the picture in accuracy since I have NO idea about what you actually saw. And what you saw is different from what I would have seen. I can apply my generic stuff I came to learn about my camera. But not correct and say "this is what you saw!!!"

I could take a snip from something you like, straight from camera. If you deleted, I don't mind until you go to the garden again.
Take a 10 sec shot from a leaf in your preferred exposure with a lot of details and we'll see what I can take out of it.

On the other hand, I can take some garden shots myself exposed of cameras own preference, my own preference, graded and ungraded and make a comparison video you'll be able to download to check out on your own tv-set.

You choose. What I hope is for you to see the difference and IF it's beneficial for you I would be happy to, unless you know/guess it yourself, to apply it in future videos.

Look, Mark - I have no problems with your videos. I just have a problem with you having a nice camera and only walk in garden and present it
as this camera would be the best thing since sliced bread. Or maybe the audience reaction. I don't know.
After I spent some time to review my initial complaints or whatever we would call it, I checked your videos of travelling and these make a lot
more sense. Not really any scripts etc but you actually do something many people are scared of - getting close-up in person shots.

@Ken - I believe your kindergarden level made you ignore me which is totally fine - I'm confident you won't benefit me with camera related stuff. I do believe you have a good expertise of televisions etc but it's OK. I wanted to just say that I have no relation to other "newbies" on the forum.
It's what we could call a coincidence.

I'm sure, if you read my posts and the others you would see I'm not natively speaking english while the others seem more "english"..

//
markr041's Avatar markr041
05:44 PM Liked: 145
post #2142 of 2816
07-14-2014 | Posts: 3,333
Joined: Jun 2005
aholck said: "I just have a problem with you having a nice camera and only walk in garden and present it as this camera would be the best thing since sliced bread.:


1. I did not say anything about how wonderful the camera is; I just posted a video that I thought people would like (look at my post). They did, and it drove you nuts. Right in your first post.


2. Your statement that I "only walk in garden" is ludicrously inaccurate. The garden is a test bed for the videos I care about - those that report on the cities I visit. I have posted videos in this forum of: Cambridge UK and Cambridge MA, Stockholm, Geneva, Palm Beach, Hong Kong, New Haven, Chicago, Beijing, Alicante, Boston, Shanghai, Richmond, Charlottesville, St. Louis, New York City, Nanjing, Berlin, London UK and New London CT, San Antonio, Dhaka, Istanbul, Philadelphia, Aix-en-Provence, Berkeley, Barcelona. I may have missed a few. I will be posting videos soon from Hangzhou and Guangzhou. Oh, a number of videos were graded, in Resolve.


You don't have to apologize; I understand that you are a newcomer here and have no idea who you are posting to. It is an example of how newcomers should always be humble (at first). We all appreciate that you have some relevant knowledge about cameras and editing.


Now, as to the posted videos. This one - the very first video I took with the AX100 - was not re-rendered at all. The clips are straight from the camera, only merged: https://vimeo.com/89673510


There are a variety of scenes - it is not just a garden video - as it was shot just to test out what the camera could do. We look forward to what you can do with it. I enjoyed grading the RAW videos from the BMPCC and even the Magic Lantern RAW videos from the EOS M. I would actually like RAW 4K.
Ken Ross's Avatar Ken Ross
07:56 PM Liked: 1663
post #2143 of 2816
07-14-2014 | Posts: 24,862
Joined: Nov 2000
Quote:
Originally Posted by aholck View Post
@Ken - I believe your kindergarden level made you ignore me which is totally fine - I'm confident you won't benefit me with camera related stuff.
My 'kindergarden level'? It's precisely this attitude and approach you've used in almost every post that makes me discount most of what you say. You don't have the slightest idea of the meaning of etiquette, no matter what country you come from. This is not a language issue, it's a lack of respect. You showed no respect to anyone from post one. In return, do you expect posters here, who have been valuable contributors for years, to respect you?

And yes, I'm sure there's nothing you could learn from me for 'camera related stuff'. With the endless depth of your camera related knowledge being so apparent, how could you learn anything from me or anyone here?
David Harry's Avatar David Harry
08:30 PM Liked: 18
post #2144 of 2816
07-14-2014 | Posts: 129
Joined: May 2014
Quote:
Originally Posted by aholck View Post
I am happy you take your time trying to understand my point. I'm known for expressing in an aggressive way w/o actually being aggressive. I believe I'm a bit too much "to the point" if you understand.
I strongly disagree w/ 100% accuracy of this device even if it's indeed nifty as it is. I bought it because of the content I saw but in mind I could make it better.
I see it needs as much work as previously mentioned BMPCC out of a grading perspective. Or, rather, correction. The BMPCC indeed needs a alot of changes from it's flat default image while this camera
rather needs to be slightly flatted.

There are absolutely NO intention to be negative regarding editing skills etc etc etc. I understand - we all have different levels and all need to practice. All of us.
But static shots, always beautiful nature doesn't show this camera. The level of how this camera is set from factory is indeed represented in a garden because this is what attracts people.
But it's not really good by default in representation of environment apart from gardens. Maybe my tolerance level is lower than the average. Doesn't mean I'm right but allergy is allergy (=.
And there are, like I already mentioned a lot more information given from the codec than the default grade actually shows.

Yes, this camera indeed has issues such as the rolling shutter problems. I don't suffer that much of it. I always run in a 180 degree shutter angle and the motion blur takes really good care of it. Avoiding too much panning while in tele range.
Auto exposures sets this camera a sunny day to f/8 - -3 gain and 1/200 shutter + 1ND f.ex. This is not a video/film exposure. The camera has it's native ASA sensitivity where the latitude of the dynamic range is at its best.
To achieve the perfect exposure, you need to bring the camera as bright as possible w/o clipping and in post turn it down. This method is as old as film and is actually the way to work with film/video. It doesn't work for people who just want clips and doesn't care.

You can read more here: http://en.wikipedia.org/wiki/Exposing_to_the_right

It's important to point out the weak parts of the camera, which is - according to me, color representation, sharpness (it's over sharp - mind the difference between detail and sharpness), rolling shutter and so on and figure out ways to make it better.
I gave one tip of how I deal with the rolling shutter. I also mentioned earlier to bring more details out of the image which is not even seen default. Also, to put just a few minutes or maybe make a preset of a slight grade (which I mentioned one)
would make it look different and, according to me, more natural and accurate.

The problem in this thread is that once anyone has a view a war almost starts and fights because of "I always do right". I'm happy to know if I do wrong and want to get better me too.

Oh well, I can write for hours...
Hi aholck.

I would just like to agree with your assessment of going as close to maximum exposure. Although it can be a little hit or miss with the AX100, due to the screen in bright light etc. If you can get it just under the point of clipping/burning, then bringing it back down in post does, in my opinion, creates a great picture. I use Edius and its YUV filter for doing this.

The really odd thing with the AX100, is that its dynamic range is absolutely bizarre. I have never seen such detail in an 8Bit 4:2:0 picture before, and more so with an Inter-frame codec. Other 8Bit cameras clip way earlier than the 100, it's even better than the EX3. You can really tell on clouds. Where the picture on first glance may look burnt out. If you pull back the HF and crush the bottom a bit, there is a load of detail in the areas that would otherwise look flat and blown. But you do have to be careful not to wreck the picture.

This of course is no help to anyone who is shooting for fun and just wants to truncate and stitch an edit together for a fast turnaround. But for anyone wanting to up their game a bit, or for pro delivery, it is well worth the effort. You can get some real bite into the picture. You can even use the clips with Da Vinci Resolve, and push some heavyish looks into them. They won't compete with 10bit+ or raw! but are flexible for what they are.

BTW, just some friendly advice. Maybe soften what you are saying, but not water it down. Some of your points are quite interesting, and may help others. But coming on strong, may go against you. Especially with the likes of Mark, who's productions are quite excellent. In fact, I first touched base with Mark outside of this forum to get advice on how he stitched his native files without a transcode. I think it was with his first 'indoor and outdoor' AX100 piece. Which was one of the AX videos that finally convinced me of just how good the 100 is. He has more practical experience with this camera than most, and doesn't always shoot gardens and flowers

Cheers,
Dave.
aholck
10:49 PM Liked: 0
post #2145 of 2816
07-14-2014 | Posts: 23
Joined: Jul 2014
Quote:
Originally Posted by Ken Ross View Post
My 'kindergarden level'? It's precisely this attitude and approach you've used in almost every post that makes me discount most of what you say. You don't have the slightest idea of the meaning of etiquette, no matter what country you come from. This is not a language issue, it's a lack of respect. You showed no respect to anyone from post one. In return, do you expect posters here, who have been valuable contributors for years, to respect you?

And yes, I'm sure there's nothing you could learn from me for 'camera related stuff'. With the endless depth of your camera related knowledge being so apparent, how could you learn anything from me or anyone here?
Oh well, you seem not to get it. I doesn't matter. It's OK. You know what? Let's not get too much into it. I will win in my end anyhow and you will win in your end. Fair enough? I don't lack anything but I might agree I could have at least waited until 2:nd post.

// A
aholck
11:06 PM Liked: 0
post #2146 of 2816
07-14-2014 | Posts: 23
Joined: Jul 2014
Quote:
Originally Posted by markr041 View Post
aholck said: "I just have a problem with you having a nice camera and only walk in garden and present it as this camera would be the best thing since sliced bread.:


1. I did not say anything about how wonderful the camera is; I just posted a video that I thought people would like (look at my post). They did, and it drove you nuts. Right in your first post.


2. Your statement that I "only walk in garden" is ludicrously inaccurate. The garden is a test bed for the videos I care about - those that report on the cities I visit. I have posted videos in this forum of: Cambridge UK and Cambridge MA, Stockholm, Geneva, Palm Beach, Hong Kong, New Haven, Chicago, Beijing, Alicante, Boston, Shanghai, Richmond, Charlottesville, St. Louis, New York City, Nanjing, Berlin, London UK and New London CT, San Antonio, Dhaka, Istanbul, Philadelphia, Aix-en-Provence, Berkeley, Barcelona. I may have missed a few. I will be posting videos soon from Hangzhou and Guangzhou. Oh, a number of videos were graded, in Resolve.


You don't have to apologize; I understand that you are a newcomer here and have no idea who you are posting to. It is an example of how newcomers should always be humble (at first). We all appreciate that you have some relevant knowledge about cameras and editing.


Now, as to the posted videos. This one - the very first video I took with the AX100 - was not re-rendered at all. The clips are straight from the camera, only merged: https://vimeo.com/89673510


There are a variety of scenes - it is not just a garden video - as it was shot just to test out what the camera could do. We look forward to what you can do with it. I enjoyed grading the RAW videos from the BMPCC and even the Magic Lantern RAW videos from the EOS M. I would actually like RAW 4K.
I make the difference between people but the repetitive stuff and lack of actually assigning a name to personality, video taste etc etc and as you said, being new could make you mix it up at occasions. I'm humble. I just had a problem with one person and the shockwave radius got slightly wider than expected.

I will check out the video from your suggestion to see if it will qualify codec wise. Now, my wife returns after a 2 month trip so I might spend some days with her now with a bit of prio. I won't forget this tho.

I like, when it's personal videos to sit and take my time to create the feeling in videos. It's a bit like, if you really like it, relaxing to sit and take your time to try and bring your emotions and mood into the finished video. What would a rainy video look like if the hat of the lady would be happily red?

RAW is fantastic but comes to a price. Size is one negative factor and the workflow is different. I've been working with BMD, Red and Arri RAW and this is time consuming rendering wise. But the fact you can basically do whatever you want w/o any direct quality loss is truly brining something extra.
4K RAW is, even on a sturdy new PC, slow.

I love working with efficient codecs. Prores 422 HQ is a great codec but not really efficient. I'm OK with large files and so on but 4K 422 HQ would be eating up memory cards in a sec. AX100 codec is efficient. I like it a lot. It's preferred to transcode it into Prores 422 to make it lighter on your PC.

Resolve, as you mentioned, is a good piece of software. Since you already bought it, you should use it.
aholck
11:21 PM Liked: 0
post #2147 of 2816
07-14-2014 | Posts: 23
Joined: Jul 2014
Quote:
Originally Posted by David Harry View Post
Hi aholck.

I would just like to agree with your assessment of going as close to maximum exposure. Although it can be a little hit or miss with the AX100, due to the screen in bright light etc. If you can get it just under the point of clipping/burning, then bringing it back down in post does, in my opinion, creates a great picture. I use Edius and its YUV filter for doing this.

The really odd thing with the AX100, is that its dynamic range is absolutely bizarre. I have never seen such detail in an 8Bit 4:2:0 picture before, and more so with an Inter-frame codec. Other 8Bit cameras clip way earlier than the 100, it's even better than the EX3. You can really tell on clouds. Where the picture on first glance may look burnt out. If you pull back the HF and crush the bottom a bit, there is a load of detail in the areas that would otherwise look flat and blown. But you do have to be careful not to wreck the picture.

This of course is no help to anyone who is shooting for fun and just wants to truncate and stitch an edit together for a fast turnaround. But for anyone wanting to up their game a bit, or for pro delivery, it is well worth the effort. You can get some real bite into the picture. You can even use the clips with Da Vinci Resolve, and push some heavyish looks into them. They won't compete with 10bit+ or raw! but are flexible for what they are.

BTW, just some friendly advice. Maybe soften what you are saying, but not water it down. Some of your points are quite interesting, and may help others. But coming on strong, may go against you. Especially with the likes of Mark, who's productions are quite excellent. In fact, I first touched base with Mark outside of this forum to get advice on how he stitched his native files without a transcode. I think it was with his first 'indoor and outdoor' AX100 piece. Which was one of the AX videos that finally convinced me of just how good the 100 is. He has more practical experience with this camera than most, and doesn't always shoot gardens and flowers

Cheers,
Dave.
I strongly dislike that AX100 totally forgets your zebra if you are waaaaay over exposed.. What I mean is that there's no zebra indication if you are like 2-3 f-stops over cutting. This has destroyed footage for me. I rather get slightly under exposed in some cases and pull the area up in post rather than loosing important stuff. I don't care about blown out skies. I just avoid filming them too much. One problem less.

If you take this camera and crop it to 100% and study the footage, you'll see it's not much better than anything else in $500 price range.
The camera skips too many lines. But, due to the higher resolution which is overfeeding our screens with details sure makes it look great.
I'm impressed with it's DR being a Handycam. When you see your footage you regret not working with a Red/Arri because this camera cuts too easy at some occations. Sure, not comparable, not same market, but you get my point. But I would say this camera gives around 8-10 stops of DR. Totally OK.

I believe the chroma subsampling could be represented on 1080p screens closer to 4:4:4 from UHD 4:2:0. I actually bought this camera to get decent 1080p footage and not 4K I film in 4K and scale it down to HD to actually get incredibly sharp 4:4:4 HD w/o any line skipping.
Sony Cinealta F900 f.ex with 444 gives amazing detail. It's only HD but the debayer as well as the lack of any skipping gives stunning results.

Regarding Mark, Im working on sorting it out. This is just a detonation shockwave spreading out a little too much

//
David Harry's Avatar David Harry
02:35 AM Liked: 18
post #2148 of 2816
07-15-2014 | Posts: 129
Joined: May 2014
Quote:
Originally Posted by aholck View Post
I strongly dislike that AX100 totally forgets your zebra if you are waaaaay over exposed.. What I mean is that there's no zebra indication if you are like 2-3 f-stops over cutting. This has destroyed footage for me. I rather get slightly under exposed in some cases and pull the area up in post rather than loosing important stuff. I don't care about blown out skies. I just avoid filming them too much. One problem less.

If you take this camera and crop it to 100% and study the footage, you'll see it's not much better than anything else in $500 price range.
The camera skips too many lines. But, due to the higher resolution which is overfeeding our screens with details sure makes it look great.
I'm impressed with it's DR being a Handycam. When you see your footage you regret not working with a Red/Arri because this camera cuts too easy at some occations. Sure, not comparable, not same market, but you get my point. But I would say this camera gives around 8-10 stops of DR. Totally OK.

I believe the chroma subsampling could be represented on 1080p screens closer to 4:4:4 from UHD 4:2:0. I actually bought this camera to get decent 1080p footage and not 4K I film in 4K and scale it down to HD to actually get incredibly sharp 4:4:4 HD w/o any line skipping.
Sony Cinealta F900 f.ex with 444 gives amazing detail. It's only HD but the debayer as well as the lack of any skipping gives stunning results.

Regarding Mark, Im working on sorting it out. This is just a detonation shockwave spreading out a little too much

//
I work in a similar way to yourself, shooting UHD for 1080. But don't get as many issues with the camera, maybe it's just a different way of operating. For whatever niggles I have with the camera, they are totally forgot with the awesome picture and detail. The comparison with scaled Red footage is quite an eye opener. It's obviously not as flexible for manipulation, but the comparison for scaled resolution is nuts.

You may want to be careful with your understand of chroma sub sampling, as in going from 4:2:0 to 4:4:4. How the pixel info is interpreted when squeezing four to one, is one thing. But you can't generate what isn't there in the first place. You can check this by editing the down scaled footage in a high bit depth timeline. You will still get a restricted colour rendition that doesn't use the full colour space. As seen by what is commonly known as banding. Plus, this camera, as well as most others out there, are working YUV and not RGB. And even if working with RGB, you need a higher bit depth at acquisition to account for the dynamic range of that colour space. I of course mean the dynamic range required in data terms as opposed to picture latitude, although one does effect the other in real world terms.

On that note, how are you getting 4:4:4 from a F900? Plus, how are you getting full resolution with no line skipping from it, as it does not output full raster anyway, so is subject to line interpolation to get full raster. It doesn't resolve anywhere near the resolution of the AX100's down scaled picture.

What are you using to down scale your footage?

Can you link to some of your videos, as I am really interested to see the results of someone using the AX100 in a similar way to myself.

Cheers,
Dave.
aholck
03:09 AM Liked: 0
post #2149 of 2816
07-15-2014 | Posts: 23
Joined: Jul 2014
Hi Mark!

Please read carefully below so we won't have any misunderstanding.

The following video will show some interesting details in your footage you didn't show
It will repeat by playing original, an applied fix and last, a graded option w/o applied fix.
Note, most playback software as well as streaming services online expects RGB while this camera records in YUV color space.

The problem is, from the ITU.709 specs an YUV decoder expects white level to be 235 at maximum. RGB decoders expect white to be 255 on 8-bit channels.
Vimeo, Youtube etc will all decode the codec as RGB footage, blowing out highlights. This is default manner of Quicktime, VLC, Youtube, Vimeo etc.

I put the output value of white at 220 in every 2nd clip to ensure a clearly visible difference.

It is important to understand that this is a very quick thing I did and I didn't split the clip to grade each clip alone.
I made a generic grade to lower the "grey" layer on top as well as removed the blue tint from the incorrect auto white balance.

The rendering engine of my NLE obviously couldn’t mix HD and UHD so my footage has a 4th of it’s quality and looks like LEGO.
So, don’t mind my quality - just see how I would prefer the coloring. I could fix it but heey - this is not a quality thingie..
Only for original => Fix => color grade (which actually also changed due to the lower res).

Some of my grades ARE too dark and I admit that it could be much better. But, like I said, I made one trying to match all
With the energy I could fix both the YUV highlight clip limit, grade and so on. My clips were exported in 1080p thus slightly lower res.
This is because I didn't insert the license USB stick to unlock 4K in davinci resolve.
Only to show color difference.

Now, please download the file on the link and hopefully your system will be able to show you what I did.

The screen used in this case is calibrated 2 month back. My two presets are still photo and other is for video editing.
The video calibration is based on REC.709/BT.709 standard from ITU.R.

http://en.wikipedia.org/wiki/Rec._709

Now, for the clip.

http://213.115.239.33/mark/mark.mp4 (800MB+)

One of the most visible differences are at around 1:09. Look at the plastic bag as well as on the sign to the left once they pass it.
In 2nd part of the same clip, you can see more details in the bag as well as the sign got letters written

Same in the first clip of the clouds.

One clip you see a colourfully dressed woman walking with a stroller. The white car passing on the white lines as well as the white lines do
have more white information now.

At around 2:35 you’ll see the people walking towards you. Notice in the 2nd clip how details came out from the blown out part further back on the road.

Further on, the bricks on the other stroller clip where the kid jumps down are blown out but I managed to at least recover the edges slightly.
If you check on the flower in the end, I’m sure you’ll see some more detail in the whites of the flower.

These are details and people often complain about these stuff on this particular camera. This is because of what I wrote above and it’s very easy
to correct to achieve details in the white areas.

Maybe I’m too sensitive but I want 100% advantage of my camera.
aholck
03:23 AM Liked: 0
post #2150 of 2816
07-15-2014 | Posts: 23
Joined: Jul 2014
Quote:
Originally Posted by David Harry View Post
I work in a similar way to yourself, shooting UHD for 1080. But don't get as many issues with the camera, maybe it's just a different way of operating. For whatever niggles I have with the camera, they are totally forgot with the awesome picture and detail. The comparison with scaled Red footage is quite an eye opener. It's obviously not as flexible for manipulation, but the comparison for scaled resolution is nuts.

You may want to be careful with your understand of chroma sub sampling, as in going from 4:2:0 to 4:4:4. How the pixel info is interpreted when squeezing four to one, is one thing. But you can't generate what isn't there in the first place. You can check this by editing the down scaled footage in a high bit depth timeline. You will still get a restricted colour rendition that doesn't use the full colour space. As seen by what is commonly known as banding. Plus, this camera, as well as most others out there, are working YUV and not RGB. And even if working with RGB, you need a higher bit depth at acquisition to account for the dynamic range of that colour space. I of course mean the dynamic range required in data terms as opposed to picture latitude, although one does effect the other in real world terms.

On that note, how are you getting 4:4:4 from a F900? Plus, how are you getting full resolution with no line skipping from it, as it does not output full raster anyway, so is subject to line interpolation to get full raster. It doesn't resolve anywhere near the resolution of the AX100's down scaled picture.

What are you using to down scale your footage?

Can you link to some of your videos, as I am really interested to see the results of someone using the AX100 in a similar way to myself.

Cheers,
Dave.
David,

You are absolutely right. In some ways I don't always know in what level I should talk to people w/o either bending the truth by claiming 444 of 420 scaled down but the fact remains you do get the tighter pixels, but not the coloring information as refered to subsampling.
There are two parts to it. Both the coloring information as well as, how can I call it - unique pixels? The latter is what Im talking about but in wrong terms.

I dont get 444 from F900 . I ment the F23 what I used at a few occations. F23 is not new, but 444 and B4 mount for regular broadcasting lenses.

There are too many cameras in mind and frankly, I wrote FS100 from start

Edit:

Most of my videos are private stuff such as weddings, a few real estate videos and so on.
I've been travelling for a while so I have a bunch of goodies to fix... one day.

I shot a car event recently which will, again, one day be edited and made public.

My goal is to actually start giving hints, tricks + making generic videos to show how I like stuff and hopefully help others.
Wife arrives today so now she will give me ideas. I'm planning a short movie soon but might be using Blackmagic cameras instead.

// A
Spok1701's Avatar Spok1701
03:23 AM Liked: 13
post #2151 of 2816
07-15-2014 | Posts: 32
Joined: May 2014
Well, I have been away for a bit (still no AX100 - Sony are out of stock again, but I'm hopeful) and this thread seems to have livened up.

I think there seems to a bit too much criticising (not naming any names, but you know who you are) and lack of respect of people who are here to help by sharing their videos and experiences. That's why I joined this AX100 thread. I need as much information about this camera as possible before I commit and buy it and I have had very helpful answers to my questions.

Nobody should be saying what others should be shooting with their camera. That is up to them. If you don't like what others have shot, or their style of editing, then please show us the way by posting some of your own work. Not everyone here is a professional with years of experience, but that's no reason to knock them down.

From what I have seen, this camera produces great images. Better than anything else in this price range. That, for me, is what counts. Codecs, sample rates, f-numbers, etc, etc, are not relevant to the end viewer. It is the quality of the final output that is important, and this camera has it in abundance. I can't wait to get mine.
David Harry's Avatar David Harry
03:42 AM Liked: 18
post #2152 of 2816
07-15-2014 | Posts: 129
Joined: May 2014
Quote:
Originally Posted by aholck View Post
David,

You are absolutely right. In some ways I don't always know in what level I should talk to people w/o either bending the truth by claiming 444 of 420 scaled down but the fact remains you do get the tighter pixels, but not the coloring information as refered to subsampling.
There are two parts to it. Both the coloring information as well as, how can I call it - unique pixels? The latter is what Im talking about but in wrong terms.

I dont get 444 from F900 . I ment the F23 what I used at a few occations. F23 is not new, but 444 and B4 mount for regular broadcasting lenses.

There are too many cameras in mind and frankly, I wrote FS100 from start

Edit:

Most of my videos are private stuff such as weddings, a few real estate videos and so on.
I've been travelling for a while so I have a bunch of goodies to fix... one day.

I shot a car event recently which will, again, one day be edited and made public.

My goal is to actually start giving hints, tricks + making generic videos to show how I like stuff and hopefully help others.
Wife arrives today so now she will give me ideas. I'm planning a short movie soon but might be using Blackmagic cameras instead.

// A
All easy mistakes buddy. Yes, the F23 was amazing back in the day and is still probably the best 2/3 HD camera ever made. But a real bitch for doing selfies

So anyway. What NLE do you use and what scaling method? I am always interested in these types of things when someone uses similar kit.

Any chance you can share some of your videos, and maybe some native camera files. Again, am just interested in how people work. Like Mark for instance. His productions and workflow are quite different from mine, but I have picked up some really beneficial and practical tips from his work.

Cheers,
Dave.
aholck
03:44 AM Liked: 0
post #2153 of 2816
07-15-2014 | Posts: 23
Joined: Jul 2014
Quote:
Originally Posted by Spok1701 View Post
Well, I have been away for a bit (still no AX100 - Sony are out of stock again, but I'm hopeful) and this thread seems to have livened up.

I think there seems to a bit too much criticising (not naming any names, but you know who you are) and lack of respect of people who are here to help by sharing their videos and experiences. That's why I joined this AX100 thread. I need as much information about this camera as possible before I commit and buy it and I have had very helpful answers to my questions.

Nobody should be saying what others should be shooting with their camera. That is up to them. If you don't like what others have shot, or their style of editing, then please show us the way by posting some of your own work. Not everyone here is a professional with years of experience, but that's no reason to knock them down.

From what I have seen, this camera produces great images. Better than anything else in this price range. That, for me, is what counts. Codecs, sample rates, f-numbers, etc, etc, are not relevant to the end viewer. It is the quality of the final output that is important, and this camera has it in abundance. I can't wait to get mine.
Point taken and understand your view. I guess, like explained earlier that the radius extended slightly outside the initial range.
Hopefully this can be recovered over time and I advice you to get the AX100 knowing what it's good at and where it falls behind.
In most cases, this can be worked around. Don't think shooting movies on ARRI 435 Xtreme is a walk in the park just because the footage looks amazing from the Kodak Vision3 stock.

To achieve the top notch results, everything needs work. There are people here with knowledge and people that use their stuff, regardless of levels to show what they can do. This is, like you said, more important than just having the camera and telling others what to do.

Now, I just did a quickie. My intention is to be kind and show some stuff to do. Unfortunately, I'm reputated low atm but time will change so
I guess f.ex Mark reads my lines being careful.

Still, there's one problem I have but I just ignore it. It's the best for all and for this thread.

Go get one and have fun. Summer arrived and babes are in all the parks

Cheers,
A
aholck
03:54 AM Liked: 0
post #2154 of 2816
07-15-2014 | Posts: 23
Joined: Jul 2014
Quote:
Originally Posted by David Harry View Post
All easy mistakes buddy. Yes, the F23 was amazing back in the day and is still probably the best 2/3 HD camera ever made. But a real bitch for doing selfies

So anyway. What NLE do you use and what scaling method? I am always interested in these types of things when someone uses similar kit.

Any chance you can share some of your videos, and maybe some native camera files. Again, am just interested in how people work. Like Mark for instance. His productions and workflow are quite different from mine, but I have picked up some really beneficial and practical tips from his work.

Cheers,
Dave.
Haha, Selfies. Well, I'm indeed a pro staying away from the end of the rifle.

Like I said, I've been travelling and still have loooads of footage to take care of. My creations with the AX100 has been delivered in 4K to clients/friends but remain at a private stage until anything else is stated.
Now, I will try and put some time into actually figure out best way of scaling down. I for one love working in Resolve and so far I've exported footage
just into 1080. I'm not sure if I'm mistaken by saying I remember an option of biqubic but might be in other NLE.

I work in two NLE's. On my laptop, I grade in Filmconvert (w/o stock emulation) and edit in Final Cut Pro X. This is because of Resolve requires a frame buffer like NVIDIA. So, I run Resolve on my PC with Premiere Pro CS6.

Yes, you can have some native files. Let's retalk once I have footage available for others.

// A
David Harry's Avatar David Harry
04:17 AM Liked: 18
post #2155 of 2816
07-15-2014 | Posts: 129
Joined: May 2014
Quote:
Originally Posted by aholck View Post
Haha, Selfies. Well, I'm indeed a pro staying away from the end of the rifle.

Like I said, I've been travelling and still have loooads of footage to take care of. My creations with the AX100 has been delivered in 4K to clients/friends but remain at a private stage until anything else is stated.
Now, I will try and put some time into actually figure out best way of scaling down. I for one love working in Resolve and so far I've exported footage
just into 1080. I'm not sure if I'm mistaken by saying I remember an option of biqubic but might be in other NLE.

I work in two NLE's. On my laptop, I grade in Filmconvert (w/o stock emulation) and edit in Final Cut Pro X. This is because of Resolve requires a frame buffer like NVIDIA. So, I run Resolve on my PC with Premiere Pro CS6.

Yes, you can have some native files. Let's retalk once I have footage available for others.

// A

If you have a PC, maybe check out Edius 7 by Grass Valley. It is hands down the fastest and most powerful NLE of them all, for real time ability without extra GPU's.

Edius uses variations of Lanczos, including 2 and 3. In my experience, this is the best readily available algorithm for re sizing. You can also find it free in many free encoders, but don't know what other NLE's use it.

I use Windows 7 PC's myself, but can also run Apple Mavericks on one of them. So will try out a down sampling test later with FCP X. I have a great file for checking aliasing/moire when down converting UHD to HD. If the differences between Edius and FCP are noticeable, I will post some results later.

Cheers,
Dave.
aholck
04:25 AM Liked: 0
post #2156 of 2816
07-15-2014 | Posts: 23
Joined: Jul 2014
Quote:
Originally Posted by David Harry View Post
If you have a PC, maybe check out Edius 7 by Grass Valley. It is hands down the fastest and most powerful NLE of them all, for real time ability without extra GPU's.

Edius uses variations of Lanczos, including 2 and 3. In my experience, this is the best readily available algorithm for re sizing. You can also find it free in many free encoders, but don't know what other NLE's use it.

I use Windows 7 PC's myself, but can also run Apple Mavericks on one of them. So will try out a down sampling test later with FCP X. I have a great file for checking aliasing/moire when down converting UHD to HD. If the differences between Edius and FCP are noticeable, I will post some results later.

Cheers,
Dave.
Will take a look at it. Yes, please post some comparison videos. I do have a PC but my laptop is a mac. I would love to have same engines on both devices.

One problem with Premiere Pro is for me, dont know if it's something wrong, is the sooo slow editing mode. If I move the bar to cut, it might update after 5 seconds with a different part of the clip. Maybe it's intended to use with proxies. Im talking about HD 422, not UHD.

// A
Spok1701's Avatar Spok1701
04:37 AM Liked: 13
post #2157 of 2816
07-15-2014 | Posts: 32
Joined: May 2014
Quote:
Originally Posted by aholck View Post
Point taken and understand your view. I guess, like explained earlier that the radius extended slightly outside the initial range.
Hopefully this can be recovered over time and I advice you to get the AX100 knowing what it's good at and where it falls behind.
In most cases, this can be worked around. Don't think shooting movies on ARRI 435 Xtreme is a walk in the park just because the footage looks amazing from the Kodak Vision3 stock.

To achieve the top notch results, everything needs work. There are people here with knowledge and people that use their stuff, regardless of levels to show what they can do. This is, like you said, more important than just having the camera and telling others what to do.

Now, I just did a quickie. My intention is to be kind and show some stuff to do. Unfortunately, I'm reputated low atm but time will change so
I guess f.ex Mark reads my lines being careful.

Still, there's one problem I have but I just ignore it. It's the best for all and for this thread.

Go get one and have fun. Summer arrived and babes are in all the parks

Cheers,
A
Aholck, you are very welcome here. You obviously have knowledge and experience that will benefit us all.

"Still, there's one problem I have but I just ignore it."... Is that a problem with the camera? Please feel free to criticise the camera, that can only be helpful. The AX100 appears to be fairly easy to use, but, again it only costs $2000. The ARRI on the other hand...

Going from a Panasonic SD700, one of the greatest things on the AX100, from a usability point of view, will be having a good quality, tilting, viewfinder. Shooting on bright sunny days will be so much easier.
markr041's Avatar markr041
06:09 AM Liked: 145
post #2158 of 2816
07-15-2014 | Posts: 3,333
Joined: Jun 2005
Quote:
Originally Posted by aholck View Post
Hi Mark!

Please read carefully below so we won't have any misunderstanding.

The following video will show some interesting details in your footage you didn't show
It will repeat by playing original, an applied fix and last, a graded option w/o applied fix.
Note, most playback software as well as streaming services online expects RGB while this camera records in YUV color space.

The problem is, from the ITU.709 specs an YUV decoder expects white level to be 235 at maximum. RGB decoders expect white to be 255 on 8-bit channels.
Vimeo, Youtube etc will all decode the codec as RGB footage, blowing out highlights. This is default manner of Quicktime, VLC, Youtube, Vimeo etc.

I put the output value of white at 220 in every 2nd clip to ensure a clearly visible difference.

It is important to understand that this is a very quick thing I did and I didn't split the clip to grade each clip alone.
I made a generic grade to lower the "grey" layer on top as well as removed the blue tint from the incorrect auto white balance.

The rendering engine of my NLE obviously couldn’t mix HD and UHD so my footage has a 4th of it’s quality and looks like LEGO.
So, don’t mind my quality - just see how I would prefer the coloring. I could fix it but heey - this is not a quality thingie..
Only for original => Fix => color grade (which actually also changed due to the lower res).

Some of my grades ARE too dark and I admit that it could be much better. But, like I said, I made one trying to match all
With the energy I could fix both the YUV highlight clip limit, grade and so on. My clips were exported in 1080p thus slightly lower res.
This is because I didn't insert the license USB stick to unlock 4K in davinci resolve.
Only to show color difference.

Now, please download the file on the link and hopefully your system will be able to show you what I did.

The screen used in this case is calibrated 2 month back. My two presets are still photo and other is for video editing.
The video calibration is based on REC.709/BT.709 standard from ITU.R.

http://en.wikipedia.org/wiki/Rec._709

Now, for the clip.

http://213.115.239.33/mark/mark.mp4 (800MB+)

One of the most visible differences are at around 1:09. Look at the plastic bag as well as on the sign to the left once they pass it.
In 2nd part of the same clip, you can see more details in the bag as well as the sign got letters written

Same in the first clip of the clouds.

One clip you see a colourfully dressed woman walking with a stroller. The white car passing on the white lines as well as the white lines do
have more white information now.

At around 2:35 you’ll see the people walking towards you. Notice in the 2nd clip how details came out from the blown out part further back on the road.

Further on, the bricks on the other stroller clip where the kid jumps down are blown out but I managed to at least recover the edges slightly.
If you check on the flower in the end, I’m sure you’ll see some more detail in the whites of the flower.

These are details and people often complain about these stuff on this particular camera. This is because of what I wrote above and it’s very easy
to correct to achieve details in the white areas.

Maybe I’m too sensitive but I want 100% advantage of my camera.

Thanks for doing this. Your modified first version for each clip is a real improvement in each case.


You did choose clips that in my view were overexposed to begin with (e.g., the little girl with family, the fish - this was my first time with the camera), and it appears that simply lowering the exposure in post results in a very visible richer picture with details revealed in the blown parts (I know you did more than that, but I think most of it is that). The changes do show that there is amazing hidden latitude. Others have reported they can equally pull up shadows and they find details. The "superwhites" issue has come up before. And I agree the camera picture at times is too blue.


Given the latitude, you may be right that one can use etr with this camera, just like the BMPCC, if you are willing to put in the work in post. It was an enjoyable experience working with etr-shot BMPCC RAW files, and watching the full colors emerge in Resolve by lowering luminance, and thereby getting exposure just right at a leisurely pace rather than in the field.


I am not fond of the grading (I ignore the terrible aliasing, as I know that can be fixed). It just looks muddied. That is a matter of taste, of course.


I really appreciate the effort. This demonstration does appear to show the camera is even better than what the straight ooc shots indicate, with clips that can be manipulated for the better in post (within limits) without degradation, at least for clips that need improvement.
Ken Ross's Avatar Ken Ross
06:30 AM Liked: 1663
post #2159 of 2816
07-15-2014 | Posts: 24,862
Joined: Nov 2000
Quote:
Originally Posted by Spok1701 View Post
"Still, there's one problem I have but I just ignore it."... Is that a problem with the camera? Please feel free to criticise the camera, that can only be helpful.
Fear not Spok, the 'one problem' he refers to is me. He has had a 'thing' for me from the get go as evidenced by his demeaning remarks. He doesn't like me and certainly doesn't like my videos. That's OK, there are many that do. Tastes differ.

However, the only thing that I truly dislike is any insinuation that in my 14 years on AVS, I have done anything but try to help people. Those that know me, are well aware of that. I do this publicly and do it via PM. So regardless of how much someone has to offer in terms of the subject being discussed, it means nothing to me if that person is insulting and can't respect the fact that videographer's here have different styles and different approaches to content. Put simply, we all don't see things the same way.

We can all learn from each other, but if we can't do it in a respectful way, the S/N ratio here will drop to zero.
Ken Ross's Avatar Ken Ross
07:34 AM Liked: 1663
post #2160 of 2816
07-15-2014 | Posts: 24,862
Joined: Nov 2000
Quote:
Originally Posted by markr041 View Post
Thanks for doing this. Your modified first version for each clip is a real improvement in each case.


You did choose clips that in my view were overexposed to begin with (e.g., the little girl with family, the fish - this was my first time with the camera), and it appears that simply lowering the exposure in post results in a very visible richer picture with details revealed in the blown parts (I know you did more than that, but I think most of it is that). The changes do show that there is amazing hidden latitude. Others have reported they can equally pull up shadows and they find details. The "superwhites" issue has come up before. And I agree the camera picture at times is too blue.


Given the latitude, you may be right that one can use etr with this camera, just like the BMPCC, if you are willing to put in the work in post. It was an enjoyable experience working with etr-shot BMPCC RAW files, and watching the full colors emerge in Resolve by lowering luminance, and thereby getting exposure just right at a leisurely pace rather than in the field.


I am not fond of the grading (I ignore the terrible aliasing, as I know that can be fixed). It just looks muddied. That is a matter of taste, of course.


I really appreciate the effort. This demonstration does appear to show the camera is even better than what the straight ooc shots indicate, with clips that can be manipulated for the better in post (within limits) without degradation, at least for clips that need improvement.
To my eyes the real improvement was in the middle clips with the family walking with the white plastic bag and about 2 or 3 subsequent to that. I remember those well and do recall you were new to the camera and hadn't yet mastered the exposure & the zebras. I suspect with the way you're shooting today, most of this would simply be unnecessary. Perhaps if there was a need to lift shadow detail at the expense of something else in the frame, you might want to alter exposure in post, but for the vast majority of clips, exposed properly, there's probably little need.

However there were other clips (dinosaur, snake, the opening clip and a couple of others) where improvements were more subtle. There's a tradeoff between subtle changes to highlights and the overall impact of the initial contrast. In those cases I'd say it's a judgment call that might be dictated by what the true subject of the scene is. Do we care if a cloud or two is overexposed if they're not the subject of the scene?

However, not surprisingly, I think in each and every case, IMO, the grading did nothing but detract from the original coloration. We went from a generally natural looking color palette to the typical graded, 'shades of brown' look that we see so often. As I've said before, rarely have I felt a need to touch up the colors with this camera. Certainly less so than almost any camera I've used in the past. But having seen this 'shades of brown' so often in graded videos, it is apparently a look that many like. For me, not so much.

Reply Camcorders

Subscribe to this Thread

Powered by vBadvanced CMPS v3.2.3