FS5: FS5 Highlight Artifacts.. Another Imaging problem and a Workaround

dvbrother said:
Interesting. And in the interest of gettting to the bottom of this, what lens did you use? I'm wondering if the Auto-Correction on the 18-105 kit lens might have anything to do with it.
Was the Color Gamut on your SLOG2 footage the SLOG2.cine/3200k? Your SLOG footage seems more saturated than mine.
Click to expand...

There is no such Color Gamut as "SLOG2.cine/3200k". I used the stock PP7 profile except to change the color mode to "S-GAMUT/5500K" instead of "S-GAMUT/3200" to account for the color temperature of my daylight balanced LED desk light. When the three S-LOG profiles are used, nothing else should be changed in the profile except to choose choose one of the three white balance presets that matches the Gamma choice and the prevailing color temp of the lighting.

The lens I used is my 50mm Cine-Xenar. I believe 90% of the problems people are blaming on the FS5 really come down to using the wrong settings and/or crappy glass. It is amazing to me that people will post camera tests using that awful 18-105 lens and then blame all the defects on the camera. That lens needs distortion control, plus it looks like SteadyShot and AF are probably being used in some people's tests too. Use a real lens folks, and then we can talk about any remaining deficiencies in the camera. I've shot the hell out of my FS5 since I got it before Thanksgiving and the camera looks fantastic.

BTW, the profile I'm using in the first half of my test has Detail at +7.
 
Cheap Glass. (is a cure)

It is worth considering IMO that some glass is over sharp and softer glass can help.

Ive certainly seen it worse on my FS7 with canon cine glass than my own nikkors which seem a little more relaxed.

S
 
morgan_moore said:
I dont think you have the really sharp edges that induces it.. or maybe not enough over exposure.
Click to expand...

Oh, come on. Do I also have to stand on my head and wait until an odd numbered day with a full moon? The footage was a close as I could make to simulating the other test. The problem either exists or it doesn't, and I say it doesn't. There ought to be a law against people posting camera tests that use cheap crappy e-mount toy lenses or focal reducers. Especially when IS, AF, and Distortion Correction are being used. They don't even disclose those shortcuts in their tests -- as if the lens that is used doesn't even matter. Well glass always matters. Put the blame where it belongs, and it's not the camera.
 
Last edited:
Narky! Clearly the problem exists in some situations - do those sitiations intersect with situations that are "competent" use of te camera? I make no coment.
 
I think Doug shot as close to the original footage as possible, but his doesn't have any fringing.
Hard to tell whats causing it, but obviously we won't know everyones settings in each test, etc etc.

Its very likely Doug knows how to use the camera better than all the other operators combined... Makes you wonder what other settings or combinations of things could be causing the fringing .. and not 'just the camera is doing it' .......(no offence to anyone of course... He certainly knows more about the cameras I make my living with than I do!)
 
chappelldigital said:
I think Doug shot as close to the original footage as possible.
Click to expand...

You can tell that from looking at video clips? Wow, you're good. Real good.

In my experience the highlight aliasing aka ringing on the FS100 became visible when a certain threshold in contrast had been exceeded. A softer lens passes lower edge frequencies so the ringing shows up less and maybe not at all because the contrast threshold is never met. The curvature and material finish of Doug's lamp stand is probably producing a slightly softer specular highlight with a slightly softer gradient which is not causing the ringing artefact to be present.
 
You people are unbelievable. How am I supposed to prove a negative? No matter what test I try to come up with someone's comeback is always going to that my procedure was somehow flawed. I could do a hundred tests that didn't show the problem, and someone would still say that I didn't do it right. Well if the problem is that hard to recreate when you are purposely trying to create it -- THEN THERE IS NO PROBLEM. I won't waste my time doing further tests. I've satisfied my own curiosity and I know where the problem lies, and it ain't with the camera.
 
Doug Jensen said:
You people are unbelievable. How am I supposed to prove a negative? No matter what test I try to come up with someone's comeback is always going to that my procedure was somehow flawed. I could do a hundred tests that didn't show the problem, and someone would still say that I didn't do it right. Well if the problem is that hard to recreate when you are purposely trying to create it -- THEN THERE IS NO PROBLEM. I won't waste my time doing further tests. I've satisfied my own curiosity and I know where the problem lies, and it ain't with the camera.
Click to expand...

There's no point getting testy Doug but the aliasing issue that is present on the FS100 (which I own) didn't always appear and required certain conditions to be met before it "switched on" and _if_ there is an aliasing issue on the FS5 it too may have similar characteristics but the FS5 is a not a FS100 and the processing and maths will be different and it hasn't been ruled out that other effects may be causing the aliasing/sharpening/ringing artefacts.

On a sunny summer's day the specular highlight hit of polished chrome was an absolute dead cert to present the aliasing but not on the highlight of a polished car. Tree branches against the mid afternoon Sun would alias whereas branches agains a red setting sun would not. I only really bothered looking out for the issue when shooting in very hard light that can't be controlled.
 
Doug Jensen said:
There is no such Color Gamut as "SLOG2.cine/3200k".
.
Click to expand...
Ah yes, I just misremembered, as I've only had the camera for three weeks. Yes, I'm aware that the gamuts for SLOG settings shouldn't be adjusted.

Well, I suppose since my credibility has been questioned for daring to use the 18-105 kit lens, when I return to work on Monday I will try the Canon 24-105 and perhaps one of our Rokinon primes. Doug, I respect you, but please don't reduce this argument down to the incompetency of the tester. It seems to be the prevailing approach on this Sony forum.

Perhaps a better way to phrase it would be that the Sony 18-105 kit lens is the likely culprit, and the solution is to use better glass. Rather than insinuate that I'm stupid for even using the lens. Thank you.
 
Doug Jensen said:
Oh, come on. Do I also have to stand on my head and wait until an odd numbered day with a full moon? The footage was a close as I could make to simulating the other test. The problem either exists or it doesn't, and I say it doesn't. There ought to be a law against people posting camera tests that use cheap crappy e-mount toy lenses or focal reducers. Especially when IS, AF, and Distortion Correction are being used. They don't even disclose those shortcuts in their tests -- as if the lens that is used doesn't even matter. Well glass always matters. Put the blame where it belongs, and it's not the camera.
Click to expand...

This is the best thing I've ever read on a camera forum. +1
 
Not stupid, I would say careless in your testing. I'm sorry if my comments have offended you, but there are just too many people ready to pounce on the shortcomings of the FS5 (and other cameras) wtihout first ruling out other potential sources of trouble. That's just Troubleshooting 101. And in the case of that lens, I think it should be quite obvious that it would be one of the first things that needs be ruled out.

You didn't just post here with some questions and ask for opinions. No, you took the grand step of doing a big blog about it and making a big stink. That's your right to do, but I think it does a disservice to all of us who have spent our hard earned money on the camera and expect it to be accepted by clients and colleagues. You have damaged all of us. The camera definately has shortcomings, but lets be careful about blaming things on it that aren't fair.

BTW, when you test with the Canon lens, please use a non-speedbooster adapter, since the optics in the adapter may contribute to the problem. Troubleshooting 101. :-)
 
I guess this could also be caused by post workflow. I've had the same footage exhibit awful moiré in one program and be completely fault free in another.
 
Doug Jensen said:
You people are unbelievable. How am I supposed to prove a negative? No matter what test I try to come up with someone's comeback is always going to that my procedure was somehow flawed.
Click to expand...

You are not obliged to do this - I don't see it reflecting badly on you not being able to recreate an unknown situation. I still guess it happens sometimes though - it does on the fs100 and does on the fs7 In both 1080 and 4k so it is not really a suprise
 
It is interesting. When I test a new camera, I try to find out how to get nice images from it, not terrible ones.

I have not seen this effect in any footage I have shot, over 20 hrs worth.

I don't spend my time pixel peeping 400%. That is a waste of time. And if I see some banding, blocking, or strange color fringing I try to figure out a setting that gets rid of it, not how to make it look worse.

The kit lens is a $500.00 lens and for that it is a very good value, but there is a ton of processing going on to make it work. So it is far from a great lens. But even using that lens I have not seen this nor will I go and try to recreate it. I will just keep on shooting with my settings I am developing and so far I am very happy with the FS5 and I even like the 18-105 kit lens for what it is, I will never use it for my most critical work, but it is on the cam a lot. And I have about 20 lenses I can use on the FS5, so...

If I had a lens that was giving me fringing in every shot, I would return it. Its is very possible there are some bad lenses out there, how good can QC be on a $500.00 lens. I have actually returned a $2,000.00 Canon lens that was defective when I got it. Canon replaced it with no questions asked.

I am getting pretty sick of the hysteria on this board. But I guess this is the nature of social media.

Anyway I am back to shooting some skin tone testing of PPs matching my FS7 and FS5 settings for upcoming shoots.

Have a great weekend and don't become frustrated by the tone of some of the posts here, if it works for you great keep shooting. If not send it back or ask for a replacement if you think you got a bad camera/lens.
 
Doug Jensen said:
Not stupid, I would say careless in your testing. I'm sorry if my comments have offended you, but there are just too many people ready to pounce on the shortcomings of the FS5 (and other cameras) wtihout first ruling out other potential sources of trouble. That's just Troubleshooting 101. And in the case of that lens, I think it should be quite obvious that it would be one of the first things that needs be ruled out.

You didn't just post here with some questions and ask for opinions. No, you took the grand step of doing a big blog about it and making a big stink. That's your right to do, but I think it does a disservice to all of us who have spent our hard earned money on the camera and expect it to be accepted by clients and colleagues. You have damaged all of us. The camera definately has shortcomings, but lets be careful about blaming things on it that aren't fair.
:-)
Click to expand...

I seriously doubt that my posts or blog entry has damaged anyone. That overestimates their reach. Nor was it my intention to jump on some kind of FS5 Bashing bandwagon. I will admit that the test was hasty, and that can be chalked up to the fact that I have other work to do at my job. Like we all do.
In my defense, I have seen this same problem on the FS700 and the choice of lens was irrelevant. Again, in my haste, I failed to reveal the entire backstory of my hunches that led to this test in the first place. I initially did a much more controlled, studio lit, locked-on-a-tripod test with the FS5 and FS700 in order to match them because our company will use them together for 2 camera shoots. I was more interested in comparing their color and dynamic range, etc. This highlight problem with the statuette showed up on both cameras. The FS700 had the Rokinon 85mm at f4.
Regarding Speedboosters, I do not have one, but I would be inclined to think it would reduce the problem, as it seems excess sharpness is a contributing factor.

And regarding "blaming the FS5 for things that aren't fair", nowhere in my blog or posts have I called the camera inferior or bad. I haven't bashed Sony or demanded any kind of fix. I posted something that I observed about the behavior of the camera. It's a visual issue that I find particularly aesthetically annoying. Which is my right as a videographer. The Panasonic cameras I have been using for the past nine years or so never exhibited this problem. Again, in my blog entry, I did find a workaround. The title of this thread says as much. Since it seems most people are using the FS5 as a 4k camera, it will not be an issue for them except in slo-mo.

I posted this just to make people aware of it and also see if others had experienced the same. I suppose I should have been more direct in stating my intention.
 
Doug, I heard through the grape thread in around 2 weeks there will be a FS5 dvd coming out :beer:
 
Ugh, that's my deadline anyway. Two more weeks before I can can up for air. I will be very, very happy to be done with this one. Posting some snarky comments over hear is really my only diversion!
 
Sounds like yr not quite a fan of the cam.

It was marketed as an easy to operate point and shoot run n gun cam. I feel it was a replacement for the Z1 and EX workhorses.

Ill buy yr DVD as a birthday present to myself :)
 
blue-ropes.jpg

Heck. I don't want to get in the way here as I do not (as yet) have an FS5 and still keeping my options open and, therefore, value ALL comment here. However, I would like to offer up something I have come across with the FS7K (kit lens) similar to posting #16.

It's an action shot and has motion blur due to the rope swing and action-follow in the pan; shot XAVC L SLOG3 CineEI 800 5500K 25fps Interlaced. The upper and lower interlace fields are the same so this is not an interlacing issue. There is a blue/cyan hue in the rope as it sits against the blue sky but adopts the correct black against the clouds. The blue sky is at about 75 IRE with clouds at peak 90 IRE on the scale. The blue/cyan hue continues for a few more frames in the action.

I don't have a problem with this as I am shooting action shots, and this is the only 'issue' I noticed in shooting over three weeks of footage on this project - I am not freeking with this. But I couldn't help noticing 'some' similarity with other sample shots with the FS5.

This is just for information. I am not about to ask Sony to send me a replacement, but accept this is one of those anomilies which arise ocassionally under very stressed conditions. However, I do feel the sample given in post #16 is well over exposed and looks stressed.

Again the issue I report is FS7 related.
 
Last edited:
You must log in or register to reply here.
Back
Top