A99 Soft Video Note~

Hey guys, I've seen tons of posts about "soft" video from the A99. I've also seen examples posted of very sharp video from A99's so what might be the culprit.....I say "aperture" and the way the A99 defaults to biggest aperture value (large hole.)

For those that don't know, the A99 defaults to the largest aperture setting when recording video. The largest aperture is usually soft on most lenses and must be stopped down a few stops to get sharp images. To enable your own aperture setting when shooting video, switch the camera to the "movie" setting and hit the Function button and select either manual or aperture priority modes- you will now have control of the aperture setting and you can manually stop down the lens to it's sharpest value.........shoot video and check for sharpness- i think you guys would be surprised at just how sharp the A99 can shoot.

I too was very disappointed at the video quality of the A99 until I took control of the Aperture setting.......sharp video is now always possible.

I hacked a few of my GH2's and know sharp video and am now impressed with my A99.

I know this is a simple solution but for those who haven't shotten in Aperture priority mode, this may be your culprit for soft video.

The fact that the A99 is capable of such great image capture with stills should indicate the sensor isn't the issue, the problem was either compression or setup.....in the case of the A99, the sensor demands the sharpest settings from your lens which is usually F8 or thereabouts.

Hope some of you guys try this and give the A99 a second chance and get good results.

Enjoy~

--
Steve Nunez~New York City
www.stevenunez.com
Sony NEX7, NEX5, Sony A99, Panasonic GF-1

I thought the problem was that the part in focus, at large apertures, was sharp, but when stopping down somewhat, to increase DoF, the parts in focus are actually softer.

I got my A99 exclusively for stills, so this aspect isn't taking anything away from my enjoying the camera. Like you said, since the stills are tack sharp with plenty of detail, it's not the sensor, so the softness issue could technically be solved via a firmware update.

Steve, I have been always defending the sharpness of the A99, but in all fairness, the discussion was never about the auto-focus video recording mode, where the A99 chooses f3.5 automatically.

There were posts where people shot in full manual mode with the aperture set to f8 or higher. (Not exactly the sweet spot of the lens which is usually wide open, stopped down to f stops)

The criticism is that the A99 does not resolve as much detail as other cameras do at the same f stop.

There have been comparison shots between the A99 and other cameras and still images were pulled of the video and compared.

I will not start again with my personal point of view and my own experience as I am very happy with the A99's video function and that is all that counts for me. I consider pixel peeping at still images from video clips a waste of time and for me the A99 is all about shallow depth of field something that I can not achieve as easily with my video cameras.

Stef.

--
http://stefaniekappel.tumblr.com/
http://www.flickr.com/photos/15931938@N05/

I shot everything full manual, everywhere from f2.8-16 but mostly at f8, steadyshot off, 180 degree shutter to avoid motion blur, various profiles with varying levels of contrast/saturation/sharpness and so on until I could find the best settings.  I even shot stills manually focused with focus magnify and zoomed in to check for critical focus before shooting video.  I also slowly racked focus while rolling in case I missed focus - but I didn't - I nailed focus on all my shots with all three cameras I used.  Its not the lens because I used it to shoot tons of stills on a paid gig and video on my Nex5n.  Its not because I got two duds a month apart and everyone else has perfectly functioning cameras.  Its not my settings because I spent hours checking footage on broadcast reference monitors.  Its not the light because I shot in bright mid-day sun.  Its not some Sony-hating agenda because I sold a lot of other gear to make the switch to the A99 and I'm very pleased with the stills I shot with the camera.  It actually made me a nice chunk of change.

Its the camera.

Chris just wanted to ask- are you shooting 25p (30p) or 50p (60p). yesterday i shot 50i -imported into fcp as normal but it looked softer than normal. Not sure why but just a thought. The other thing is have you tried different lenses as perhaps you have a faulty lens?

Digitalrev, a usually Nikon friendly and Sony opposing reviewer preferred the video quality of the Sony A99 over the Nikon D600.

http://www.youtube.com/watch?v=_SShHCsuI94&feature=youtube_gdata_player

Or here:

"Sony A99 is a winner with terrific video..."

http://www.newsfactor.com/news/New-Sony-A99-Camera-Is-a-Winner/story.xhtml?story_id=012001D1AZTO&full_skip=1

We are still missing more comments from professional reviewers and all the anti- A99 hype on this forum is based on a couple of posters that post over and over again their findings. We have though at least an equal amount of "professionals" on this forum who are very happy with the video performance of the A99.

IMHO the jury is still out on this one, but slowly the reviews start to come in. Let us wait and see?

Stef.

--
http://stefaniekappel.tumblr.com/
http://www.flickr.com/photos/15931938@N05/

It seems confusing to me.

We have a small number of ( it must be said, very vocal and persistant ) members on DPR who are unhappy with the A99 video performance.  Fair enough,  I have no reason to doubt their position and taken at face value they do seem pretty unhappy.

And yet, I have seen A99 video that looks just fine to me.

And, it must be said,  I have'nt seen any rival FF video ( Canon for example ) that really looks any different.

It may well be that there is something in the software that is making the A99 underperform for many people.  But if you dig around there are plenty of people complaining that the Canon 5D3 is soft as well, so it is hard to make any sense of it.

The A99 is still very new.  Hopefully a clearer picture can be extrapolated from when there is more evidence out there.

Right now we have the same small number of people complaining in the same circular way.  We need a larger sample size IMO.

What sets this apart is No one can contradict their views.  Sony released a defective product?  Sony has no experience at this? Not at all.  The subject is flogged so hard that the dead horse is a red blotch in the sand.  Time to move on.

--
proud father of an a100, a77, and a99
John D. Conley, Canuck

I'd like to see more testing done on this personally.

Specifically on the lens and sensor feature side of things. ie. when I was looking over the Nex5n video test + mini A99 hybrid shooter, I found the differences akin to an optical blur effect. Which leads me to believe that the cause may potentially exist between the lens and the sensor. At which point, I'd try to eliminate as many of the potentials problems from the equation as possible. ie. the use of an m42 prime lens on both test systems would be a good start to such a case study. Likewise, the addition and removal of as many system features as possible would likely be another helpful step in the evaluation process as well.

From where I stand, I see no need to hide from the fact that the A99 could potentially be malfunctioning at this point given that we have yet to find out the cause of the discrepancy. Though more importantly, I'd add that the soon we can get the bottom of this, the sooner we can expect a fix as well. And so I'd say there's that to consider also.

PS. anyone in the mood to perform some tests?

--

This is where I'd normally write an impressive summary of my skills and proficiencies.

I have a Nex-7 with a SEL18-55 and an A-99  with a SAL28-75  and give equally nice video with equal sharpness both on automatic mode.....just an observation

I don't thinkit's flawed- quite the opposite.

I have a US camera, so all my shooting was at 24/60p - the US version doesn't have a 30p option. Interlaced video is always going to be softer as you're taking images and splitting them into two parts - I always shoot progressive.  The lens is fine, I got great results shooting stills and in the side-by-side with the 5n I used the same lens on both cameras via the LAEA1 adapter.

chris

What's lens do you have photo chris?

The sample videos speak for themselves comparing the a99 to other cameras.  DPR's own videos they took speak volumes.  If other cameras don't look any better then you really haven't done due diligence looking at other videos.

It's not just a few either - there are many truly objective people on here that have also commented about the video IQ of the a99 - in fact, I would submit there's just a few who refuse to admit it; not the other way around.

IQ shapness  CAN BE a problem, on a LIMITED number of A99,i have nex-5n,7,A77 and the LE-A2 adapter and make some video testing on  the sharpness newpaper' s letter and all these cameras give very very near the same results.........

The Zeiss 24-70 was used in all my tests.

I have a NEX7 and adapter for Sony Ff lenses- I'll try shooting video using identical settings and see if I can resolve more detail in the NEX7 shots as opposed to the A99 footage......hopefully within the next week.

If someone else can do the same test using the same lens- I think we'd all like to see the end results....we'll need to see exported frame grabs of course.

--
Steve Nunez~New York City
www.stevenunez.com
Sony NEX7, NEX5, Sony A99, Panasonic GF-1

i'm on the "a99 video is soft" side of the fence, and i've commented a couple of times about this and otherwise generally steered clear of what i think is pretty fruitless discussion on this forum about it.

i've mentioned this before, and i don't know if this has been mentioned by anyone else here recently, but i am quite certain the problem is caused by the line skipping technique used to downsample the 24mp image off the sensor to the 2mp video frame. that is pretty much the end of the story, and it has nothing to do with settings or lenses or faulty cameras or avchd compression, and can't be fixed by firmware updates. the a99 exhibits similar video quality problems (softness, moire, aliasing) that the 5d mk2 has, which famously line skips (moire problem exacerbated due to line skipping combined with the olpf). the 5d mk3, while still relatively soft, doesn't have moire and aliasing problems because it no longer line skips and instead has a 3x3 pixel binning technique (allegedly). the gh2 doesn't line skip (exact technique unknown, but supposed to be some combination of intelligent pixel sampling to a 4 mp image followed by downscale to 2mp) and exhibits far less problems with moire and aliasing, and has arguably the most detailed video of any of the dslr style cameras.

so i just did a quick google (https://www.google.com.au/search?q=a99+video+line+skipping) and to not much surprise, many other people on other forums and blogs are suggesting line skipping as the problem both on the a99 and the vg900. take that for what you will i guess, as i'm sure most of the replies to this will be that all those people are stupid, don't know how to test, have faulty cameras, faulty lenses, aren't professionals, shouldn't be using a dslr to shoot video, etc.

now i am about to completely speculate on this next point, but because the a99 was engineered to have 1080p60 as a headlining feature, the line skipping downsampling algorithm was compromised even further to throw away more data so that the downsampling could be applied 60 times a second. perhaps had the 1080 video mode been limited to 30fps, fewer (half as many?) lines could be skipped and the video image quality at least improved over what we have now, perhaps to a tolerable level.

the problem with this forum in particular is that it is filled with a) people who can't possibly fathom that an otherwise great photo camera could possibly have a poor video mode (and can't possibly understand that it is worth criticising it so it will get improved as it did when canon listened and fixed the moire/aliasing on the 5d3), b) people who have great technical photography knowledge, but don't have the faintest clue about video, c) people that on one hand will tell you that the a99 really does have a great professional video quality, and on the other hand will tell you that you should be shooting video with a proper video camera because the a99 video is not professional enough.

I agree with this assessment, I've said from the beginning I think its caused by hatcheting the 24mp image down to ~2.1mp, line skipping is not the way to go, All-I from what I've seen produces better results.  The A99 does exhibit very similar traits to the 5d2 in the area of moire and aliasing, but the 5d2 is sharper.  I cut footage from my Nex5n with the 5d2 - using Zeiss ZE lenses via a fotodiox dumb adapter on the 5n - and side-by-side the 5d2 footage was sharper after adjustments were made in post.  I'm holding on to my Nex because its a crazy cheap backup that shoots nice 60p.  Discussions that level any criticism at the A99 really seem to rub some people the wrong way, I'm checking out of these threads until I have other test cameras to compare with the A99.  Enjoy your cameras.

When an end result is the objective you state...I need to do...x .  So I want to produce an award winning tv documentary.  What gear do I need?

The hunt is on.  You DO not buy a dslr!  You go pro.

I wanted to take good stills, video is a bonus.  Under 5k.  ....a99!

Sony has a formidable lineup of pro cameras....if you want video first, you go there.

Easy right?

John D. Conley, Canuck