AC130: OIS - Am I expecting too much?

A

amendegw

Active member
My brand new AC-130 was delivered yesterday and I'm putting it thru some tests. It seems to me that the OIS should be doing a better job. Here's a short test I made in my family room with comparison to my Panasonic TM700. Full zoom on each, handheld as steady as I could.


What are your opinions? Should I be calling back to B&H and requesting a replacement?

...Jerry
 
Jerry:

That is normal performance. Nothing is wrong with your AC-130. There is no point in asking B&H to replace it with another.

I could make a video just like this, comparing image stabilization on my AC-130 to my Canon Vixia HF G10, and the comparison would look very similar to your video.
 
I noticed that my XH-A1 had a better OIS then my AC130. It appears your camera is in-line with what I have seen. Little disappointing but not nearly as much as the bad auto-focus on the AC130.
 
The main reason the OIS is better on the smaller cameras is because the lens element that is being driven is much smaller.
 
Full zoom enhances the movements of any camera though. I just tried this out myself yesterday without being fully zoomed in and thought it did a pretty decent job

I would also argue that you could definitely hold a small camera more stable than a larger one. weight distribution and all. ;-)
 
Yeah, when zoomed in, you just see how much your really moving. Really got be using a tripod or some form of stabalization when zoomed in all the way on any camera.
 
Not to mention, the zoom on the 130 is much longer than the tm700... you must have been standing twice as far away using the Ac130 than with the TM700, no? The greater the distance, the greater the effect of camera motion/shake. Try them both at about the same zoom level!
 
Please bear with me, I hate to belabor this point, but I'm still concerned. Here's another test.

Methodology was to point the AC130 out my Family Room door, on a tripod and on my soft carpet. Zoom in to max, turn OIS off and shift my weight back and forth behind tripod. Turn OIS on & repeat.

Shouldn't the OIS=On show some improvement over the OIS=Off?


I'd love to see someone with an AC130/AC160 attempt to duplicate this test.

...Jerry

btw: In the earlier test, I had both the AC130 (22x) and my TM700 (18x) zoomed to the max shooting form the same location - my family room chair.

btw2: Can others see vignetting in these images?
 
Last edited:
Perhaps a sideline… but an interesting question… the tm700 has a 12x optical zoom, and 18x "intelligent zoom" (which I think is just using a smaller area of the sensor, which has a higher than 1080p native resolution). Now, shod this setup make it more stable and/or stabilizable than the AC130?

12x zoom of course should be much more stable, but does the smaller effective sensor size exaggerate the motion enough to make up for this?

OIS works by changing the optics to keep the image steady in the same place on the sensor. Non-optical image stabilization moves the image "window" around the sensor to counteract motion. The hm700 sensor, especially zoomed in to 18x, has a LOT of leeway for non-optical stabilization. Could it be combining OIS and EIS? The AC130 sensor, at 1080p native, does not have leeway for EIS.

So, as an experiment, try them both at 12x!
 
ullanta,

I appreciate your comments but I think getting a 100% apples-to-apples comparison between the TM700 & AC130 may be a diversion to my real issue - is there a malfuction with the OIS in my AC130 that requires repair? Yeah, I'll accept the criticism that I probably named this thread in error. Look at the video segments for the AC130 with OIS=On & OIS=Off. It appears to my eyes like very little (if any) stabilization is being done.

If I get some time, I'll try to put together a 12x comparsion. If I do this, I'll probably start a new thread.

...Jerry
 
Okay, I couldn't resist - by education I'm an engineer and I love to test! Also, I guess I'm going to have to eat my words about this test being relevant (certainly not the first time, nor the last [grin]). It turns out that the TM700 stabilizaiton is still better (at least to my eyes), but the interesting thing is that at 12x the AC130 does show obvious stabilization compared to the OIS=Off mode. As can be seen in the previous video, that difference does not appear in the 22x comparison. Wonder why?


...Jerry

PS: A couple notes about my methodology.

First, on the TM700, I turned the Zoom mode to optical only (i.e. 12x) and OIS to normal mode (not Active Mode) - then zoomed the TM700 to the max (12X).
Next, on the AC130, I set the display to Zoom = mm, then manually set the zoom to 46.8 (Max = 86, Min=3.9, 86/3.9=22, 46.8/3.9=12).

Note that the bird fills the frame differently in each camera. I suspect that is because there is a difference in Full Zoom Out between the cameras. It may have been a better test to match the size of the image in the 1920x1080 frame???
 
Last edited:
I'm not sure I like the test! Testing OIS on a tripod, shifting weight… the cameras different sizes and weights… I dunno. To me it seems like OUS was working in your first test, until the shaking or whatnot got a little more vigorous in the middle. Anyway, try some handheld? My concern is that the tripod mounted approach is restricting the the motion of the camera in strange ways, and there are a lot of threshold issues with such tiny and slow motions...
 
ullanta said:
I'm not sure I like the test! Testing OIS on a tripod, shifting weight… the cameras different sizes and weights…
Click to expand...

ullanta, We're not having a meeting of the minds here (maybe I'm not explaining these tests very well). The tripod test was with the AC130 only - 2 clips one with OIS on and the other with OIS off.

The other tests (first in this post) and the one of the plaster bird (12:06 today - 02-23-2012) were handheld and used both the TM700 & the AC130.

I agree that I can see a small amount of stabilization in the AC130 in the video of my cable box. To my eyes there's a more dramatic difference in AC130 (OIS=on vs OIS=off) in the 12X only test of plaster bird video. Of course I'd really love to see really great stabilization at 22x.

...Jerry
 
Oh, see, I've been lookin at these via my phone... and it seems the plaster bird video is private, but somehow it's opening in my YouTube app and showing the most recently watched video instead... the tree...

Anyway, my mind is muddled today... but I am a little unclear on this shifting weight thing and what I'd expect from it. Remember, IS wants to counter accidental motion (shakiness), and interfere as little as possible with deliberate motion (pans, etc), so I assume there are various thresholds and such of smoothness and degree of motion... and when I picture shifung weight around on the floor with the camera on the tripod, that seems like a sorta weird case that might stymie the thresholds. I'll look for the bird when its public.

I'll try again soon, but I did testing shooting a full moon handheld at full zoom, and definitely saw clear results of OIS...
 
Sorry, Sorry, Sorry. The plaster bird was, indeed, "private". I normally keep my uploads private until everything is checked out, then change them to "public" or "unlisted" when I see everthing is okay.

...Jerry
 
About the only complaints I've had with Panasonic cams are the poor OIS. DVC80, DVC30, and DVX100B that I own are noticeably poorer at stabilizing than equivalent Canon or Sony. A Sony FX1000 I have is much more stable at full zoom (20X) than the DVC30/80 and DVX (10X). Don't know if the 130/160 share the same system, do they have the infamous Pany "clunk" when powered down and tilted back and forth? Your videos look similar to what I get from the tape model Pany's.
 
amendegw said:
Note that the bird fills the frame differently in each camera. I suspect that is because there is a difference in Full Zoom Out between the cameras. It may have been a better test to match the size of the image in the 1920x1080 frame???
Click to expand...

A proper test is to match the field of view. The TM700 doesn't go as wide as the AC130 so the zoom factor relates differently.

That said, this last test shows me that the AC130 OIS is working as I would expect. Active OIS work by using 2 methods in conjunction. The first is movement of a small group of internal lens elements. The effectiveness of this system is heavily dependent on the mass of the movable lens assembly, the power of the motors driving it, and the speed of the vibration. The second method in active OIS is digital and done by cropping the sensor slightly and electronically moving the scanned area sent to the output. This system has the advantage of no inertia at all and is theoretically possible to dampen small vibrations 100%. In practice, this has to be explicitly reduced by the software algorithm or the slightest deliberate camera movement would cause major snapping of the picture as the limits are reached.

The 130 has much more mass in the movable optical elements than the 700, mostly because of the larger sensor, but also partly because the lens is a much more complicated design.

I really don't think your 130 has a fault with the OIS.
 
tmuss said:
the infamous Pany "clunk" when powered down and tilted back and forth?
Click to expand...

That clunk is typical with almost all mechanical OIS systems. When the OIS motors are unpowered the movable lens elements are free to move on their gimbals. Very few lenses have a power-down lock system on the OIS.
 
You must log in or register to reply here.
Back
Top