AC130: Video footage of the Auto Focus in action

Hi Rod

I guess it's now time to be patient and wait for a response?? Funny the 1/3" chip thing attracted me too but it seemed to cause a lot more issues than benefits for me and although the light performance is obviously better it wasn't a big deal for me at weddings as I had to use some lighting anyway!! We cannot have the perfect setup but for me it's as close as it will get.

I think you could also give your dealer a few nudges this week so he knows that your request is urgent and important and you want answers fast!! Sometimes if you are a moth you fly towards the light a little too eagerly...I will not look at the AC-90 until people have used it and raved about it for at least 6 months or more.

Chris
 
More AF testing whiilst Rod waits for a response.

I ran a number of tests today on my AG-AC160A with AREA set ON.

Included in this set was a repeat (10 times) of my "anchor/church" test under cloudy skies (25 feet, 1.5 miles) Z86.

The AG-AC160A achieved quick focus lock-in at both the far and near objects.

No special attention was paid to trying to ease into the far/near targets through high contrast edges. The camera performed consistently. No complaints.

I also ran a timed (10 minutes) static test on some reeds at about 30 feet away (wind blowing/not blowing) - the camera did not get "bored" - it did not go to background scenery. I walked through the scene a few times and the focus went back to the reeds. No inappropriate behavior.

Interestingly, I tracked some water fowl moving slowly beyond some shore reeds. It seemed that so long as you kept the AREA box on the birds, the camera tracked them and did not try to pick up the foreground reeds.
 
The autofocus also worked for me 99% of the time and you will find that the majority of people will not even get into a situation where the AF acts up. Most people might not even use 22X zoom and have a couch to orange cross pan situation like Rod does. I do worry a bit about ELN614 though....his camera does go crazy in bright sunshine with objects coming towards you..... If you are kicking your heels on a Sunday why not setup on the kerbside and see if filming cars approaching you and kept in focus?? That's the closest I can think of without actually finding a paddock with a horse in it. Might also be worth taking the grandkids to a park and get them to run towards the camera at full zoom???

Chris
 
Hi Chris,
As Mike pointed out the contrasts were low on the orange/curtains test, hence my second test with the test chart shots; The issue is ,when trying to recreate a random event one needs a constant subject to keep the variables at a minimum, although in real life this camera is designed to shoot random events under ever changing conditions.

I have now had some communication with Panasonic here who agree that the test charts have more than sufficient contrast. The conversationwith them is, so far, very encouraging for me at this stage and now continues.
 
I used this camera on a job last week and noticed this problem. It was a day-long shoot, outdoors, in flat, overcast conditions. I'd say 97% of the shots I had no problems, none. But on a few shots where zoomed in to full telephoto, or near it, I would get these strange pauses where it didn't want to focus, or seemed to take a nap for a moment before slowly finding focus on moving objects. Not that different than the video of the horse on the first page. It wasn't enough for me to shut AF off, and I don't believe it ruined any key shots (I'm not the editor) bit it puzzled me. I didn't think anything of it until I came across this thread. But now I have to say it would make me think twice about purchasing such a camera.

Otherwise I liked the cam quite a bit.
 
snowleopard said:
I used this camera on a job last week and noticed this problem. It was a day-long shoot, outdoors, in flat, overcast conditions. I'd say 97% of the shots I had no problems, none. But on a few shots where zoomed in to full telephoto, or near it, I would get these strange pauses where it didn't want to focus, or seemed to take a nap for a moment before slowly finding focus on moving objects.
Click to expand...


That is pretty much exactly what I noticed too. Under flat, low contrast lighting, and when zooming out quite a bit, things got dicey. I would also notice movement within the scene throwing off the focus, making it reset to find focus again.

But as I said before, this is nothing new. We are just sharing the same information over and over, re-confirming what the issue is.
 
1 year old ac 160

1 year old ac 160

This day is my AC160 1 year birthday did so manny job with it now i use only manual focus and iris overall this camera is work horse just like my 150 which i sold to my friend that also has manny hours .
Don't know how long these camera using cards last for, I know the old tape camera about 900 hours to 1500 hours. My friend sold his 130 cause he can not deal with manual focus i stop complaining about the focusing i just except it.


P.S I just have 1 more complaint about MOIRE - more people wearing stripe shirt and jacket i had a groom wearing a stripe jacket don't know how to get rid of it I use Progressive like 24p 60p.
 
Last edited:
dhardjono said:
... just like my 150 which i sold to my friend that also has many hours ...
My friend sold his 130 cause he can not deal with manual focus i stop complaining about the focusing i just except it.
Click to expand...
150 or 130? the one 130 does have auto focus if he can't handle manual.


dhardjono said:
...Don't know how long these camera using cards last for, I know the old tape camera about 900 hours to 1500 hours.
Click to expand...
Yeah I wonder the same thing... I assume a lot more hours than that.... no head to wear out.
I know I have gotten to around 2000 hours on my Supercams until the head went. I have 600 hours on my miniDV and still running. My DVCPro I think I only have 200 hours on so I am sure it has plenty more to go.

My 1 year anniversary is around now, too. I didn't start using it until they fixed the time date stamp issue (??January??). I presently have 356 hours on my camera. so 3x = just over 1000 hours in the three year warranty time period.
 
Going to check my time hours ELN also i don't think i will upgrade to A.


I check i have 589 hour on my ac 160 3x is 1767 so if i had my old dvx 100 a or b using tapes it might need a new drum.
 
Last edited:
I'm back, and I previously promised to do an autofocus test of my HPX250. The test was just completed and I will post again in a few days when the shots are on youtube.

Here are some preliminary observations: For the test, I fabricated a dual camera mount to run the HPX250 and a HMC150 on the same tripod. I'll edit the footage with both cameras in a split screen. I tried to simulate Rod's near/far test chart setup, at 30ft and 3 ft. Each chart was individually lit for a similar exposure.

For test 1, I matched the HPX250 zoom to approximately that of the HMC150 (around 53mm), The HPX250 was set to f 3.6 and the HMC150 was "open" at max zoom. The HPX250 acquired focus perfectly in about 10 round trips. The HPX250 autofocus system seems well damped, in that it slowly changes focus, and then locks on target and stays there. I assume the HMC150 was faster, but will have to wait till I sync the clips in the editor.

Test 2: This time I maxed the HPX250 zoom to 86mm, and now the HPX250 had less than a 50% success ratio, as it often first observed the transition target as a complete blur, and it just stayed there (no hunting). I found I could return to 100% focus hits if I moved the tripod head more slowly and let it focus on intermediate objects between the targets.

Test 3: I had to pluck a lemon of a tree and replace the near chart. Ultimately this is less scientific because of background differences due to different framing each time the cameras swung around. Again the HPX250 occasionally failed to focus, but so did the HMC150. Until I get a chance to review the clips, it seemed like the HPX250 did better than the HMC150.

I come away with the view that the HPX250 has a highly damped autofocus feedback system, and given any contrast to work with, it will find focus and stay there. This is fine with me, and matches my experience in the field.

I'll do my best to get theses clips put together, and up on you tube, by this weekend. Given the importance of this issue to many AC130/160 owners, I would like to minimize my opinions, and just contribute whatever raw data I can.
 
Terry_Martin said:
Test 2: This time I maxed the HPX250 zoom to 86mm, and now the HPX250 had less than a 50% success ratio, as it often first observed the transition target as a complete blur, and it just stayed there (no hunting).
Click to expand...

Given this scenario is problematic, what happens if you pull it yourself when AF gets confused? Will it then stay locked on after the help or will it get confused all over again as soon as the subject's distance to the cam changes (assuming a moving target) or even if the subject is static?
 
ajcourtney said:
Given this scenario is problematic, what happens if you pull it yourself when AF gets confused? Will it then stay locked on after the help or will it get confused all over again as soon as the subject's distance to the cam changes (assuming a moving target) or even if the subject is static?
Click to expand...

Aaron,
I only way I got it to lock up was by a rapid change to a "contrast free" image. With any help to a moderately defined blob, it would follow through to focused, and then stay there. I suppose a fast moving close in target could outrun the focus.

After I get the video up, I'll comment further on some conclusions.
 
OK, my HPX250/ HMC150 autofocus test is now up on youtube.

Unlisted: http://www.youtube.com/watch?v=v7SfM12G4BU&feature=channel&list=UL

This is certainly the most boring video I've ever put on youtube, at least in my opinion.

Turns out that I think Rod's near/far test is quite useful in understanding the autofocus dynamics, but I'll suggest some caution is warranted in inferring real world results from laboratory tests.

In the first test, the HPX250 zoom is set to match the HMC150 max zoom. There are 9 round trips, and the HPX250 nails each focus, but you can see it is definitely slower to respond than the HMC150 (higher damping factor). Settings 1080, 60i, lighting set to grey card 50 ire with the lens open at max zoom. Targets at 3ft and 30 ft. The HPX250 does not have the latest firmware upgrade, and is set to default PAP2.

The second test sets the HPX250 to max Zoom, and now the HPX250 frequently fails to find focus on the far target. Toward the end of test 2, I reduced the speed of the pan from near to far, so that the camera would track intermediate features along a wall, and now the camera reliably locked onto focus.

The third test replaced the near target with a lemon (nothing implied here, I have a lemon tree and not an orange tree). You could infer here that the HPX250 faired better than HMC150, but the real lesson here is that a hasty setup where the cameras are framed differently and background becomes a factor, well that can drive the outcome. My bad.

Conclusions:

1) The HPX250 has an acceptable autofocus, but it has a high damping factor (slow response).

2) Barry is right that there is some Physics involved in the large zoom range. When you go to the maximum zoom setting, and then make a "rapid" object distance change, then the newly acquired image pops in so out of focus that there is inadequate contrast to guide the control system. By panning slowly, the lens can "keep up" and maintain enough contrast to keep the control system going in the right direction. This is an interesting result that helps describe the lens function, but a circumstance that would likely never show up in my work.

3) The slow response of the autofocus works fine for my work, but I could see circumstances that would make it difficult to track a fast moving target, like a running horse.

So what does this mean for the AC130/160? By reviewing Rod's near/far test:

http://www.youtube.com/watch?v=Vsjvns8YKjc&feature=player_embedded

I see a similar slow response but his camera started tracking in on the near target, and then stoped before achieving focus. The HPX250 plowed along, but always got there once it started. Also, I didn't see anything in my tests or field experience that would explain jumping off a focused target for no reason. From these issues, I "speculate" that there may be some differences in the way autofocus is controlled in the HPX250 vs the AC130/160.

Final Words:

The HMC150 definitely focuses much faster than the HPX250, but now I understand why I liked the HPX250 autofocus much better than the HMC150 when I first used it in the field. Turns out I shoot run and gun live music, and have lost a lot of HMC150 shots when the subject moved and a mike stand appeared in between me and the subject. Result, a nicely focused mike stand. This has happened many times with the HPX250, and the focus stayed locked on the subject.

I commend the folks in this thread for addressing this issue with real testable experiments. You got me interested even though I don't have an AC160, and as a result I now know more about my own cameras.
 
Hi Terry

That must have taken a fair amount of time and effort and it's much appreciated I'm sure. Rod will also appreciate the effort too. I guess one simply has to work around the issue ...It will be interesting to see what ELB614 has to say about the horses issue too...I guess the HPX250 might also battle at full zoom on a horse running towards you??? Last night, (purely for interest) I was shooting wedding speech cutaways on the HMC82 second camera at about 50% zoom and the camera totally lost it as well!! That's only 1/4" chips and 12X max zoom ..the EVF just went a total blur and locked up!! A crash zoom to full wide and then zoom in again solved the problem so it does go to show that even the older cams do it (It was the first time I had it do that!!!)

For those comtempating a change to the JVC HM600 cameras, I see my supplier has just listed them and they are substantially more expensive than the 160A by nearly AUS$2K!!!! I'll stick to Panasonic ..been with them since 1989 so I'm a dedicated user!!

Chris
 
So this extremely simplistic test video is supposed to support a conclusion that the camcorder's autofocus performance in real world shooting scenarios is going to be, as you conclude, "acceptable"??

That would appear to be what you are saying.
 
@ LanceOregon

I'll stress again what I said originally "I'll suggest some caution is warranted in inferring real world results from laboratory tests".

Rod's and my tests confirm two things. 1) The AC130/160/HPX250 auto focuses much slower than a HMC150 class camera and 2) the AC130/160 behaves differently than the HPX250 in achieving final accurate focus.

One fact often overlooked in these forums is that video production is a big world and we all may have widely varied subjects and technical challenges. I'm in a group that relies on autofocus, and a heavily damped servo (ie slow) is an asset for me, ... but it is not so great for what ELN614 needs to do. Also, the performance of the AC160 in Rod's video would not be acceptable to me.

My intent is to figure out how the gear works, and then we all can pick what is useful for our needs.
 
Terry as an HMC150 / AC160 owner I would certaily agree with your first point but not necsasarily the 2nd. The AC160/HPX250 have the same AF system according to Panasonic, so any differences in AF performance could be due possibly to individual tester technique or manufacturing variances in assembly.
Can't help wondering what would happen if you actually took an offending cam back to Panasonic and had them "blue print" the lens assembly. I know precision assembly, and know that mass production often can mean making "compromises" in the name of "manufacturability". Been there, done (as in seen first hand) that.
 
I too, have the understanding that "the AC160/HPX250 have the same AF system". If this is indeed true, then one would expect to see an HPX250A.

If not, why AC160A but not HPX250A?

Most instrumentation does not perform very well at its limits (going full zoom to a close object). Terry and others seem to have demonstrated that slow panning avoids focus lock-in failures. This type of testing (fast versus slow), seems to take contrast at the near or far target out of the equation.

In the testing that I did, it seemed there were significant differences between AREA ON versus AREA OFF.

I cannot imagine in the manufacture of any precision device that . . ..mass production often can mean making "compromises" in the name of "manufacturability".
 
Update;

following further control testing by Panasonic Service Company here in New Zealand, and tests duplicated on my camera at their request it has been assessed as having an identifiable fault and is going in for service work to correct.

They emphasize that it is not a generic fault with the model and is demonstrably specific to my camera.

Just to refresh, I never had the searching/hunting focus that ELN614 had. My 160A always nailed focus well.(allbeit very slowly ) It was just the occasional focus lock up on far/near shots on full zoom as shown on my initial trials that highlighted the issue.
 
Last edited:
You must log in or register to reply here.
Back
Top