AG-AC90 starts shipping tomorrow. How many have one on order?
- Thread starter LanceOregon
- Start date
ChrisHarding
Well-known member
Hi Zim
Barry has already gone into specifics about the camera's sensitivity and it has roughly the same sensitivity as the HMC40 but the image noise at high gain is a lot cleaner. It could very well do better than the DVX in low light simply due to the fact that the DVX technology is old on the gain circuitry so despite the bigger CCD's any gain added will be a lot noiser ! Once the cameras start getting used maybe someone will do a comparison test for you??
Chris
Barry has already gone into specifics about the camera's sensitivity and it has roughly the same sensitivity as the HMC40 but the image noise at high gain is a lot cleaner. It could very well do better than the DVX in low light simply due to the fact that the DVX technology is old on the gain circuitry so despite the bigger CCD's any gain added will be a lot noiser ! Once the cameras start getting used maybe someone will do a comparison test for you??
Chris
The DVX100b is actually pretty good in low light. Much better than the original 100 which I also own. If the AC90 is as good or better than the 100b then I will consider it. Weddings are my bread and butter and low light performance is the number one priority for me. Canon will get my money next year when I finally upgrade unless the AC90 proves that it's as good as the XA10.
Here are a few grabs:
First one, gamma HD NORM, 1/60 sec, f 11.0, knee auto, matrix NORM1:
Second pic:
GAMMA CINE_LIKE D, 1/60 sec, f 11.0, knee auto, matrix NORM1:
Third Pic:
gamma HD NORM, 1/60 sec, f 11.0, knee auto, matrix NORM1:
Fourth Pic:
GAMMA CINE_LIKE D, 1/60 sec, f 6.7, knee auto, matrix NORM1:
5th and last Pic:
GAMMA HIGH; 1/60 sec, f 1.5, GAIN +30db, DRS (dynamic stretch) 3, DETAIL CORING +7, KNEE AUTO, MATRIX NORM1
As to the last pic (above), I thought the low light was impressive for 1/4.7 inch chips. It was very dark in that hallway, I was standing in my office, my office lighting was all there was illuminating the foreground, and you can see the warehouse lighting leaking through the door at the other end. But what do I know, my other cam is a Sony PMW350 with F13 sensitivity
rogar-Happy(DBG):
(I like this little cam! It's a definite improvement over my other little cam, the TM900, particularly in the areas of control of flare, very clean and organic gradations and texture, natural color and seemingly good dynamic range, sensitivity and latitude. The daytime shots could have used external ND which I did not have time for, but they don't seem inordinately softened by such a small iris setting. The six blade iris is another improvement as is the hybrid stabilization. As to the HMC40, sorry can't help you there. Decide for yourself. I'm a Sony guy if not brand neutral, but I think this little AC90 is going to serve me well and cut with the PMW350 just fine. )
Edit: I posted these pics at full rez, not sure why they are resized, but...se la' vi
First one, gamma HD NORM, 1/60 sec, f 11.0, knee auto, matrix NORM1:
Second pic:
GAMMA CINE_LIKE D, 1/60 sec, f 11.0, knee auto, matrix NORM1:
Third Pic:
gamma HD NORM, 1/60 sec, f 11.0, knee auto, matrix NORM1:
Fourth Pic:
GAMMA CINE_LIKE D, 1/60 sec, f 6.7, knee auto, matrix NORM1:
5th and last Pic:
GAMMA HIGH; 1/60 sec, f 1.5, GAIN +30db, DRS (dynamic stretch) 3, DETAIL CORING +7, KNEE AUTO, MATRIX NORM1
As to the last pic (above), I thought the low light was impressive for 1/4.7 inch chips. It was very dark in that hallway, I was standing in my office, my office lighting was all there was illuminating the foreground, and you can see the warehouse lighting leaking through the door at the other end. But what do I know, my other cam is a Sony PMW350 with F13 sensitivity
rogar-Happy(DBG):(I like this little cam! It's a definite improvement over my other little cam, the TM900, particularly in the areas of control of flare, very clean and organic gradations and texture, natural color and seemingly good dynamic range, sensitivity and latitude. The daytime shots could have used external ND which I did not have time for, but they don't seem inordinately softened by such a small iris setting. The six blade iris is another improvement as is the hybrid stabilization. As to the HMC40, sorry can't help you there. Decide for yourself. I'm a Sony guy if not brand neutral, but I think this little AC90 is going to serve me well and cut with the PMW350 just fine. )
Edit: I posted these pics at full rez, not sure why they are resized, but...se la' vi
ChrisHarding
Well-known member
Hi Tom
Impressive!!! open iris and 30db is quite incredible for those tiny chips ..OK there is noise there but if I had shot the same with my HMC80 at 24db it would have been a LOT noiser!! I don't see any of the "smear" (spray paint hair effect) that the guys were worried about on low light shots....Despite the huge gain that's still a usable image!!! Meant to ask ...is the scene file setup like the HMC 40 which has just 2 files and you have to power down to change them ?? On the 130/160 it's switch selectable but I wasn't sure if the 90 would keep the old method or you could actually change scene files with the camera on.
The outdoor shots are pristine!!
Chris
Impressive!!! open iris and 30db is quite incredible for those tiny chips ..OK there is noise there but if I had shot the same with my HMC80 at 24db it would have been a LOT noiser!! I don't see any of the "smear" (spray paint hair effect) that the guys were worried about on low light shots....Despite the huge gain that's still a usable image!!! Meant to ask ...is the scene file setup like the HMC 40 which has just 2 files and you have to power down to change them ?? On the 130/160 it's switch selectable but I wasn't sure if the 90 would keep the old method or you could actually change scene files with the camera on.
The outdoor shots are pristine!!
Chris
Hi Chris
On the scene files, well...I don't know if it is as easy as it seems, or perhaps I'm missing something, haven't studied them much, but there are six of them, chosen from the touch screen. As far as I can tell, it's as easy as touching the screen to choose one, and it does not appear that the camera has to be powered down or anything special to select among them. As for the smear or sprayed hair, personally I never worried about that. I'm just speculating, but it could have been Youtube compression, jpeg compression, who knows what. That's why I posted grabs instead of online videos, the latter has the potential for getting too mucked up. The TM900 was tack sharp for me, there was no reason to believe the AC90 was going to be worse. The improvement is better than I was hoping for, the DR and sensitivity seem quite noticeably better, the video looks more professional, cleaner. The one area still flawed, is the jello cam along the edges when the optical stabilizer group in the lens shifts. The ATW white balance seems to track more reliably as was mentioned. But what I would remark, is that the value for $2k is very solid. Panasonic is offering up the best value and class leader in AVCCAM.
Zim
Veteran
I am about to put in my order with B&H and there are few more tests I would love to see from the pros who have one already (and have the time). I have not researched the similar models in these areas so it may be helpful to others. I would really appreciate Vimeo/”quality” ungraded footage of:
1). Scenery with motion (tilt/pan/zoom).
2). People and especially well lit faces (like for an interview).
3) Street scenes and/or objects are in motion.
4) Outdoor low light (city lights).
5) Tests of aliasing where you pan camera right & left on tripod while focused on a vertical bar(s).
6) Sound tests with a professional microphone in. I am particularly interested in the pre-amps and noise levels (especially silent). Oh and sound with the various internal microphone (especially that zoom function). The 5.1 I don't think I would use.
7) It would also be interesting seeing one shoulder mounted via on a rig and the resulting hand held footage.
8) I am really curious about with the built in stabilization and how a steadicam type device impacts the footage.
9) Oh and a dynamic range test would be good. So shooting directly at a car head lights at night. Or out a well lit window. A still would work great for this. Not expecting a great result but who knows.
10) A still of a test pattern would be very interesting, but I assume folk have few of those lying around. Especially testing the lens at different zooms.
11) Green screen would be also interesting if it is even very doable.
12) Is there a way to test the built-in ND? A light on a dimmer?
Anyone else have tests they would like to see? I know all the above would be awesome if tested against another known camera at the same time, but that of course takes quite a bit more time.
1). Scenery with motion (tilt/pan/zoom).
2). People and especially well lit faces (like for an interview).
3) Street scenes and/or objects are in motion.
4) Outdoor low light (city lights).
5) Tests of aliasing where you pan camera right & left on tripod while focused on a vertical bar(s).
6) Sound tests with a professional microphone in. I am particularly interested in the pre-amps and noise levels (especially silent). Oh and sound with the various internal microphone (especially that zoom function). The 5.1 I don't think I would use.
7) It would also be interesting seeing one shoulder mounted via on a rig and the resulting hand held footage.
8) I am really curious about with the built in stabilization and how a steadicam type device impacts the footage.
9) Oh and a dynamic range test would be good. So shooting directly at a car head lights at night. Or out a well lit window. A still would work great for this. Not expecting a great result but who knows.
10) A still of a test pattern would be very interesting, but I assume folk have few of those lying around. Especially testing the lens at different zooms.
11) Green screen would be also interesting if it is even very doable.
12) Is there a way to test the built-in ND? A light on a dimmer?
Anyone else have tests they would like to see? I know all the above would be awesome if tested against another known camera at the same time, but that of course takes quite a bit more time.
magicdavek
Well-known member
Like Zim, I'm interested in the AC-90's low light performance vs. the DVX100. I have the DVX100A, not B. I didn't realize there was a difference between the A and B on low light performance, as was implied above. I thought the B was made from more environmentally friendly materials but otherwise the same. The 30dB shot, though noticeable noise, was much less noise than the DVX100A at 12dB.
Michael Reichmann
New member
A comparison of the C90, lens wide open, shutter at 1/50th; at 0db, 18db and 30db, top to bottom. Fascinating.
No post processing – straight screen grabs from FXP-X.
No post processing – straight screen grabs from FXP-X.
Michael Reichmann
New member
None at all. It's AVCHD which FCP-X handles smoothly.
Michael
Michael
Zim
Veteran
Thomas Smet
Veteran
Anyway to see non scaled versions of the grabs?
Look very good so far. Really good actually.
Look very good so far. Really good actually.
LanceOregon
Well-known member
Where is the noise? There should definitely be noise visible at 30 dB.
Barry_Green
Moderator
Working on it.
There is basically NO aliasing. I mean, I tested this thing on The Wringer, recording to an external NanoFlash, and it delivered the best results I've ever seen from any camera I've ever tested on that chart. It's incredibly clean.
Will try with the SteddiePod. In general I don't think it's a good idea to use two stabilizing systems together, as they can end up fighting each other and causing complications.
There's no need to test it, it's just a built-in, seamless part of the exposure system. Any time you have an iris smaller than f/2.8, the ND filter will be engaged to some extent or other.
Barry_Green
Moderator
The DVX100B had a substantial reduction in noise as compared to the prior DVX100A. There were a number of changes from the "A" to the "B"; I did an article on it here at the time.
I haven't compared a 100B to the AC90, but when I get back in the States I can give that a try.
Barry_Green
Moderator
Yeah, that's the wacky thing -- there really isn't much noise at all. Even at 30dB. Unless you crank the detail up a lot; if you set detail to +7 and then 30dB, you'll see noise, but even then, a lot less than other cameras.
I attribute it to some pretty aggressive noise reduction. The improvement in the AC90 vs. the prior HMC40/80 is so significant, at high gain levels, that it's kind of magical -- and I'm not much of a believer in magic, so I'm attributing it to aggressive, substantial noise reduction. Which explains why the 30dB images can get a tad softer and "mushier". 30dB is still surprisingly usable, but in practice I try to keep it to 20dB or lower, to avoid the mushing that can happen at higher gain levels. However, if you need to get the shot and you need 28db or 30db to get it, go ahead and get it -- you'll be surprised at how good it looks, as compared to how you might expect it to look! On the HMC40/80, I really didn't like to go to +12dB, that was where it got too noisy for my taste, and +24dB was a hailstorm of noise. On the AC90, +24 looks pretty good, all things considered. And +30dB is usable. But you don't really see what you're giving up unless you see them back to back, 0dB vs. 30dB, and then you can tell the colors are softer, the detail is a tad mushed, etc.
So in general noise reduction seems to be quite effective, but one place where it can come into play is in shadowing/trailing artifacts. You might see some trailing shadows on fast pans on 30dB footage. Can't expect the world for a $1995 price, so I'd say try to avoid 30dB when you can, and if you have to go 30dB, try not to "blair witch" the camera too much...