or Sony RX10: What we've been looking for from Panasonic?

damn you all, i won't get mine in my hands for another day and a half...

i did have an annoying realization though, that the 1300 for the rx10 plus selling my 5d2 body would equal an on sale 5d3 body..... and i have lenses.... and 5d3 ML raw is ridiculous.... and theres the 3x crop mode...

ok, bring me back to my senses
 
dustylense said:
Camera just got dropped off. Pulled out of box and did my standard aliasing test. All IS features off set to 60P. I'M AMAZED. I shoot the house across the street that is flanked on all sides with pine trees to check for aliasing. Upon first inspection (charging battery so I had limited time) I see near zero aliasing! Sharp! Detailed! Also, the camera does not clip at 100%. Check out the waveform on my over exposure iris ramp. So far (again limited by battery) so good.
View attachment 79179
Click to expand...

Many thanks for the good report dustylense. Looking forward to your further RX10 adventures.

Shooting my favorite aliasing proving ground tomorrow & Grand Canyon next week. If the RX10 succeeds at both locations, it'll be a keeper.

Best of success...
 
Darren Levine said:
damn you all, i won't get mine in my hands for another day and a half...

i did have an annoying realization though, that the 1300 for the rx10 plus selling my 5d2 body would equal an on sale 5d3 body..... and i have lenses.... and 5d3 ML raw is ridiculous.... and theres the 3x crop mode...

ok, bring me back to my senses
Click to expand...

Makes me nostalgic for the days when good, better and best were our only choices :)
 
OK kids. Was able to play a bit with the RX10 today. But not as much as I would like to have. But so far I think the camera is pretty nice. For a camera with a 1 inch sensor it does a darn good job. I shot in 60p highest quality. No 24p testing today.

THe camera is able to shoot a detailed image. It does alias though. But it's highlight aliasing where fine highlights meet mid tones or shadows. Most cameras do that. Hell, even my C100 will pick it up once in awhile. The aliasing reminds me of the Gh3's highlight aliasing. It's there but I've seen FAR worse in a camera in this price range.

I shoot the neutral profile with contrast at -3 (all the way down) and sharpness at -3. Just didn't have time to play with profiles. So I can't really say if the codec is good or bad because I'm not sure if another profile would shoot cleaner. I shot mostly at 640-800 ISO. (So we have to do an ISO test and find the sweet spots).

So I was doing some real estate for a client and while I shot some slides with the C100 recording to the Ninja I did a few slides of the same shot with the RX10. On the C100 I am using Cowpunks profile Rec709 (see C100 section to understand that) and recording at 30P. On the RX10 I shot 60p and am dropping the footage into a 30p timeline. I matched the kelvin temp on both cameras at 5400. My C100 profile hues toward greenish and the RX10 hued toward magenta. I did minimal CC to each to match them in the sample I'm providing. Also. I left the RX10 in continuous AF for all shots.

So here is a dropbox to a ProRes file of 3 scenes and you will see 6 shots. These scenes are lit with 100 watt LED and some fluorescents. Lets play the game of which is which. ****(keep in mind that the C100 is recording to the Ninja)*****

(MAKE SURE TO DOWNLOAD THIS FILE. Don't stream it).
https://www.dropbox.com/s/i31b4qkt6i9ytxh/RX10 V c100.mov

Some more about the camera that I found. I shot some footage with the DR increase activated. It was 1600 ISO and I would say DON'T PLAY with the DR increase function. Maybe at a low ISO it would work, but stay away with higher ISO's.
The camera has zebras and I find it wonderful! Why cameras like the Gh3 are missing real zebras is beyond me. Peaking in the camera is also a plus. AF works pretty darn good. It's responsive anyway. THe lens doesn't show very much distortion. It's very minimal anyway. The lens is pretty darn nice.

Right now I'm playing with it hooked up to the Ninja2. (bought a mini HDMI cable on the way home). So far so good. So even if we find the codec does suck, at least there is external recording be it 422 8bit. I do shooting a lot of 30P and so I gave the MP4 setting a try to see if I could record it's signal out to the Ninja. It works and does record at 30p but I've not read anywhere if it passes uncompressed in that MP4 mode. I'm gonna assume it does.

More to come but so far the camera is nice. We'll have to discover it's sweet spots more through user discussion to see if this camera is what we truly sought after. Again, I didn't have the time I wish I had to play and test it today. But lets just say that for a 1 inch sensor camera with a lens that reaches 24-200 constant 2.8, I'm impressed. It could make for a nice run and gun camera or maybe even a replacement for those shooting 1/3 chip cameras. I look forward to spending more time during the holiday discovering it's potential. I feel it's there.
 
Thank you for posting the footage, Dusty. I'm not sure which is which because I don't know which lens you used on the C100, but I prefer the second shots of the 3 scenes. The biggest difference for me was in the third scene, where the second shot brought out more detail in the orchids. Did you match the depth of field in both cameras (eg., the RX10 at f2.8 means the C100 should be at f2.8 * 1.8 = f5.04)? Did you turn off SteadyShot in the RX10?
 
noirist said:
Thank you for posting the footage, Dusty. I'm not sure which is which because I don't know which lens you used on the C100, but I prefer the second shots of the 3 scenes. The biggest difference for me was in the third scene, where the second shot brought out more detail in the orchids. Did you match the depth of field in both cameras (eg., the RX10 at f2.8 means the C100 should be at f2.8 * 1.8 = f5.04)? Did you turn off SteadyShot in the RX10?
Click to expand...
I use the Canon 17 TSE on the C100.

OK, So this morning I was testing with the Ninja2. It's nice having the option of a better codec. Good stuff and the images are super clean and detailed.

More on the aliasing and moire. Most aliasing occurs with thin highlights against midtones and shadow. Like I said, lots of digital motion capture cameras will exhibit that.
Moire. I have two homes that I use to test moire across the street. Here is a grab from a wide shot doing so.
Screen Shot 2013-11-27 at 10.30.25 AM.jpg
You see some moire on the house in the middle. But if you look at the roof left of that with the orange tiles, it's super clean. When I did this test with my Gh3, both roofs exhibited moire. For a camera that shoots such high detail as the RX10 handling the orange roof is good news. Now look at the thin line at the bottom of the sliding door on my white van. This is where you would expect to find aliasing on wide shots with the RX10. Honestly, it does a better job at handling it than my Gh3 did. (I don't have the Gh3 anymore so I can't do further side by sides). So what I am saying here is that for a Hybrid camera with a 20mp sensor, you should expect some aliasing and moire certain patterns at certain focal lengths. From my recollection of the aliasing/moire of the Gh3, I would give the advantage to the RX10.

Here is a shot of a palm against a white cloudy sky. Other than the motion blur from the grab, I will say that when watching this I was pretty impressed with the detail. No colored aliasing of the palm leaves against the sky and zero chromatic aberration.
Screen Shot 2013-11-27 at 10.31.47 AM.jpg

A shot of the front of my van. I shoot this to see highlight handling and look for aliasing along the grille. When I look at the wiper blade arm and study the "specular highlight" against the black of the arm, this is where you see the highlight aliasing I speak of. Even my C100 and other cameras will exhibit this. A Canon HDslr (like the 7D or Mk2) would totally fail this and show aliasing with colored stair stepping. The RX10 just shows what I call acceptable NON-colored aliasing and again, reminds me of the Gh3 and maybe even better.
Screen Shot 2013-11-27 at 10.31.10 AM.jpg

Here is a link to the grabs on dropbox is you want to see better quality:
https://www.dropbox.com/sh/qbjzbk7zxobfkg8/0wXcRt1bQX

Still early with the camera but I will say this.
If you like the Gh3 and accept the images from it, then you will be pleased with the RX10. Having a camera that can PACK detailed images that is so small along with a 24-200 constant 2.8 is some serious power considering what that camera is. If you accept the fact that certain patterns might exhibit aliasing and moire but don't completely kill the shot than you might find a place in your bag for the RX10. Plus the ability to record to the Ninja or the like is just another plus. Add the external XLR unit to the camera further enhances the cameras capability. Remember, this is a hybrid camera. It's taking on image quality seen out of the Gh3 (in my mind anyway). While playing this footage, if I think to myself that I'm looking at my C100 footage I can easily trick myself into think that.

Is it the end all be all camera killer? Of course not. That camera just does not exist. Not for under $20,000 anyway. Maybe not at all......

Oh and by the way....Don't forget to wipe the nose grease off your monitors after pixel peeping. One thing I learned when viewing HD footage is to BACK OFF your monitor about 5 feet just as you would when watching your television. Have you even been 1 foot away from a printed billboard? No, you are 100's of feet away. Back up off your monitors and you might just like your camera a bit more.....
 
Last edited:
philiplipetz said:
^
what about "codec breakdown" reported elsewhere?
Click to expand...

Not sure of that yet. Untouched and not acting like I'm grading RAW I don't see any issues with the codec. It's AVCHD so it's only going to go so far. But looking at some footage today with the internal vrs Prores recording, I don't see where this breakdown is happening, untouched that is.
If your referring to EOSHD's report on the codec...All I can say is this. Has he EVER persuaded you to live life and shoot any differently?

Look. I posted some footage between the RX10 internal codec and the C100 recording to a Ninja2. Other than DOF and camera profile differences, after a quick CC of the footage to mimic the other and reviewing the exported file, I had to go back and check which was which to make sure. So what does that speak of the internal codec of the RX10? Obviously not anything aggressive in CC that I did. But I also know that I'm not treating Alexa or Red Dragon footage either.

Again, this is a hybrid camera. It's still developed and focused on taking still shots. It cost $1300 and you get THE LENS! It's not a BMD Pocket Camera, but it does handle moire better than the Pocket Camera (from what I've seen). It's not a C100. It's not an FS700. And you could by how many of these compared to an Alexa? I will call the RX10 a "Pocket Gh3". How about that?
 
^
I am finding highlight falloff at default settings. have you seen it? Still testing to find workaround, but I am using a torture test designed to quickly show DR and highlight problems, not representative of real life, designed to quickly show changes in profile settings. So no "real life" tests yet.
 
My preferred settings on NEX cameras: Portrait, minimum contrast, saturation halfway between center and min, DR-O set at -1 (move it further than that and it screws the light response function)
And make sure you are getting the whole DR in the editor: in Premiere, for example, you have to apply 16-235 output levels, otherwise highlights suck big time

downloading the test...
 
Samuel H said:
My preferred settings on NEX cameras: Portrait, minimum contrast, saturation halfway between center and min, DR-O set at -1 (move it further than that and it screws the light response function)
And make sure you are getting the whole DR in the editor: in Premiere, for example, you have to apply 16-235 output levels, otherwise highlights suck big time

downloading the test...
Click to expand...
I use FCPX.
 
Dustylense, first let me say it was great of you to post your RX10 footage and I appreciate your efforts. The footage you posted was static indoor scenes with slow camera movement. I would not expect the codec to breakdown under these conditions. Outside with fast moving water, traffic, blowing leaves etc. that is what I would like to see if anyone has the time.
 
Dusty, does the camera behave well with the Ninja 2? There are no pull-down issues or the like?

I'm strongly thinking about picking one up. Doing a documentary on a bridge right now, and so following architects 15 years my junior around on catwalks 300 feet above the water... currently lugging the C100/Ninja with 18-135 and a 5d2 with 17-40 as the usual run and gun package (doing stills and video), though sometimes I go very light and bring the hacked GH2 and mk2 14-140 instead of the C100.
 
Strange, the last shot, the highlights are falling apart on the C100 and the RX10 before looks solid. Are you sure those didn't get reversed?
 
Dusty - thanks again for posting the test and analysis. It looks like I prefer the C100 footage :) Did you test the RX10 with SteadyShot on or off?
 
Mark Williams said:
Dustylense, first let me say it was great of you to post your RX10 footage and I appreciate your efforts. The footage you posted was static indoor scenes with slow camera movement. I would not expect the codec to breakdown under these conditions. Outside with fast moving water, traffic, blowing leaves etc. that is what I would like to see if anyone has the time.
Click to expand...
Can you list the number of AVCHD cameras that handle moving water, blowing leaves, traffic, high motion? Isn't that the reason we shoot higher bit rates, be it external or in a $18000 camera?
So I guess my plan for shooting the next Iron Man is out of the question with the RX10.....
 
You must log in or register to reply here.
Back
Top