RX100 vs LX7

[Glenn Thomas]

I'm looking at making the switch back to using a compact camera for video. I had an S95 a while back that I shot over 20 music videos with, and got some great results from it. Then I picked up an NEX5N and have already shot over 20 videos with that too, about half of which are still yet to be edited. It's got a great look, and I've been using some nice lenses with it including a few old Nikons, but I'm missing the freedom I had shooting with te S95. The overheating slows me down, I've had dust on the sensor that I couldn't see on the LCD, there's more to carry, I miss having decent macro capabilities, and changing lenses all the time is a pain too.

So I've been checking the RX100 and LX7. Comparing the two, the LX7 has caught my attention the most. I've looked at a lot of RX100 footage, but when it comes to shallow dof, I'm yet to see any clips that offer much more than what the LX7 appears to offer, judging by clips I've seen from that. But not a big deal anyway. I was happy with the look I got from the S95 with an ND filter attached.

Here's a comparative list I've compiled so far.

Panasonic LX7 advantages:
120fps at 720p. There's a waterfall clip on Youtube which is a good example.
3 step ND filter.
Cheaper by at least $200 here in Australia
24mm wide
Timelapse function
1cm macro
Hot shoe

Sony RX100 advantages
Better photo quality due to larger sensor
Camera is smaller
Better in low light?

Opinions anyone?

[alexander]

Low Light of the RX100 will be extremely hard to beat! Especially concidering how bad Pana with its camcorders at this

[jgharding]

Sounds like, with your list of advantages, you've already chosen which one is best for you and your way of working, the LX7!

I picked the RX100 for low light sensitivity and sensor size (DOF control), also because I know the menus and software are fast and easy... I had the HX9V til I dropped it in the sea, and it was only low light and DOF that it failed at, other than that it was an ideal B-Cam for 550D. I've never been happy with Panasonic menu navigation when I've tried them, so I thought I'd leave the LX7.

But of course some people like different things!

It all has to start with "what do I want to use it for, and how". Look at how the features stack up against your intended purpose rather than as a spec sheet and you'll make the right choice. It looks like both are great for different reasons. I love shooting in dark environments and natural light, so the RX100 took it because of that, as well as my positive experience with the HX9v.

A lot of people have been mentioning the LX7's wider aperture at the long end, forgetting that this is completely written off by the tiny sensor: the Sony is still more sensitive and has shallower DOF at the long end. But of course, if you need a hot shoe and a mic input and often use super close macro and ultra slow motion, AND if you don't always want to shoot at 50p or 60p, then there's only one choice here, the Panasonic.

[Glenn Thomas]

Low Light, ok, that's a good reason. I'm yet to find any decent low light footage shot with the LX7, so it's hard to know. The only clip I've been able to find is this

Pretty badly shot, but doesn't look too bad. Dynamic range would probably be better with the RX100 too. I'm still undecided, but that 120fps option would be nice. If it were just used for a few random shots here and there, could easily be uprezzed to 1080P.

Your "what do I want to use it for, and how" makes perfect sense. Although to be honest, my only real requirement is a camera that can shoot video at 1080 50P minimum, with either manual control or some kind of exposure lock. Anything else would a bonus, so I thought it would be nice to get a few opinions here on what additional features others find most appealing. Which I appreciate, thanks!

In fact, if Nokia announce a Pure View Windows Phone 8 device at the Microsoft conference next week, I might even consider that.

[jgharding]

That LX7 video looks pretty nice, it's always hard to know which is right til you try them out though! Such is the most annoying part of these decision :S There doesn't seem to be much footage of the LX7 used in anger on proper projects...

I like the sound of 720/120p though, that's pretty cool from a little camera!

All the shots of the guy with backpack walking and running around in this video are from the RX100, as are some others like the birds flying away at the start. The band is all 550D though, as is a lot of the other narrative material.

https://vimeo.com/48174944

I was really pushing it, running about with it and shooting high-contrast shots. It still has the small-chip 'look' to it -- kind of flat -- but it cuts together with the 550D OK. You can still spot the signature Canon milky movie look though ;)

[Glenn Thomas]

True, that footage does look quite nice. If only it had been shot manually with the contrast, saturation and sharpness reduced. There are parts where the exposure looks good for a second, but is then lost. These companies really need to hire a few good shooters and editors to show people what the cameras are capable of. Every dodgy bit of test footage posted online by people who don't know what they're doing could easily reduce the number of sales of the product.

Nice video you've done there too JG. Was the shot of the guy with the train behind him shot on the RX100? If so, how far were you zoomed in there?

[XXX]

Damn, you did it again, great video JG!

@ Glenn
By music video do you mean live recordings? Being able to add an external mic on the LX7 might outweight a not as good image compared to the RX100 in this case...

[jgharding]

Cheers! :) That shot with the train behind was the 50mm f1.4 Zeiss on the 550D. I'd guess at f2 or 2.8 with a fader ND. There's a lot of post-production scaling in there too, more than I'd usually push it, but the feel of that zoom out is more important than the fidelity of a few frames, I reckon! ;)

The first shot of him is 5500D as well, as is the last. All the rest of him are RX100, like the picking the leaf, the close up of his face in the forest, the two far away shots in the forest, looking at the train tracks (so you can see it's possible to focus pull with the RX100, though I'd love if the focus moved more quickly), and the running about shots, which do fall apart a bit under codec stress and jello, but they're good enough for a punk video with a gritty feel.

I can't say if it would have been better with the LX7 though, as I've not used it, but I'd like to try it out.

I use the Sunset profile on the Sony, with a bit of contrast, saturation and sharpness reduction. That profile is known for smoother roll-off apparently, though I've not had time to test extensively, just to use it in anger on a tight shoot! It seems to give nice colours and good dynamic range. I can't recall ever really finding the time for extensive tests TBH, though when I have briefly they're very informative

I no longer pull sharpness all the way out with the 550D, nor do I use Cinestyle or ultraflats (unless it's a super dark shot where I want shadow details, and thus I want everything out the H264 data free zone that is black) as they seem to kill the skintone and other subtle shade details. Just one of the downsides of the 4:2:0 8-bit bullshit... the more range you try and cram into the limited data, the worse the colour transitions become :(

If you want to prove it with a fun test, try shooting really flat then converting to black and white, then afterwards try shooting to black and white in camera. It's usually pretty noticeable how much nicer the black and white in-camera footage is, detail wise. The data is used for luminance only, it seems, at least on the EOS

[Andrew Reid]

Had a little try of the LX7 today.

Nice screen, best I've yet seen on a Panasonic camera. Full manual control in movie mode, but not in the HS (high speed) movie mode. That was set at 720/100p and full auto. No 120p on the PAL model. The camera is not switchable to NTSC for the even higher frame rates. There's no lower resolution 240fps either like on the larger FZ200.

Also the Samsung EX2F is worth a look. Though the 1080p mode is only 30p not 60p, and the image quality at the higher frame rates (up to 480fps) is poor.

The RX100 of course wins on image quality for me. 1/1.7" cannot compare to that larger 1" 20MP sensor, even though the lens on the LX7 is lovely and much faster at the long end.

[Glenn Thomas]

XXX, no actual music videos. You can check my vimeo link. There's probably 30 or so there by now.

JG, yeah, I thought that shot looked a bit too good to be the RX100.

I use the sunset profile on my NEX5N also, having read Andrew's post here about that a while back. I reduce the sharpness too and haven't had any issues there. I don't know if that helps much with aliasing. Reducing contrast and saturation on the NEX5N doesn't flatten the image too much. It would be much the same as the RX100 I'd assume.

Still undecided on a compact. I've checked a few more LX7 videos, and noticed plenty of blown out highlights.

Andrew, 100P woud be better for PAL anyway with playback rate of 0.250. Easier to calculate although I'm not sure how the camera records the clips.

[Julian]

The LX7 records at 120/100 fps but the files are saved at 30/25 fps by default.

I have a LX7 around, will do a test later this week.

[jgharding]

It'd be good to see how the LX7 handles a rather dynamic scene. I've found that when protecting highlights with the RX100, there's still quite a bit of detail in the blacks...

One would think the LX7 would drop a stop or two to the RX given sensor size...

[XXX]

Glenn, checked your Vimeo and now understand your original post better. Inspiring what you pulled off with "just" a S95. I can see now why slomo is of importance for you. But I can also see the RX´s larger sensor helping you separate the talent, just like in your NEX5 vids.

Besides that, since you work with minimal setup, I can´t really see, why you wouldn´t want to shoot with a bigger camera. A used 550D with the kit zoom lens and ML would shine in your hands and wouldn´t break the bank.
But that´s your call and you´re doing great already.

[jgharding]

Yeah just has a look too, you've definitely got a good eye for it, it'd be a shame to totally remove the APS-C aesthetic for the sake of a little hassle. The RX100 is a kind of middle ground, a big enough sensor to create shallow DOF but in a body small and well stabilised like a compact, with 50p.

Trouble is, it's still very long GOP AVCHD footage on a smaller than S35 digital sensor, footage feels kind of electronic. I think you'll miss the APSC NEX pretty swiftly. Both is a good shout, but just the RX100? That's brave...

[jgharding]

Don't get me wrong, the RX100 has a hell of a load more detail than the NEX or EOS stuff, even with the 550D hacked upto 100mbps or so. the RX100 just processes the sensor data more effectively and probably doesn't bin so many rows of data, but actually scales, and I'll bet Sony's codec implementation is just better. But it has a very different 'feel'. It's sharp, but a little flat front to back, if ya get me.

[meanwhile]

Having used both cameras at a little, I think a lot of the above makes two classic mistakes - too much attention to a single easily measured characteristic, in this case sensor size, which you use to make overly definite statements about two cameras one of which you haven't used. because when you look away from that single stat and at the two cameras in real use, there are several factors that even things up in the LX7's favour. I.e.

 

- The LX7 has a jaw-dropping, I-can't-believe-this-isn't-a-Canon-L-Series, A+ lens. In fact, looking at stills, I honestly believe that a series of primes couldn't do a noticeably better job. But the Sony's lens, well, it's is a fuzzy-cornered B.

 

- The LX7 lens goes down to 24mm; the Sony's 28mm is nowhere as useful in interiors or for getting dramatic perspective effects. And to get even that wide you have to switch off stabilization - otherwise the Sony barely goes wider than 35mm equivalent. 24 vs 35mm is a huuuuge difference.

 

My own limited epxerience - which seems confirmed by the few comparison videos I can find - suggests that the LX7 stabilization and autofocus are more robust (I found the manual focus on the Sony to be too annoying/difficult to use because I couldn't get focus peaking to work in video mode - was this me or the camera?)

 

As for blown LX7 highights in sample videos, when people have agreed that they can't find ones shot "professionally".... this doesn't really say much about the results someone who knows what they are doing will get. Camera exposure and white balance systems are stupid and getting a good exposure is hard, which is why it is the subject of entire books. The Sony's sensor size should be an advantage, but only if the entire scanning chain is equal in every other way. If you look at empirical measurements of DR (which I could ony find for jpgs but which should be indicative) then the cameras run evenly until 1600 iso: http://www.techradar.com/reviews/cameras-and-camcorders/cameras/compact-cameras/panasonic-lx7-1089288/review/5

 

Neither camera is perfect. Both are amazing in some ways but severely limited on others - you have to *carefully* consider which limitations matter most to the work you want to do. For example, if I had to shoot in interiors like this

then I'd regret giving up the RX100's (limited but valuable potential) for dof effects, but I'd do it in a heartbeat (which is what I did.) Better DR over 1600 iso I didn't give a thought too, because I don't need it - ymmv.

[meanwhile]

...And the other factor that influenced me: rolling shutter is likely to be a LOT less of a problem. The larger the sensor, the longer (all things being equal, which is a reasonable assumption here, especially given the top end of shutter speed for each camera) the transit time for the shutter window across the sensor and the more easily that your shot will become jello flavoured.  So far, I'm definitely but cautiously pleased with what I'm seeing from my tests with the LX7 in this area. There really is a lot more to camera choice than sensor size - and sometimes, when you don't need the advantages it can bring, sensor size may impose handicaps.