New Panasonic 4K 60P Camcorder HC-X1000

I think the color sampling thing is only relevant/important on certain types of projects and for certain types of clients. Yes it would be wonderful if everything was up to broadcast spec or beyond, but for me most of the folks i work for dont know and couldnt care less about that stuff.
 
This could sell in great numbers, its a cracking little camera for the video journalism set, I can see the BBC ordering truck loads for its news operation.
 
Last edited:
Wow , really interested in this camera. I wonder if it uses the same sensor as the AG-AC90. I remember someone stating that the AC90 down sampled from 4K
I think we can answer this pretty definitively -- no, it's not the same sensor as the AC90.

The AC90 uses three sensors, each 1/4" in size, and a prism to split the light so that R, G, and B each get sent to a different sensor. Then it uses a 1/2-pixel offset to increase the resolution, effectively meaning that it's working on a 4K pixel-shifted image internally, which then gets downconverted to 1080p for recording & output.

The HC-X1000 uses one single sensor, which is almost 4x as large as each of the sensors in the AC90. That single sensor is a 4K sensor, so it's 4x as large, with 4x the res, so I think we can probably say that the pixel density is about the same between them. But the AC90 only outputs 1080, whereas this new camera outputs 4K, records 4K, and can also record 1080p from the full sensor scan. Accordingly, it could potentially deliver even sharper 1080p than the AC90 does.

It remains to be seen how the footage really compares, but in adequate light the AC90 was simply incredible for the price. This new camera has comparable pixel density, so it should do at least as well, but it's also about three years' newer technology. The AC90 came out in 2011, IIRC -- so this new camera benefits from three years of technological advances. It very well could end up being more sensitive, less noisy, more dynamic range, etc., than the AC90 has.
 
I think the color sampling thing is only relevant/important on certain types of projects and for certain types of clients. Yes it would be wonderful if everything was up to broadcast spec or beyond, but for me most of the folks i work for dont know and couldnt care less about that stuff.
If you're talking about a single sensor, the chroma resolution will be far lower than luminance resolution in both the horizontal and vertical axes.

Consequently, there's little point in 4:2:2 sampling in a camera of this type - all it will do is waste bitrate. Unfortunately "it can't just be 4:2:0!!" has become a cry in it's own right, even when it's not a valid technical point. Believe it or not, on the whole the designers do know what they're doing to optimise features for the best OVERALL result at a given price.

Secondly, the main problem with this type of camera (4k with small sensor) is down to diffraction limiting, and limited amount of iris adjustment between wide open and the image starting to soften because of diffraction.

Which is why 1/3" cameras may have been pretty common for SD, but are far more limiting for HD, and very much so for 4k.
 
So given the modest sensor size, is this cam just going to be trash in anything not out in the blinding sun? Like worthless in lower light situations?
 
So given the modest sensor size, is this cam just going to be trash in anything not out in the blinding sun? Like worthless in lower light situations?
No, that's not the issue. It won't be as good (for sensitivity) as a bigger sensor camera, but the diffraction issue is the lack of iris CONTROL or range you'll have without it going soft. Exactly the same problem as the similar Sony cameras that came out recently. With a Standard Def 1/3" camera you may be able to use it wide open and if the sun comes out stop down to f 8-11 and be OK. Keep that sensor size for HD, and you can't stop down as far without softening, probably only to around f5.6. Go to 4k and the problem gets worse still - you may only get good sharpness between (say) f2 and f2.8! You have to keep the lens virtually wide open to maintain performance.

And in blinding sun, I'd expect it to be grossly overexposed with wide open iris, hence need a lot of ND as you can't use the iris to the same extent as on other cameras. It's actually indoors in average light that it may be a case where with the lens wide open you then don't need ND.

This is not a new problem, and in consumer cameras with very small sensors it has been the case that to deal with differing light conditions then closing the "iris" has actually moved a variable ND between lens and sensor.

It's really a case of getting what you pay for, and there's a price to be paid for (small) size as well.

It's also why we haven't yet seen a pro 2/3" 4k camera. At 4k even that would only have a couple of stops of usable aperture range. Hence most true pro 4k cameras (Red, Sony F55, etc) also being s35 sensor.
 
If you're talking about a single sensor, the chroma resolution will be far lower than luminance resolution in both the horizontal and vertical axes.
In a single-sensor bayer system, this is true. This affects all single-sensor bayer cameras equally -- the Red One, the Alexa, the F65, F55, FS700, and HC-X1000. It also equally affects every DSLR ever made, from the $45,000 Hasselblad on down to a compact point 'n' shoot.

In a 4K bayer sensor, luminance resolution will typically max out at about 3200 x about 1728, and chroma resolution tops out at 1920x1080.

There is an alternative sensor type, the RGB stripe system as used on the F35 and Panavision, which delivers full 4:4:4 color, but it does so by limiting the amount of horizontal resolution it can create. It uses 5,760 pixels wide, but only creates a 1920x1080 image.

Consequently, there's little point in 4:2:2 sampling in a camera of this type - all it will do is waste bitrate.
Well, not entirely. At 4k res, yes, you're accurate here. 4:2:2 color sampling would give a color matrix of 1920x2160, and the sensor only has 1920x1080 color samples on it, so it would be a complete waste of bandwidth. But when considering 1080 recording, it sure would be nice to have 4:2:2 since the sensor is fully capable of filling out that recording space properly.

Unfortunately "it can't just be 4:2:0!!" has become a cry in it's own right, even when it's not a valid technical point. Believe it or not, on the whole the designers do know what they're doing to optimise features for the best OVERALL result at a given price.
Absolutely!

Secondly, the main problem with this type of camera (4k with small sensor) is down to diffraction limiting, and limited amount of iris adjustment between wide open and the image starting to soften because of diffraction.

Which is why 1/3" cameras may have been pretty common for SD, but are far more limiting for HD, and very much so for 4k.
Yes, this is something people need to be quite aware of. Shooting at f/11 on a camera like this is likely to result in quite unsatisfactory footage. You pretty much will need to limit the iris to stopping down to no more than, oh, probably f/4 given the pixel density. Once I get one I'll test for diffraction performance. Sometimes the engineers do crafty things, such as the AC90's integrated ND filter which gives you the perception that you can stop down to f/6.4 before diffraction comes into play (even though internally the iris is limited to f/2.8).

The good thing I see here is that they've integrated a proper three-stage ND filter to help deal with the situation. Six stops of ND should be enough to ensure that you can get decent exposure outside without running into the diffraction limit.
 
So given the modest sensor size, is this cam just going to be trash in anything not out in the blinding sun? Like worthless in lower light situations?
My guess -- most definitely not! The AC90 is surprisingly useful in lower light situations; it was a huge step forward from the HMC40. So far we've got three generations of super-tiny-pixel cameras to consider, the HMC40, the AC90, and now this HC-X1000.
The HMC40 was just superb in daylight, but was not sensitive at all and turned into a grainy mess at anything above 12dB of gain. Which, at 12dB of gain, it was about equivalent to an HMC150 at 0dB.
The AC90 greatly improved that situation. It is more inherently sensitive, and its gain is basically whisper clean, you can crank that thing up to 24dB of gain and still have a pretty good-looking shot.
The HC-X1000 has about the same pixel density, but is two years newer technology. What that means for sure, we can't yet say, but I'm optimistic that the trend will hold and that it will be innately more sensitive and simultaneously the gain will be cleaner.

Of course, there's another answer for low light -- the HC-X1000 has infrared night vision, so you can literally take footage in pitch black conditions. No idea what the quality of that footage will be yet, I'm sure it'll be the grainy green x-files night vision look, but it does bring a new dimension to "low light" shooting...
 
Thanks, guys. Intriguing camera, to say the least. I'll await footage. But I can definitely see this as something to use in docos.
 
Well, not entirely. At 4k res, yes, you're accurate here. 4:2:2 color sampling would give a color matrix of 1920x2160, and the sensor only has 1920x1080 color samples on it, so it would be a complete waste of bandwidth. But when considering 1080 recording, it sure would be nice to have 4:2:2 since the sensor is fully capable of filling out that recording space properly.
Yes, I was just presuming it was the 4k mode being referred to.
Yes, this is something people need to be quite aware of. Shooting at f/11 on a camera like this is likely to result in quite unsatisfactory footage. You pretty much will need to limit the iris to stopping down to no more than, oh, probably f/4 given the pixel density. Once I get one I'll test for diffraction performance. Sometimes the engineers do crafty things, such as the AC90's integrated ND filter which gives you the perception that you can stop down to f/6.4 before diffraction comes into play (even though internally the iris is limited to f/2.8).
Again, agreed. If anybody is in any doubt, it's worth looking at a recent report on the JVC GY HM-890 camera by Alan Roberts for the BBC - https://tech.ebu.ch/docs/tech/tech3335_s14.pdf . What's relevant here (and is equally applicable to all 1/3" cameras, from any manufacturer, not just JVC) is when he says:
Since the sensors are small, iris diffraction starts to be visible at between F/4 and F/5.6, so the lens should not be stopped down much beyond F/5.6 for the best effect.

Figure 4 shows .......Clearly significant resolution is already being lost; F/5.6 should be regarded as the absolute limit for good HDTV production, and even F/4.0 is marginal. Fortunately, the camera has good neutral filters which should be used for exposure control, and has a viewfinder warning to use filters rather than aperture when resolution loss would become noticeable.
And I'm relieved to see his figures of f4-5.6 agree with what I said in post 69! :) It must be stressed they are for 1/3" sensor size and 1080 HD. For 4k, the relevant figures are f2-2.8.
The good thing I see here is that they've integrated a proper three-stage ND filter to help deal with the situation. Six stops of ND should be enough to ensure that you can get decent exposure outside without running into the diffraction limit.
To an extent - but the real problem is less ability to deal with a given lighting level, than the useful iris RANGE. With this size sensor, for 4k you only really get an adjustment range of something like 1 stop or so between (say) wide open at f1.8 and the diffraction limiting point of f2-2.8.

Last week I found myself operating in variable lighting conditions (with no chance to swing filter once set) with ND set to give a range between about f4 at darkest and f11 at brightest - and a usable range of 3-4 stops is not an uncommon thing to require. Since this was with a 2/3" Standard Def camera, no problem. It should have been OK down to about f16-22. But with a 1/3" HD camera it would become a real problem, even if you can arrange ND for wide open to correspond to the darkest point of the scene. There are barely three stops of usable range.

And with 4k, and only about a single stop of usable range....... That's why bigger sensors have far more validity if we're talking about 4k.
 
To an extent - but the real problem is less ability to deal with a given lighting level, than the useful iris RANGE. With this size sensor, for 4k you only really get an adjustment range of something like 1 stop or so between (say) wide open at f1.8 and the diffraction limiting point of f2-2.8.
I understand exactly what you're saying, and I did find this occasionally troublesome with the 1/3" HD AC160 and HPX250. Once the ND filter is set, there's not a lot of range you can work within; totally fine for staged shots but for reality or news it could be a tad troublesome.

And you'd think it'd be worse with the pixel density of the AC90 or the HC-X1000 but -- it isn't. Both cameras change the rules a little to give you a ton more flexibility than you might think you'd otherwise have. With the AC90, they included the automatically-integrated ND filter so you could seamlessly stop down to (ostensibly) f/6.4, while the physical iris stayed at a physical diameter of f/2.8. So you had a complete range from 6.4 up to wide-open with zero image/exposure problems caused by having to swap in a filter. And, the AC90 takes it a step further because the gain is integrated into the iris ring -- once you hit wide open, you can just keep opening up more and more, and it'll start adding gain in 1dB increments. It's seamless, and vastly more usable in a real-world scenario than the broadcast standard has been (meaning, a three-stage filter wheel which you're not allowed to rotate once the shot has started, combined with an instantaneous clunk of gain if you reach for the gain switch, usually at least 3dB if not 6dB or more all slammed in at once).

Last week I found myself operating in variable lighting conditions (with no chance to swing filter once set)
Totally understand where you're coming from. All I can say is -- try it before you say it wouldn't work. As you said before, the engineers do have an understanding, so they've come up with something pretty useful to solve the issue. In the HC-X1000, the ND filter is actually seamless. I don't know how they do it, because I don't think it's electronic; it sounds like and feels like a physical filter wheel swapping in place, but there's absolutely no frame or ruin of the shot. Swapping in the ND filter looks about like an iris change. It's really vastly better than the situation with a conventional broadcast camera. It's not as seamless as the AC90 (where the filter is gradually introduced in 1/6-stop increments); it looks like you just stopped down a couple of stops on the iris. But if you're working in a range of 4 or 5 stops anyway, all I can say is -- this works. I'd be okay with it. Functionally, you basically ride the ND filter like as if it was a second iris wheel. I had no problem keeping the iris in the sweet spot of f/2-f/2.8 while in broad daylight and pointing the camera every which way, whether into shady trees or bright sky, you could easily quickly get it right in the sweet spot.

It's not perfect, but it's serviceable. If someone needs perfect 4K iris control, they're probably looking at something a tad more expensive than a handycam. For the price, I think they've approached the problem quite intelligently and provide a workable solution.

That said, I just played with it, I wasn't able to use it in the field on a job obviously, so firsthand user reports would be appreciated. But I found it far more usable than I was expecting, and I'd recommend trying it -- it may surprise you as to just how workable it is.

That's why bigger sensors have far more validity if we're talking about 4k.
No doubt. But to be fair, I was one of the guys who said we'd never get 1/4" HD cameras because the diffraction would so severely limit their usefulness that they'd never see the light of day. I turned out to be 100% wrong on that, as the AC90 has ended up being one of my preferred cameras in spite of the tiny chips. I think the HC-X1000 involves compromises, as all low-cost cameras do and as all smaller-sensor 4K cameras will. The question is whether you can work within the compromises. To be brutally honest, I think I'm more annoyed by the lack of WFM and SDI than I am with the diffraction situation, as I think they've got it workably sorted out.
 
Last edited:
You can do a lot with clean, noise free images off of the sensor. In the past, the zero gain images were pretty clean but allowed for little to no wiggle room with regards to manipulation. Without the noise to begin with, the digital side can be used to work around the limits of the optical side. I have been shooting with the PX-270 and am still in shock about the lack of noise. No Panasonic camera I have ever come across is as clean as this camera, so therefore I can use auto-gain to make up for the lens ramping. (I wish this feature was available in manual mode). This is clearly the way forward with all cameras (including broadcast) with the lens playing less of a role in the entire equation than it used to.
 
It does have three user customizable function buttons, though I have no idea if these would let you swap settings.
It actually provides the ability to customize up to 10 buttons. But, as said before, that's not really how you change the settings; it has six internal "scene file" setting slots, and you can change which scene file is active either by going into the menus, or by touching the scene file "virtual button" on the LCD.
 
Okay, officially, hands are now on -- I've got one I can play around with for a little while.

Some questions that have been asked, here or elsewhere --
1. Does it have time stamp for legal videography?
Yes, yes it does.

2. Wouldn't it be cool if it had the focus transition from the EX1?
Well, apparently it does. Still don't have an instruction manual (this is an early prototype) but there are menu settings for establishing focus transition. I don't know how to use the feature yet so I don't know how useful it might be, but I'm just reporting that it's there.

3. Is a 1/2" chip too small for 4k?
This one I can definitively say -- it is WAY sharper than the AC90. Hella sharper. I put both cameras on the 4K res chart and the single-chip 4K HC-X1000 is just vastly sharper than the full-HD, three-chip AC90 (which is, on its own, a very sharp 1080p camera.) I'll try to get around to testing it against the HPX250, but the 4K on this camera is serious.

4. Low light?
Much testing needs to be done. Initial impression is that it seems noticeably brighter than the AC90 when both are on the same iris. The AC90 was no low-light superstar, but it wasn't bad, and this seems better.
 
"4. Low light?Much testing needs to be done. Initial impression is that it seems noticeably brighter than the AC90 when both are on the same iris. The AC90 was no low-light superstar, but it wasn't bad, and this seems better."Oh God. My wallet now lives in fear.
 
It's great that you've got the camera, Barry. I always like it when somebody 'responsible' gets a hold of new equipment. If it's at all possible, could you post some camera original clips?
 
It's great that you've got the camera, Barry. I always like it when somebody 'responsible' gets a hold of new equipment. If it's at all possible, could you post some camera original clips?
Well, understand that this is a pre-preproduction unit. It's very early. Footage from this camera is not necessarily representative of what footage from the production version will be like.

That said, understand that there's roughly seventeen billion different file formats this thing records. Which may be a tad of an exaggeration... but there's Cinema 4K, and regular 4K at 60/30/24, and FHD at 60/30/24, and then all of the above at 25/50, and then there's the dizzying assortment of bitrates, and then compound that with the file formats (.MOV, .MP4+AAC audio, .MP4+LPCM audio), and then there's all the 1080 formats also available in AVCHD.

I know some of the modes are one-recording-mode only, but others of them appear to be available in all four...

So if you're looking for file compatibility, I might be able to finagle one. But if you want it for image quality assessment purposes, I'd have to ask them if they want footage released from this camera. It may not be representative of the final product, and if it isn't, then it would do no favors to anyone to get the footage out there yet.
 
Back
Top