xf-300 vs sony EX1R sensor size?

[Steve Phillipps]

"it does seem that in picture quality terms the HPX370 is beaten overall by the EX1" Does it? What's that impression based on? Not saying you're wrong, just haven't seen any evidence that clearly shows that.

"if all else is equal, a 1/2" will have a 1 stop advantage over a 1/3" camera" That's definitely a relevant point - all else is rarely equal, and as Barry Green's tests showed, there's actually very little in it. Not saying that I trust Barry Green more than anyone else, but I do get the feeling he's a fair commentator, and above all his is the only comparison I've seen!

"There are reports that a difference between the 300 and the 370 is that the 370 gets the noise level down by some form of electronic noise reduction" The only suggestion I've seen of that has been by me!!! I mentioned that in response to Andy Shipside's comparison but got no response. It certainly did seem to me,as you say, that the image was less noisy but a fair bit softer - just as you'd get if you put a still through Noise Ninja.

I really feel that a lot of people discount 1/3" needlessly. Even the depth of field argument is often not that big a deal. For example for most documentary stuff any shot you'd do at f4 on an EX1 will look the same on a 1/3" chip camera if you set it f2.8. It's only for drama focus pulls/shallow dof that it might be a bit of an issue IMO.

I think IF you've got a lens that you are happy with from f1.8 you've got a useable range of f1.8-f5.6 before you'll get any noticeable diffraction effects, so as long as you've got enough NDs to allow those wide apertures in your lighting situation then there's not a problem.

Just my opinions though, I fully respect anyone's opinions and disagreements!

Steve

[David Heath]

Originally Posted by Steve Phillipps

"if all else is equal, a 1/2" will have a 1 stop advantage over a 1/3" camera" That's definitely a relevant point - all else is rarely equal, and as Barry Green's tests showed, there's actually very little in it.

By "all else equal" I was really thinking of such factors as CCD or CMOS? Total no of pixels on chip? Do the chips have microlenses? It's very difficult for manufacturers to defy the laws of physics as they apply to silicon - it's a bit like aircraft design, manufacturers put the numbers into computers and tend to get the same results, which is why all airliners tend to look much more similar now than they used to.

In this case, both the EX and the 300/370 are CMOS and 1920x1080. I believe the EX chips use microlenses, the 300/370 doesn't, so combined with the chip size difference the EX would be expected to have more than one stop intrinsic difference.

Now maybe Panasonic have access to alien technology which would explain why Barrys results don't match expectations. An alternative might be down to lineup - quite a few factors could affect the noise level, and it's possible to degrade certain performance factors to favour noise. Another alternative is if active noise reduction is used, which is down to the same thing - degrading other aspects of image quality in favour of reducing noise. So in this case "all else is not equal" benefits the HPX370 in one regard, diminishes it in another.

I really feel that a lot of people discount 1/3" needlessly.

It used to be a case of going for 1/3" because you couldn't afford cameras with bigger chips. A general acceptance that 1/2" was better, 2/3" better still, but hey! I can't afford them and 1/3" is good enough for me. Alternatively size/weight was a crucial factor so 1/3" it had to be.

The EX changed all that by giving 1/2" chips in a camera at a price/size/weight that previously had been that for 1/3". It's ceases to be a case of quality versus price - it's one of why settle for less if you can get better for the same price?

[Steve Phillipps]

That all seems reasonable David, you may well be right. But until you've done tests it's just supposition as to which is better. Barry's done that and I have no reason to doubt him.
Steve

[Brian Woods]

This is all relatively moot, because the fact is, all things are not equal. Sensor size (in terms of 1/3 vs 1/2) is a tiny tiny part of the whole pie. Here's the difference between 1/2" (pink) & 1/3" (yellow) sensors:

At that level, lens quality, codec, processing power & implementation and a host of other factors play a much more crucial role in determining overall IQ - and that's why Panny's HPX300's & 370's have been so successful at securing a majority market share in broadcast television vs any of the Sony XDCAM line (ex1, ex3 & even pmw350).

If bigger is always better, Sony should be winning the game. But they're not, and there's a reason for that.

Originally Posted by Cristian Adrian Olariu

Yes, you are right on the outdoors usability of Sony LCD's. But you don' know how good is the Canon's LCD in practice, it's only an assumption that is better in bright conditions. We will have to wait and see.

I was able to use the Canon XF305 outside in bright light. The LCD performed very very well, although I always prefer to use a hoodman or the like. However, it is, without question, the best LCD on any camera that I've had the pleasure of using. It really is a step up.

[David Heath]

Here's the difference between 1/2" (pink) & 1/3" (yellow) sensors:

The fact is that in terms of area a 1/2" sensor is twice as big as a 1/3" sensor. For chips of each size, each 1920x1080, that means each individual photosite will be more than twice as big in the 1/2" chip case. (It'll be more than twice because it's reasonable to assume the inter-site border will be the same in each case, hence takes up less percentage wise in the 1/2" case.) That means twice the light gathering power/photosite, and therefore an inherent advantage of at least a stop if the chips are otherwise the same.

No matter how good or bad a manufacturer may make a camera with 1/3" chips, the same manufacturer making the same camera with 1/2" chips will inevitably be better - a lot better. You only have to compare Sonys EX with Sonys NX5. (They keep the photosite area the same, but the NX5 has about 1 megapixels/chip compared to double that for the EX.)

Hardly surprisingly, manufacturers who only sell 1/3" designs tend to disagree with that - at least in public. It's marketing, not science.

[Mark Andersson]

Oh you make me happy Brian on the confirmation of the LCD. Im going to 100% buy this beasty XF300.
The EX1 is just too awkward to use hand held days on end.

I will mainly shoot in 35mbs and for special occasions i.e very low light ive got the option to go 50mbs!!

Im still praying Canon will bring out some firmware to allow Push-Auto-Iris on the handgrip hehee :)

[Peter Moretti]

The DOF argument goes in favor of the EX-1. But the LCD w/ waveform monitor and vectorscope is going to make composing a good, well lit shot quite a bit easier on the Canon.

[Leon Lorenz]

I wonder why Canon didn't go with 1/2" chips for their 300 / 305 models as it would of pleased more people for sure. I used an Canon L1 in Hi8 way back in 1991 that had a single 1/2" sensor that was better in low light than the XL1 or the XLH1 that I used since for wildlife filming.

As handy as this camera will be to pack around, the lack of lens power won't cut it for most wildlife shooting I do. I'm hoping for their interchangeable model soon ( anyone have an idea when? ), maybe they will surprise us with 1/2" chips. In the meantime I'll keep using my XLH1 that has never missed a beat.

Leon Lorenz
Canadian Wildlife Productions: Grizzly Bears, Bighorm Sheep in Alberta & BC Rockies DVD Videos

[Ronan Fournier]

Well, maybe because 1/3" chip sensor allows long range zoom for an affordable price. As a wildlife cinematographer too, I prefer a 1/3"chip with a 18x zoom than a 1/3" chip with a 14x zoom.
Of course, an interchangeable lens would have been even better, but I bet that, if Canon make one, it won't have a 1/2"chip, because Canon wouldn't want to redesign the zoom.
I also keep my XLH1 with a NanoFlash and it rocks!

[Steve Phillipps]

Agreed Ronan, and it's not just lens price it's also the size. A 100-300 f4 Sigma is a lot more compact than a 200-600 lens that would be the equivalent on a 2./3" sensor. Bigger lens, so bigger tripod too - before you know it you'll be in hospital with a broken back!
And with that in mind I'd suggest that the wildlife community should be crossing their fingers that Canon don't put 1/2" chips in an interchangeable lens model.
Steve

[Brian Woods]

Originally Posted by David Heath

No matter how good or bad a manufacturer may make a camera with 1/3" chips, the same manufacturer making the same camera with 1/2" chips will inevitably be better - a lot better. You only have to compare Sonys EX with Sonys NX5. (They keep the photosite area the same, but the NX5 has about 1 megapixels/chip compared to double that for the EX.)

I think you're absolutely right about this. However, the same argument doesn't work when comparing cameras from separate manufacturers because of how many variables are involved in ultimate performance. That's why any argument stating that an EX1 is better than a XF300 because the EX1 simply because it has a marginally larger chip size just doesn't hold water. Truly, both cameras have pros and cons over each other - testing has shown that the EX1 has about a 1 stop light advantage over over the XF series, naturally due to it's larger sensor and photosites. A one stop light advantage does not necessarily make the EX1 a superior camera, unless all you care about is low-light ability (some may argue a shallower DOF feature of the EX series as well, but in my experience, there is no easily perceptible difference in DOF between 1/3" & 1/2"). Early images suggest better dynamic range & more resolution in the XF line (and certainly more color information via 4:2:2 out of the box). There's a lot of factors to consider here.

A lot of people/productions have very different needs when it comes to a video camera. In this day and age where nearly all cameras are way past "good enough", bigger is better regardless of all else is a failed argument.

Going back to the original poster's question about the sales guy pushing the Sony over the Canon b/c of the sensor size - well, from my experience working in sales, there's one simple reason for this: that sales guy can sell a Sony right now, today. He can't sell the Canon. Any decent salesman will always push a product he can sell immediately over one he can't sell for the next month or two. Which is just one of the reasons why I think it's very dangerous to take buying advice from a salesman (but that's getting into a whole other can of worms)!

[David Heath]

Originally Posted by Brian Woods

That's why any argument stating that an EX1 is better than a XF300 because the EX1 simply because it has a marginally larger chip size just doesn't hold water.

Firstly, I wouldn't call the difference "marginal" - 1/2" chips are twice the area of 1/3", and the practical difference is more owing to the likelihood of the inter-photosite boundaries being a lower percentage of the total in the bigger chip case.

In principle you're absolutely correct, you can't say that any camera is necessarily better than another purely because of chip size. BUT, for two cameras with the same underlying technologies (no of pixels, CMOS/CCD, 3 chip/1 chip etc) the manufacturers of the larger chip version are starting off with a huge advantage in design terms, and whilst desirable, I don't think the 50Mbs codec will be anything like enough in itself to make up the difference. Whether Canon can pull anything else out of the bag remains to be seen - it's not impossible, but I'll believe it when I see it.

Whatever happens in practice, however good the XF300 turns out to be, there's no getting away from the fact that it would have been far better with 1/2" chips. If I'd heard that, it would be my clear expectation that we'd have a likely "EX1 killer", as it is we'll have to wait and see......

[Leon Lorenz]

I have to agree with David. All things being equal the camera with 1/2" sensors should perform better in low light than a 1/3'' camera. Also wouldn't the larger sensor camera have a larger subject size at the same lens power? This shoud be a big plus for seeing and keeping fast moving small wildlife ( like flying birds ) in your viewfinder easier. I believe if Canon made a 1/2"sensor camera or larger they would have a huge market, I for one would buy it in a heartbeat. I looked at Sony's EX3 about a year ago and it looked very cheaply made compared to the Canon XLH1 which I now use.

Leon Lorenz
Canadian Wildlife Productions: Grizzly Bears, Bighorm Sheep in Alberta & BC Rockies DVD Videos

[Brian Woods]

Originally Posted by Leon Lorenz

Also wouldn't the larger sensor camera have a larger subject size at the same lens power? This shoud be a big plus for seeing and keeping fast moving small wildlife ( like flying birds ) in your viewfinder easier.[/url]

No, other way around. A smaller sensor will give you a larger subject in the frame @ same 'lens power.'

Originally Posted by David Heath

Whether Canon can pull anything else out of the bag remains to be seen - it's not impossible, but I'll believe it when I see it.

In regards to Canon pulling some other magic out of the bag, I will say this: When I was playing with the XF305 at Cinegear, I zoomed all the way in (18x), pointed at some high contrast vertical lines, and whip panned away at various speeds. I couldn't see any rolling cmos skew in the LCD. I'm sure there's some there, but compared to what I'm accustomed to seeing in the HPX300 & and the EX1/3 - massive difference. i don't know what they're doing, and i haven't heard a whisper of 'global shutter,' but I'm just saying there seemed to be far less rolling shutter than any other professional cmos camera I've ever used.

Also, I'm not convinced it's an EX1 killer - in fact, I don't think it is. But I would put it on par, as the EX1 certainly isn't an XF killer. And I can certainly see why some would choose it over the EX1.

[Tans Mark]

another compare video and some more:

YouTube - CANON XF300 ?SONY XDCAM PVW EX-1 test.wmv

http://www.youtube.com/user/2009keiko#p/u

not mine, etc