Haven't seen this posted yet. Canon just released an update to the XF-line

Full press release at New Canon XF200 and XF205 Camcorders at DV Info Net (http://www.dvinfo.net/news/new-canon-xf200-and-xf205-camcorders.html)

Canon U.S.A. : Professional Imaging Products : XF205 (http://usa.canon.com/cusa/professional/products/professional_cameras/hd_video_cameras/xf205)

B&H product page: http://www.bhphotovideo.com/c/product/1043891-REG/canon_9593b002_xf200_hd_camcorder.html

I'd imagine next up is the XF400/405 probably in the Fall maybe? These specs do give an idea of where Canon would take future products with DV DIGIC IV.

Canon XF205 Professional Camcorder - YouTube

From the link:

Looks like a loaded camera overall. Can't find info on the sensor but it has the picture profile Wide-dr it seems and also 3 Card Slots! MXF + AVCHD formats. I'd imagine 1080/60P in AVCHD.


"26.8mm Wide, 20x Optical Zoom Lens with 3 Separate Lens Rings
AF options (Instant AF, Face Detection) and Reduced MOD
Enhanced Image Stabilization Performance
Ergonomic, Rotating Grip
Equipped with Wide DR Gamma (600%) Mode and Refined, 8-bladed Circular Aperture
Infrared, High-sensitivity Shooting Function with Enhanced Exposure Performance
Professional 0.45-inch Color Viewfinder and Precise 3.5-inch OLED Panel
MXF and MP4 Dual Codec Recording
Slow and Fast Motion Function Supported (MXF)
35 Mbps 1440x1080 Recording Supported for Broadcast Usage (MXF)
Three Memory Card Slots
HD-SDI Monitor Output, 3G-SDI Output, Time Code and Genlock
5 GHz and 2.4 GHz Dual-frequency Built-in Wi-Fi® Capability
Ethernet LAN Terminal
Browser Remote, FTP File Transfer (MXF/MP4) Functions Featured
Four-channel Audio Recording"

It says "1/2.84-inch sensor of the XF205, XF200, XA25 and XA20" in the brochure, so I'd take that as 1/3" sensors.

Not bad at all.

20x zoom. Built in slo mo.

I like it.

It looks like the best of the XF100 and XA20 worlds together in one camcorder. Though, at an even € 800 higher price than these two cameras together! Pretty expensive extra zoomrings.

Very practical.

So it's pretty much an XF105/305 hybrid with a few extra features.

It's exactly what I need and have been looking for, a more portable XF300. The problem is I would have to sell both my XF100 and XF300 to be able to afford an XF200. In the UK the price is about £4000 for the 205.

BH has it for $3,499.

Canon XF200 HD Camcorder 9593B002 B&H Photo Video

Curious to see the 5-axis stabilization in use and also how this matches up to the XF300. in terms of matching cameras for say a 3 camera wedding shoot.

Wow. A rotating grip like the EX1 and EX3. Funny how Sony bagged it in the PMW-200 but brought back in the 300 after feedback. Nice.

It does look good and also makes me wonder about an update to the XF300/305. All the speculation I've seen about that seemed to say "don't hold your breath" but perhaps one will be forthcoming.

So, the XA20/25 sensor / processor in an XF100 body? Is that a smaller or larger sensor than the XF100? With some features from the XF300? And no announcement of an XF300 successor? Hmmm... Same CODECs? As the XF100?

My problem with this is that the XA20/25 don't have great dynamic range. And no 1080p60 in the XF200. Even the XA20 has that, so you know the sensor can do it. So... why not in this new camera? Seems like Canon has a thing about 1080p60 in it's "professional" cameras. Why is that?

I guess I'm not clear on the market for this camera. Maybe it'll all clear up after NAB.

Bruce this camera does 1080/60P but to the 3rd Card slot which is SD. It does have an unpgraded AVCHD codec same as the AX20/25 in addition to the MXF codec. Lastly they have Wider-DR the same as the Cinema line. How much can it pull from the actual sensor? I have no clue nor am I interested. The specs page on B&H has it all detailed.

Wow. A rotating grip like the EX1 and EX3.

Not to mention the independent zoom, focus and iris rings, (including HDSDI) as found with the older XH-G1S, and now at roughly half the price. Nice.

I'm also impressed with the wireless connectivity for monitoring and control. One less cable to wrangle, tape down or trip over.

Certainly worth a look, gonna have to rent one.

Cute little thing. Reminds me greatly of my very first camera, the old Canon GL2. Pretty steep price though, be much more interested in a 3-chipper XF cam...

Yeah. I like NAB surprises. This is a surprise for me as I thought Canon was just waiting. Will an XF400 come out? Tough to know. The Panasonic PX-270 has all of this functionality with a better choice of codecs and three chips. I am sure Canon is aware that Panasonic is coming out with a winner so maybe it forced their hand a bit.

Now in turn, it would be a nice surprise to see panasonic with a PX-170. But in the end, it is the image quality and light sensitivity that really make the final decisions.

Interesting, but it seems expensive considering it's a single 1/3" chip, and at 1920x1080 can't give full HD resolution.

For that sort of money, I'd really expect three chip at least, and personally find 1/2" models like the PMW200/300 well worth the extra in this class of camera.

I don't see that spec about upgraded AVCHD like the C series has - the only 1080p60 is recorded as .mp4 - no thanks.

I love my XF100, and this has some nice features, but as far as IQ - I'm definitely not tempted to upgrade.

I got my XF100 in May 2011, and the only competition from Panasonic was the HMC40 - its poor low light knocked it out of the running.

Today the Panasonics are stronger but pricier competitors, although I love the small size of my XF100 and have been happy with the 10x zoom.

What I really wish is that Canon would introduce a C150 or something, so I could buy a used C100 for about $2500 :-)

And, the XFs do have full 1080 HD resolution, just not 60p (that's at 720p only). The PMW200 is twice the price of this - a totally different class of camera. The PMW 100 competes with these.

Wow. Remote control with IPad or IPhone? Three ring circus on lens. I'm in.

Interesting, but it seems expensive considering it's a single 1/3" chip, and at 1920x1080 can't give full HD resolution.

For that sort of money, I'd really expect three chip at least, and personally find 1/2" models like the PMW200/300 well worth the extra in this class of camera.I have to side with David on the HD resolution. I've shot XF105 and XF305 side-by-side many times and yes, the 305 wins but the image quality difference is surprisingly minimal. I often can't tell without direct A/B comparison during editing at full resolution which camera shot which clip (or cheating and looking at the file names).

As far as cost, value is subjective. The XF105 started out at the same price so I guess I personally wouldn't consider the XF205 overpriced.

EDIT: I was agreeing with David Dixon about the "full HD" quality of XF cameras.

And, the XFs do have full 1080 HD resolution, just not 60p (that's at 720p only).

I think that David was reffering to the fact that a single sensor chip can not achieve a full 1080 lines of resolution once the imaging color processing is completed. I am no expert but I think this is called de-bayering. Three chips can achieve the full 1080p lines of resolution as the color information is divided between the three chips - RGB.

That is for lab techs to consider :) I would be confident that the XF200 will have a great image.

This is a complete surprise. Looks like a sweet little camera package.

I think that David was reffering to the fact that a single sensor chip can not achieve a full 1080 lines of resolution once the imaging color processing is completed. I am no expert but I think this is called de-bayering.
Correct. It's not just resolution, but coloured aliasing and various other things will be worse as well.

A 1920x1080 Bayer single chip has red and blue counts of only 960x540 - which is simply not as good as a three chip design when all the colours are full 1920x1080.

The other point is that being single chip, it will inevitably be less sensitive than a three chip design (all else equal), and by a factor of over a stop. Arguably, this may be more relevant than the resolution differences.

None of this is to say it will be "bad" - but it won't - can't - be as good as an equivalent three chip design, of the same basic sensor size.

First post in this thread has been updated to include a link to the full press release here at DV Info Net.

Also, I've uploaded eight images of the XF205 to that post as well.

Meanwhile I'm digging the rotating handgrip (XF200 shown here):

According yo Cnet besides the optical stabilization it uses electronic stab. That's why the chip is slightly larger than 1/3.
Nowhere is said that the chip is only 1920X1080 pixels. Could be more...

The extra pixels on that chip belong to an EIS system that works with the OIS system (in the lens) to provide a five-axis stabilizer. This from CUSA's Larry Thorpe via the B&H video embedded in the first post of this thread.

now we have cameras following the latest fashion closely : the front finally shows the tongue out ala miley cyrus . Interesting. and the front is what they see when we shoot. Impressive.

From Canon website:
"Capture more than ever thanks to the XA20’s stunning new 26.8mm wide angle lens. A 20x zoom gets you closer to the action, while Dynamic Optical Image Stabilisation compensates for shake in 5 directions to provide ultra-steady images. f/1.8 maximum aperture, an 8-blade iris and EDM technology enable beautiful ‘bokeh’ effects.
Canon XA 25 1/2.84” HD CMOS Pro sensor
With an effective pixel count of approximately 2.91 megapixels, the new Canon HD CMOS sensor delivers stunning picture quality – even in low light – as well as outstanding dynamic range and realistic colour reproduction."


Since it seems that the 200/205 matches exactly the front element of the XA20/25 isn't far fetched to assume that it will sport the same higher resolution 2.91mp sensor which I think has enough pixels for true 1080 debayered image?

Actually it does.

In the newest Canon brochure, it states that the 200 has 2.91 mp vs 2.07 mp of the 105

http://downloads.canon.com/cpr/software/video/Canon_Professional_Camcorders_Brochure_Spring2014.pdf

The new cameras appear to have the same sensor as the G30/XA20/25. The extra pixels are used for dynamic image stabilization. They seem to have the same lens as well, and definitely the same processor.

In fact, these two new cameras look like repackaged XA20/25s, with a few new bits added (such as CF slots and three user rings instead of one), bundled with the MXF meg2 codec. So, they have cannibalized parts of the XA20/25 and XF300/305, and stuck them together.

This has to be a stop gap product to cover for their 4K technology in that market group not being ready yet.

I'm not seeing any built-in ND filters on this one.

Wow. Canon has been reading the criticisms of the PMW-200 on DVInfo! That periscoping mic holder is easily removed/assembled (albeit with a screw driver) for packing and travel. And a shoe/LCD design that doesn't interfere with each other. Plus 3 rings as noted. Several well done ergonomic improvements for a camera in this class shows they listen.

I'm not seeing any built-in ND filters on this one.

There's a switch labeled Iris (ND) right where you'd expect an ND switch to be.

Isn't the ND electronic?

Wow. A rotating grip like the EX1 and EX3. Funny how Sony bagged it in the PMW-200 but brought back in the 300 after feedback. Nice.

Even more funny that the rotating grip was on the JVC HD1 / HD10 camera. Probably the only cool thing about that camera, so I can see why people forgot about it.

The new cameras appear to have the same sensor as the G30/XA20/25. The extra pixels are used for dynamic image stabilization. They seem to have the same lens as well, and definitely the same processor.

In fact, these two new cameras look like repackaged XA20/25s, with a few new bits added (such as CF slots and three user rings instead of one), bundled with the MXF meg2 codec. So, they have cannibalized parts of the XA20/25 and XF300/305, and stuck them together.

This has to be a stop gap product to cover for their 4K technology in that market group not being ready yet.

Actually this is the same approach they did with the XF100/105 - those have the same processor as the G10/XA10 but with more features and a better codec. The 10x series already has dual CF slots, so the .mp4 recording to SD is all that's new in that area. I'm not sure if the XA10 and XF100 shared the same lens though.

I'm just glad I'm not in the market right now - with $2K 4K cameras and the C100 only $1500 more than this (if you already have Canon glass, which I do), I don't know what I would buy today.

.......isn't far fetched to assume that it will sport the same higher resolution 2.91mp sensor which I think has enough pixels for true 1080 debayered image?
You really need more. That gives less than 20% extra, and 25-30% is normally taken as the minimum - the Alexa has 50% more. It appears more likely the extra are being used for image stabilisation, the image being formed from 1920x1080.

And a reason for a single 1920x1080 sensor is that it's relatively easy to process. It has to be deBayered, but the result comes straight as a 1920x1080 raster. Oversample, and you have an image which then needs to be downscaled which is not easy to do well. If you're going to take that approach, it's only really worth it with a greater degree of oversampling, and tends only to be found on higher priced cameras.

Either way, that's only part of it. Single chip will always lose you over a stop of basic sensitivity compared to three chips of the same size. I'm not saying this will be bad, but am saying it can't be as good as an equivalent camera with 3 comparable chips. Roll on the XF400!

We had a XF100 and then we got an C100 and the difference in resolution was astonishing, like between SD and HD.

I'd like to know if it has picture profile adjustments like saturation, sharpening and contrast like my HF200's. It could then be a good match as an A camera.

We had a XF100 and then we got an C100 and the difference in resolution was astonishing, like between Sd and HD.

An ENG camera and a cinema camera are two very different things, for two very different situations.

An ENG camera and a cinema camera are two very different things, for two very different situations.

I don't find that a useful distinction. In real life, both cameras are capable of being used for some of the same things.

Regardless, the issue being discussed is the image quality and light sensitivity of a single small chip of the XF100 and XF200 versus 3-chippers and single large sensors. The C100 is a good example for explaining it.

I'd like to know if it has picture profile adjustments like saturation, sharpening and contrast like my HF200's. It could then be a good match as an A camera.

The XF series has complete control over picture profiles, do not worry about that. Resolution and sensitivty are the issues. Think of the XF200/205 as an XA20/25 with all the bells and whistles you could want, however, the chip is still the same. The IQ will be better since parameters are optimized, adjustable and color depth is greater; but the underlying resolution and sensitivty cannot be similarly optimized.

It looks like the Xf200/205 lies inbetween the XF100/105 and the XF300/305. They added many of the XF300/305 features while staying with the single small chip design of the XF100/105. Within its resolution/sensitivity limitations it should give an amazing image, and work well in fast paced environments.

If the resolution and sensitivity are all you need then the XF is an great camera. Really very nice to shoot with.

You should also look at the Sony PMW series of similar cameras. If you like the "Sony look", and I do not, they are worthy competitors for the same situations. Except for resolution and sensitivity, I preferred our XF100 images (with Cowpunk52’s profile adjustments) to our FS100. Not everyone will.

Edit: I used the Xf100 for interviews, particularly of older people, where I needed to diffuse facial details. The color gradients were very flattering.

Does keeping the XF200/205 with a single small chip add weight to the possibility of an XF400/05 in the future I wonder. If they had wanted to consolidate the XF line they could have added 3 chip to the XF200 and left it at that.

I wonder if an XF400/05 with an interchangeable lens is something they might be looking at to rival the PMW 300. Maybe even 1/2 in chips. I'd be seriously interested if they did.

I don't find that a useful distinction. In real life, both cameras are capable of being used for some of the same things.



The shallow DOF in large sensor cameras would be a problem for run & gun ENG work. Likewise would the lack of a long servo zoom for most cinema cameras.

The shallow DOF in large sensor cameras would be a problem for run & gun ENG work.
Not necessarily. Large sensor cameras only have shallower depth of field than smaller sensors ones if you compare at the same f stop.

So take a 2/3" sensor and s35 as an example, and if the 2/3" is at f2, you'll get identical depth of field on s35 at somewhere around f5.6.

Next thought may be "what about light levels!?" - but generally large sensor makes for a more intrinsically sensitive camera, so likely similar performances in that respect also between 2/3"/f2 and s35/f5.6. That's the theory anyway!

It's generally wrong to say "large sensor gives shallow depth of field" - much better to say "large sensor gives the POSSIBILITY of shallow depth of field." It's all down to the f stop.

You really need more. That gives less than 20% extra, and 25-30% is normally taken as the minimum - the Alexa has 50% more. It appears more likely the extra are being used for image stabilisation, the image being formed from 1920x1080.



According to Cnet:
"Like the XA models they're built around the slightly-larger-than-1/3-inch (1/2.84-inch) sensor, though they're effectively 1/3-inch: Canon uses the extra pixels on the sensor for electronic image stabilization, though their new 20X 26.8-576mm f1.8-2.8 lens also has optical stabilization."

with my calculations...
Knowing that total pixel count is 2910000, the difference between the actual and effective image area is 5.34% which translates into 155394 pixels.
If we subtract them from the total, that leave us with 2754606 pixels which are 681006 more than the nominal HD resolution of 2073600 pixels (1920X1080) or 32,84%.
In theory and according to your 25% rule is adequate for a true 1080 debayered image.

David, since you are far more knowledgeable than I am, correct me if I am wrong.

But, getting back to the posts that started this tangent...

Many of us shoot in a variety of settings that often fall somewhere between ENG and Cinematic. For my amateur mix of interviews and promos for local non-profit arts and school groups, my XF100 has been great, except that I wish it had more DOF flexibility. I also have a 60D but have just never made my peace with shooting video on it except in rare, specialized situations.

However, a C100 would also work for everything I do, especially now that it has the autofocus option. So, yes, if I were replacing my camera today, the XF200 AND the C100 would both be on my list for consideration (since I already have Canon glass). They may seem like highly different products, but either would fit my needs.

Not necessarily. Large sensor cameras only have shallower depth of field than smaller sensors ones if you compare at the same f stop.

So take a 2/3" sensor and s35 as an example, and if the 2/3" is at f2, you'll get identical depth of field on s35 at somewhere around f5.6.

This is true only if we're talking about the same focal length. For DOF, you have to compare at the same aperture, but also at the same focal length. Which is problematic when the sensor size is different, because a given focal length gives you different angles-of-view with different sensor sizes.

So if you use the same prime lens on the 2/3" camera as you do on the s35, then move the cameras relative to one another until the framing of the image is the same with both cameras, then you get the same DOF if you stop the s35 camera down an additional 3 stops from the 2/3" camera.

This is why people think, in practice, that a large sensor camera gives shallower depth of field than a smaller sensor camera. Because in practice light levels tend to dictate aperture, and distance to the subject tends to dictate focal length.

That's not to say that you can't get a reasonably shallow DOF from a 1/3" camera. I've seen interviews shot on 1/3" camera where the camera was shot through a door way and was in the middle of an adjoining room, just to get the distance from the subject needed so that the camera could use a long focal length (shooting nearly wide open), and the subject could be pulled far enough away from the background. Between the three (focal length, aperture, and subject's distance to the background), they got sufficient blurring of the background to make it work.

But it would have been a whole lot easier with a larger sensor. Which is what people mean when they say they can get shallower DOF with a larger sensor camera.

This is true only if we're talking about the same focal length. For DOF, you have to compare at the same aperture, but also at the same focal length.
Not so. My comment does deserve a caveat, but it's the same *angle of view* - not focal length.

(The other caveat I'm taking as read is same system resolution.)

If you keep the same focal length and aperture - but vary sensor size - the depth of field will stay the same, let's say between 5 and 7 feet. What will change will be angle of view - and hence that's why a large sensor can give shallow depth of field with a relatively wide angle lens.

If you change focal length with sensor size (to keep angle of view constant), then with the same f stop it's true the depth of field will change - it will be shallower with the large sensor/longer focal length lens. But stop down the appropriate amount (2 stops if you're quadrupling the sensor area and doubling focal length to compensate) and it's back to square one.

David, since you are far more knowledgeable than I am, correct me if I am wrong.
Well, I'm not sure about the first bit of that....... :-)

No, your calculations are correct, but I may have been clumsy with my original statement and misled you.
with my calculations...Knowing that total pixel count is 2910000, the difference between the actual and effective image area is 5.34% which translates into 155394 pixels.
If we subtract them from the total, that leave us with 2754606 pixels which are 681006 more than the nominal HD resolution of 2073600 pixels (1920X1080) or 32,84%.
In theory and according to your 25% rule is adequate for a true 1080 debayered image.

The "25% rule" is normally taken on a linear photosite count and my apologies for not making that clear. Same with "preferably 50%", which is what the Alexa does - it's 2880 horizontally, or 1.5x1920. (Likewise 1.5x vertically.)

So talk of total numbers and you'd really expect (at least) 1.25x1.25 as many (56% extra), preferably more - in the case of the Alexa it's 125% extra!

If you go down the oversample, deBayer, downscale route it's a lot of extra processing and I'd only expect it to be used when well worthwhile. And doing all that work for only 32% extra photosites somehow seems a lot of work for not that much potential gain........ But this is all theorising.

The shallow DOF in large sensor cameras would be a problem for run & gun ENG work. Likewise would the lack of a long servo zoom for most cinema cameras.

Yes. I think everyone understands that. My quoted post intentionally uses the phrase "some of the same things" referring to the reality of the middle area between the extremes where either can be used for the same thing.

I think good examples are documentary and wedding work. You can shoot an interview with either a Cinema and an ENG camera but you use both in the production... some at the extremes and some for the same thing. Sometimes you use the one you have with you, or both as interview A and B cameras. Then pack up and shoot some b-roll with an ENG and some with SDOF with Cinema ... some other b-roll or stand up with large DOF using a Cinema ... some SDOF with either, depending. There's plenty of stuff that either can be used for with certain conditions and tradeoffs.

To the point of this sub-discussion, a single chip small sensor has resolution and light sensitivity differences (usually inferior) vs a 3 chip small sensor. And Philip's point was moving from the small single sensor XF100 to a C100 with a single large sensor gave a noticable improvement in resolution. Presumably shooting the same stuff he was shooting with the XF100... aka the middle.

Not so. My comment does deserve a caveat, but it's the same *angle of view* - not focal length.

If you read the rest of my opening paragraph, I think you'll find that we are in agreement even if we say it differently. What I said was:

This is true only if we're talking about the same focal length. For DOF, you have to compare at the same aperture, but also at the same focal length. Which is problematic when the sensor size is different, because a given focal length gives you different angles-of-view with different sensor sizes.