Here's a video that shows the DSLR's have a ways to go yet before challenging the Cinealtas for low budget supremacy. YouTube - Canon 5D Mark II Vs. Sony PMW EX1 (http://www.youtube.com/watch?v=ewyV2_VRLfc&feature=related)

If only the shooters knew what they were doing....

I expected the problem to be skew. Instead, this was some of the worst stuff I've ever seen from the 5D. I'm pretty sure that they had fog between the lens and UV filter, if not on the sensor. That will kill contrast and colors immediately.

BTW, don't bother asking how I know that fog behind a UV filter will kill your image... ;)

Yeah, EX series may indeed produce the better pic, but something has to be amiss here. That looks like it was shot with a Nokia.

Useless comparison. I don't know how and when it became more important to present camera tests as music videos (or even more pretentiously, "films") and make it about the editing and soundtrack instead of what really counts: in this instance, controlled, apples-to-apples duplication of the same setups as shot by both cameras.

Yeah, it's the cameras. Just like it's the brand of piano for the musicians. And the type of stove or the cooks. And the brand of thread for the tailors. And the type of clippers for the landscapers. And the scalpel manufacturers for the surgeons.

You know, that viewpoint is so inarguable that I'm going to go dump all my Beatles vinyl - since it's clear that they were made on substandard instruments and limited electronics and would have been SO much better if only those guys had had better gear.


(sarcasm mode off)

I think someone got a little confused and imported their iPhone footage.

this video sucks. even the EX1 looks dark and flat.

hmm ... both cameras are not being done justice here. Specially the 5D!

Talk about picking a day with possibly the worst lighting to do a comparison or anything.

What Charles said.

I didn't watch it all, but the Canon footage I saw was always a longer shot than the EX, and underexposed and out of focus. The EX wasn't much better.

Here's a video that shows the DSLR's have a ways to go yet before challenging the Cinealtas for low budget supremacy. YouTube - Canon 5D Mark II Vs. Sony PMW EX1 (http://www.youtube.com/watch?v=ewyV2_VRLfc&feature=related)

Ian Provo might dissagree with you.

The EX1 certainly cannot do this;
Search videos for 'ian provo' on Vimeo (http://www.vimeo.com/videos/search:ian%20provo)

and the EX1 cannot do this either;
3 candles on Vimeo (http://www.vimeo.com/2666377)

Looks like the EX1 has been well and truely superceeded by the 5D2...

Looks like the EX1 has been well and truely superceeded by the 5D2...

Well and truly disagree, have both cams.

You know, that viewpoint is so inarguable that I'm going to go dump all my Beatles vinyl - since it's clear that they were made on substandard instruments and limited electronics and would have been SO much better if only those guys had had better gear.
(sarcasm mode off)

Maybe it's like....

"....looking through a Glass Onion"!!

Skilled cabinet makes can all use a framing hammer, some framers can use a mallet or tack hammer exquisitely. As photographers (moving image or not) we are getting some amazing tools and better hammers for different jobs these days. I would have killed for a camera capable of 3 lux (or 300 lux) a year ago - even if it was a half inch chip. Some are using these tools to amazing effect. I'm grateful to this forum for leading me to Dan Chung and his extraordinary work in China, a part (okay REALLY BIG part) of the world he illuminates every time he goes on a job. If the 5D is a better tool for him we all benefit. This camera, the EX, the 7D, Scarlet, or whatever comes next move the boundaries of what we can do in the field, and these changes are happening more quickly than ever before. I'll probably never shoot time lapse stuff like the Timescapes Inc (http://www.timescapes.com) guys, but I like knowing the techniques they've shared and now if I have the opportunity I have a whole new hammer I never knew existed in my bag of tricks. Commonly available cameras and freely shared knowledge make so much possible.
I never understood the guys driving Chevy trucks (in the US) with an image of a kid pissing on Ford. Probably a bad example, maybe no one else in the world develops allegiances to brand names - oh, wait, (or as we say here) fuggedaboutit!
We have new tools almost everyday and for those with skills ( and without skills) this is a most interesting playground.
Michelangelo used a chisel and brush, W. Eugene Smith shot in black and white, they would have envied our largess.
Stop fighting brands boys (and women) - enjoy!

Looks like the EX1 has been well and truely superceeded by the 5D2...

Completely agree, have both;

I have three 5D MK2's an EX-1 and an EX-3. The 5D isn't even close to even the EX-1. In every measurable way the EX-1 is better in terms of video quality.

Recently I did a shoot for a professional pool player and the pool table material had horrendous moire patterns on the 5D footage. Totally absent from the Sony's, flawless footage.

-Most of the SLR glass is pretty slow
-Low angle is a total nuisance without a seperate LCD, very easy on the Sony
-Built in ND's on the Sony
-720 60p on the Sony's (although admittedly I shoot 1080i 60 and deinterlace most of the time.)
-very sub-par audio on the 5D, low rent connectors, no phantom power, auto gain, yuck.....
-Sony offers a million ways to paint the camera, various settings where you can shoot a signal out SDI port to a production monitor to dial in the camera
-Canon puny 17mbps datarate, Sony 35mbps with the option to run out to a AJA I/O, Matrox MXO2, etc... Shoot into ProRes 422 to really leave the SLR's in the dust.
- thats just a quick off the head list....

The one thing I like about the 5D is that it is very convenient to carry around one camera that also shoots video. This is the ultimate location scouting camera. Also an excellent macro video camera. I use the 5D with the 100mm, 65mm, and 180 Canon Macro lenses, really beautiful video. And we recently picked up a quality microscope and the Canon t-mount. So that camera is very handy and has its place, but its no EX-1.

-Most of the SLR glass is pretty slow

True. But some lenses are quite fast.

-Low angle is a total nuisance without a seperate LCD, very easy on the Sony

Yep. Budget for a monitor. (Hopefully, we will get HDMI HD during recording soon.) And budget for a loupe, if you use a shoulder rig.

-Built in ND's on the Sony

Yep. Budget for NDs - in money, and an extra minute to find and mount it.

-720 60p on the Sony's (although admittedly I shoot 1080i 60 and deinterlace most of the time.)

The 7D also does 720 60p. Hopefully, the 5D2 will soon too, among other new rates. ;)

-very sub-par audio on the 5D, low rent connectors, no phantom power, auto gain, yuck.....

With a juicedLink preamp and Magic Lantern firmware, the audio is completely transformed. The results completely exceeded my expectations for the camera.

Budget for a preamp, and plan to use Magic Lantern - not just for audio, but for zebras and other goodies too.

-Sony offers a million ways to paint the camera, various settings where you can shoot a signal out SDI port to a production monitor to dial in the camera

Yeah, the Picture Styles are so promising, yet so disappointing. If control of Picture Styles were more versatile, with processing before the encoder, the results could be fantastic.

-Canon puny 17mbps datarate, Sony 35mbps with the option to run out to a AJA I/O, Matrox MXO2, etc... Shoot into ProRes 422 to really leave the SLR's in the dust.
- thats just a quick off the head list....

The 5D2 does 40 mbps or so. The 7D does 48 mbps. They aren't uncompressed, but solidly beat the other HDSLRs. Too bad we can't get uncompressed out of the HDMI port, though frankly it would tether the camera to a PC, or cost some serious cash for a portable recorder.

The one thing I like about the 5D is that it is very convenient to carry around one camera that also shoots video. This is the ultimate location scouting camera. Also an excellent macro video camera. I use the 5D with the 100mm, 65mm, and 180 Canon Macro lenses, really beautiful video. And we recently picked up a quality microscope and the Canon t-mount. So that camera is very handy and has its place, but its no EX-1.

I like the way you are approaching this. Use the EX-1 where it does well. And use the 5D2 where it has a unique advantage.

I'm starting from the other point - the 5D2 is our main camera. So first, we maximize what it does well. We learn what not to do with it. And we either avoid shots that bring out the weak points, or rent/buy other gear when we want to cover the 5D2's shortcomings.

-Most of the SLR glass is pretty slow
True. But some lenses are quite fast.
And the EX1 lens isn't exactly fast -- 5.8 mm f/1.9 is equivalent to a 31 m f/10 on the 5D Mark II. That is worse than the cheapest kit lens out there for a SLR.

The 5D2 does 40 mbps or so. The 7D does 48 mbps. They aren't uncompressed, but solidly beat the other HDSLRs. Too bad we can't get uncompressed out of the HDMI port, though frankly it would tether the camera to a PC, or cost some serious cash for a portable recorder.
The EX1's 35 mbps is MPEG-2, while the 5D and 7D are the much more efficient MPEG-4.

Here's some free advice.

Telling anyone that Camera A is "better" than Camera B, at best exposes your lack of real experience, and at worst, is insulting to all involved.

It implies that YOU know what THEY need, better than they do. And guess what? You don't.

Personally, in the past 12 months I've shot with NINE different cameras on professional gigs. Including both of the cameras referenced above.

And here's the truth.

Each have strengths and each has weaknesses. Each got selected when their strengths matched the needs of the project.

In fact, I'm shooting this weekend in San Diego for a Not-for-profit. For a while it was going to be an EX-1 gig. Then the client mentioned that one part of the shoot was going to be newsworthy and they wanted us to distribute SD DV clips to various media types to see if they can get PR coverage.

Suddenly an ONLY SHOOTS HD camera like the EX-1 is out of the running since screwing around with transcoding files screws the delivery schedule and means we'll potentially miss evening news broadcast deadlines.

The 5dMkII I'm taking on location for stills and B-roll video with some interesting special effects lenses isn't a smart choice EITHER since tethering the still camera to active storage kills the portability that I need for those special shots.

So NEITHER makes the cut. And a DSR-series Camera with docked hard drive instead of tape storage and the ability to drag clips to a DVD burner fast makes a LOT more sense.

So sorry, but what's "best" for you isn't what's "best" for me. Heck, what's best for me today isn't what's best for me tomorrow.

So sorry, but your advice = Massive Fail.

To ease the pain let's rewrite your advice in a fashion that gives it the benefit of truth...

"If you shoot the same kind of work that I shoot, and if your quality and functional standards match mine precisely - then MY PERSPECTIVE is that the XL-1 is a better camera CHOICE than the 5dMkII."

See. Point still made and yet we elevate the discussion from Colbert style truthy-ness - to actual, defensible truth telling.

Let the celebrations begin.

Have a nice day.

The EX1's 35 mbps is MPEG-2, while the 5D and 7D are the much more efficient MPEG-4.

While that's pretty much true, MPEG only specifies the decoder. The encoder is left to the developer. You could have a multi-pass MPEG-2 encode to a small size with a good result, and an quick-and-dirty MPEG-4 encode with lots of bits that looks terrible. Or vice versa.

The MPEG-4 encoder in the 5D2 is designed to fit in a camera with long battery life, so it's quite minimal. Then again, 40 mbps is pretty fat. HD at 10 or 12 mbps can look wonderful from a top-of-the-line encoder. The large file sizes make the 5D2 and 7D encoders viable.

How does this compare to the encoding in the EX-1? I have no idea! The EX-1 encoder won't be world class either as it also runs on batteries in real time. There is a whole science to evaluating codecs. You can start with an RMS error analysis, do human measures, develop measurement algorithms that take the human visual system into account, and so on.

Personally, I'm impressed with the 5D2/7D encoder. It stomps that of the other HDSLRs. It's not perfect, but it's impressive for the price. The encoder was what convinced me to stretch my D90-sized budget to buy the 5D2 last year.

I guess I don't really care how the encoder of the 5D2 compares with the EX-1. The cameras are so fundamentally different that I wouldn't choose one over the other on the basis of their encoders. Aliasing, rolling shutter, recording length, DOF, and ergonomics are better categories for comparing the two cams.

Here's some free advice.

Telling anyone that Camera A is "better" than Camera B, at best exposes your lack of real experience, and at worst, is insulting to all involved.

It implies that YOU know what THEY need, better than they do. And guess what? You don't.
So sorry, but what's "best" for you isn't what's "best" for me. Heck, what's best for me today isn't what's best for me tomorrow.

So sorry, but your advice = Massive Fail.

To ease the pain let's rewrite your advice in a fashion that gives it the benefit of truth...

Let the celebrations begin.

Have a nice day.


AMEN to that.

I'm sorry, but add a couple fast lenses (maybe $2500), ND's (~$200), external mic ($200), XLR adapter ($200), external monitor ($500), and all the other gadgets to a 5D and how much are you really saving over an EX-1. Probably nothing.

As far as anyone comparing camera "a" to camera "b", sure you could make the case that this is silly when comparing two dedicated video cameras. Alas this is not what is happening here, a true video camera is being compared to a still camera with some very limited video capability.

I have both cameras and would not give up either one as I stated earlier, but to make the laughable case that the 5D has video that compares to the EX-1, well that is ridiculous when you view properly exposed footage on a production monitor. 40mbps from the Canon?, I've never heard that before and the footage is undeniably softer (no matter the lens cause we have all the best Canon lenses) and has more artifacts than any EX-1 footage I've ever shot. I know Canon does some weird stuff where they skip every other scan line, maybe this is why I have seen some of the artifacting.

The 5D is an low budget dream, but so is the EX-1, and the EX-1 has all the valuable features good video cameras have like peaking, features that relegate the 5D to a b-roll camera IMO. The 5D is a total focusing crap shoot unless you are using a external Marshall monitor with peaking, or have people hitting marks where you can effectively pull focus to very specific points. If you could run expanded focus on the 5D while shooting that would help significantly.

Again both cameras have their place, I mentioned the macro photography/video where the 5D is a real life saver. It is simple to attach a SLR to a microscope, and one that shoots video, well that makes things very interesting. I also like to use the 15mm fisheye, and the 24mm tilt/shift (still trying to get a 17mm TS but they are hard to get!). One of our shooters brought the 5D on a amusement park ride at night with the fisheye. One of those roller coasters where your feet hang down and you are suspended from the top, the 15mm fisheye footage is amazing, and I could not have shot that with the EX-1 the way it was done with the 5D.

The 5D can also gives you that ability to hit a 21MP still while shooting video. If you plan out a shot, you can do some pretty neat stuff with that feature. It gives you some interesting options for transitions, pan+scan, etc...

I'm sorry, but add a couple fast lenses (maybe $2500), ND's (~$200), external mic ($200), XLR adapter ($200), external monitor ($500), and all the other gadgets to a 5D and how much are you really saving over an EX-1. Probably nothing.

...

The mistake made here is that no one is shooting the 5D to match the EX-1. If you were trying to match the EX-1 to the 5D to get the film camera like performance we look for in the 5D, you are adding a 35mm adapter, the same lenses, and off camera mic an external monitor anyway. The whole point of the 5D from the filmaker stand point is trying to emulate images from a 35mm film production. The EX1 has to have a lot more added to it to get it to that level, and when you compare images at that point (see Phil Bloom's work for instance), the 5D image and ease of use seems to trump.

I'm sorry, but add a couple fast lenses (maybe $2500), ND's (~$200), external mic ($200), XLR adapter ($200), external monitor ($500), and all the other gadgets to a 5D and how much are you really saving over an EX-1. Probably nothing.

True, if you need all that it's easily possible to spend more than an EX-1. However it's also possible to go with the body + a few used fast nikon primes and an H4n, come in significantly cheaper than an EX-1 and still produce some amazing footage. Again, it all comes down to what and how you want to shoot.

40mbps from the Canon?, I've never heard that before

Really? This has been common knowledge for over a year, long before the camera shipped. It does have artifacts - all compressed video does - but personally I rarely see them except when using very high ISOs, even when pushing the color quite a bit in post (I convert to ProRes first of course). Aliasing/moire is by far the most noticeable source of visual artifacts on the 5D, and probably the single area where I'd really like to see Canon put in some work to improve things.

I also have both cameras. It is quite clear that this snowboarding video isnt much of a comparison. Moral of the video is that snowboarding looks cooler shot telephoto or at least if youre gonna shoot snowboarding on superwides you need to clean your lens, have some sense of composition or at least do some crazy wideangle POV stuff.

I'm not sure where anyone got the idea that 35mm lenses are slow. I admit they are a bit slower than smaller format lenses, but not by much. ex1 lens is f1.9-f3.4 so if you really want a constant fstop you need to shoot everything f2.8-f3.4 or only shoot very wide all the time.

Tramm, I agree that the depth of field of 5.8mm f1.9 is equivalent to 31mm f10, but im not sure what the significance of that is, since it doesnt make those two lenses equivalent. having a lens that is f1.9 is obviously better than f10 given that lighting for an f10 is an order of magnitude more complicated than for a f1.9... especially if you were to stop down from wide open to minimize aberration/maximize sharpness...

Also, I would definitely trust 35Mbps XDCAM over 40Mbps h264 as it is encoded by the 5dmk2. the encoder at 40Mbps on the 5dmk2 is probably better than most of those <25Mbps AVCHD codec cameras, but still im not convinced of the quality encoder in the 5dmk2. The EX1 on the other hand in my experience (of course anecdotally) is more reliable even at its lower bitrate.

That said, the 5dmk2 is still king in terms of latitude and high iso performance. EX1 is great for a 1/2" HD camera, and it might even come a little bit close, but i generally wouldnt go over 6db on the EX1 (depending on the picture profile)which equates to about iso800, whereas the 5dmk2 i find pretty acceptable as high as iso1600 or even iso2000 (i dont like iso3200 but id i might use it in a pinch same as with 12db on the EX1).

"Really? This has been common knowledge for over a year, long before the camera shipped. It does have artifacts - all compressed video does - but personally I rarely see them except when using very high ISOs, even when pushing the color quite a bit in post (I convert to ProRes first of course). Aliasing/moire is by far the most noticeable source of visual artifacts on the 5D, and probably the single area where I'd really like to see Canon put in some work to improve things."



Well Sony is already there, which is my point. If I had a DP that insisted on shooting an indie film on a 5D that I was financing and I saw those artifacts and moire, I would fire him/her immediately. I've had it absolutely ruin shots, and these are not stills that you can just goof with and get by.

40mpbs I stand corrected,although that number is meaningless since the Canon at a higher bitrate is totally and undeniably second rate to the Sony 35mbps Long GOP. The image quality isn't even close in certain cases, that doesn't say much about the Canon video processing engine. Hook the EX to a MXO2 and the 4:2:2 10 bit off the Sony is leagues ahead of the 5D, especially for keys.


If Canon can get some kind of RAW video situation going ala RED, that would be a potential game changer. I do worry that Canon's video division will fight to keep the SLR's neutered for internal political reasons.

As far as cinema style lenses go, this lens speed issue is relative. 2.8 is a fairly slow cinema lens, you may not shoot wide open, but most do wider apertures than 2.8. I have pretty much every Canon L lens I can think of, and many of them are F4. I have some old 60's/70's Nikon glass but I simply think the newer stuff is better in this regard. There isn't a old SLR lens that can come close to something like the current Nikon 14-24.

Frankly if RED would ever get their act together that would be the camera for me, but I think that the under 10k camera is just not going to happen any time soon.

I have pretty much every Canon L lens I can think of, and many of them are F4.

14mm f/2.8L
17mm f/4L
24mm f/1.4L
24mm f/3.5L
35mm f/1.4L
50mm f/1.2L
85mm f/1.2L
100mm f/2.8L
135mm f/2.0L
180mm f/3.5L
200mm f/2.0L
200mm f/2.8L
300mm f/2.8L
300mm f/4L
400mm f/2.8L
400mm f/5.6L
600mm f/4.0L
800mm f/5.6L
16-35mm f/2.8L
17-30mm f/4L
24-70mm f/2.8L
24-105mm f/4L
28-300mm f/3.5-5.6L
70-200mm f/4L (two versions)
70-200mm f/2.8L (two versions)
100-400mm f/4.5-5.6L

And that's just the ones currently in production. I wish I had just a quarter of those in my camera bag!

Sony isn't "already there", or there wouldn't be interest in 5D/7D video. Just about everyone here has used a lot of different cameras.