Judge these 3 camera IQ's
 

I'm looking to buy a camera to allow wide angle and continuous footage in conjunction with a 7D and T2i. Much of my work (hobby) tends to be in dimly lit areas capturing performances or other interior spaces. I came across the camera-test site of slashCAM.com and looked at test images in a staged environment from 3 different cameras ranging in size and price. How would you judge the resulting IQ's from these cams not knowing which cameras are being tested?



Note: The images were processed in Sony Vegas 9.0d to improve the images and achieve similar brightness and color levels (analysis and CC done independently from slashCAM.com). I believe the slashCAM tests took the images of all cameras in 60i in Auto mode so the camera's ability to optimize the image on its own is likely a significant factor in these images.

Aside from some odd color correction, all would seem usable...

I won't comment on the problematic nature of taking test samples that tend to vary (the site you pulled these from is better than most review sites, but their "stills" are not without significant variation... which to me weakens any conclusions you might come to). I'll presume you're comparing the Panasonic TM700, the Sony CX550V and the Canon HF-S21... no clear winner from everything I've seen, just up to you what features you want.

Dave, from the images, which cam do you feel presents the best image?

I didn't spend a ton of time on CC but tried to make each image look reasonably good in a few minutes versus trying to make them match each other exactly. I found pretty quickly that to get some tones to match, say skin, some of the other tones went wonkers.

I guess it would depend on your shooting conditions - all look pretty close in good light, but in lower light cam 2 starts to get noisy, and cam 3 goes very soft (which may just be bad focus?).

On another thread they linked to the three cams mentioned, and I noticed a "low light optimized" shot was added to the Panny and the Sony columns... compare those two, if low light is a consideration, I think that answers it.

BTW, don't assume the 3 cams you mentioned are the ones being tested.

I tend to take comparisons like this with a slight pinch of salt, as differences can often be down to individual setting, and the web presentation means you're getting a somewhat altered view. The colour correction can also be highly subjective.

That said, camera 2 does seem noticeably worse in lowlight than the other two......... If the lighting level was the same for all three, then for the same output brightness it seems to show more noise than the other two.

At first I thought I liked, over-all, camera 2's colors and brightness. Then after reading the above posts I went back and watched it again and agree about the grain in camera 2's low light. Camera 3 was out of the running for me from the beginning so I guess I would choose camera 1.

Do I win the prize?

I didn't go hunting on the links, but what camera is camera 1?

Whatever the cameras are, from the given footage, camera one would seem to be the "winner" if low light is a consideration. Lots of variables to the equation, but given what's presented, seems like we're coming to the same conclusion.

FWIW, I too thought cam 2 looked better until I went full screen... the miracles of size and interpolation... upon enlarging the clip, cam 2 just fell apart noise wise.

Sorry if I presumed what direction you might go for a second angle cam... the three I mentioned were logical, although if budget were expanded significantly, there's LOTS of other choices... and judging from the low light samples, I'm going with you're comparing smaller cameras, whichever ones are in the running... small cams tend to fall apart in bad light, some less gracefully than others, that's just the way it goes!

From experience, something with the EXMOR R sensor is as "graceful" as you'll get in a small camera, and you indicated you shoot in bad light conditions - so I logically figured you'd have a XR or CX series Sony in there somewhere...

Sorry, no prizes! One of the cameras is the CX-550V. The others are XF300 and EX1R. Without looking at the site, can you match them up?

XF300 That would be cam #2.

I'm torn on cam 1, as the EX1R should be head and shoulders sharper and better in low light, so that would seem to be cam 1.

A bit uncertain on cam 3 though, as the CX550 would be the "logical" choice on price alone, the camera isn't generally THAT soft, although in comparison to a camera costing multiples more...

As I've looked at the site recently, I'm going somewhat from memory, so its a bit of a cheat... I know I concluded at that time that the CX550V, as with it's immediate EXMOR R predecessors did a darn good job holding its own against cameras many times more expensive - particularly in bad light when optimized.

Um, DANG... just went and cross checked... I knew the CX550 looked good, but that's sort of embarrasing. Swap my guesses for 1 &3...

I've got to suspect that there's something amiss with the focus on the cam 3 samples.

Pretty interesting isn't it?! Cam 3 does look a tad soft, I agree.

Camera 3 is broken or set up wrong.

Good evening,

I watched it carefully and I personally prefer the overall images of the number two, I feel the fidelity to clearity of color and detail was slightly better. Number 3 is closer to two. I do not like number one backgrounds and colors just seem further out. all three could be color corrected and be useful.

the real question is which one requires the least color correction, or a more pristine original image!!

In the end you have to pick the one you can make the look you like most easily.

I am partial to colors that pop better and are technacolorish. which is probably why I like number two.

that is also why its nice to have a camera you can adjust the color to some degree.


Dale

PS I just read the posts, I never enlarged the pictures so I was not even really looking at gain.
Lots of variables!!!

Lay out which camera is which, so we know!!

Cam 1 = CX550V; Cam 2 = XF300; Cam 3 = EX1R

I prefer the image of Camera 1. Not saying it's a better cam, just in this instance the image looks cleaner, sharper, and was easier to color correct. The lack of noise in the 550V is amazing to me. Obviously this isn't a thorough exam, we don't have a clip of frames to see motion.....but it is interesting that the 550V holds up this well against the others in a frame grab.

Roger -
I still suspect that something is wrong with the EX1R samples... That camera is very clean and sharp from everything I've seen... BUT...

Having shot with the XR500 and CX500, I know that those cameras can produce some extremely good footage, with the least noise I've seen - there is so much less"creepy crud" that I hate seeing when looking at video. And the footage just looks sharper and crisper. The jump from the SR11 to the XR500 was so noticeable to me it was huge (the primary difference being the "R" CMOS). That a "consumer" camera in the $1K price range can produce such a clean image is pretty impressive.

I've been looking for something "big" that can match it, and honestly not finding anything that compelling. This isn't the first time I've looked carefully at "comparisons" and thought the 500/520/550 looked as good or better than cameras many times the price. It shouldn't, but it seems to...

I suspect the EX1R would look better "in motion", but the new higher bitrate of the XR/CX550 may negate even that advantage.

This certainly makes the "mystery cam" that was shown at NAB that was clearly a hot rodded "pro" version of the CX550V, a lot more interesting... If they added some truly "pro" features, and there's no reason they couldn't IMO, that cam could be a real serious contender.

BTW, on the AVCHD thread here, we've been (strangely enough) discussing the probability that the frame grabs on the CX550 are a bit "off" to the warm side...

Thanks Dave. I know the EX1R can do better, I've seen excellent footage from it and certainly in manual mode with a thinking operator it would perform better and be in better focus. my guess is that both the EX1R and the XF300 in Auto mode both cranked up the gain to insane levels in the 12 LUX scenario. I also understand that any test can have errors in setup or processing. There is no denying, however, that the little 550V produces a sharp image with low noise; or at least one frame of it! I've read enough reviews to know it's IQ can surprise professional users who bought it for a family cam. And of course it is 60i which is a big deal to a lot of people. I don't find 60i objectionable and converting to 30P is easy and effective when needed.

I certainly think the 550V is up to providing a continuous wide interior shots while using the DSLR's for high impact accent shots. It may also be perfectly fine for capturing A footage when not trying to create a cinematic look or otherwise needing all the manual controls for travel, hobby, or lower end professional work.

Making judgments about the quality of camcorders from downsized compressed still images placed in a compressed video doesn't make a lot of sense. In particular it removes the benefit of higher data rate and a larger color space. Even if the test were performed correctly, pro videocams should be judged with what a competent operator can shoot with moderate effort.
Look at how the testing website rated camera 3. They don't even look at their own test images. Why would we want to look at their test images?

This is the challenge with "review" sites - there will always be SOME biases, and there's no guarantee of optimized samples from a camera you've had hands on for maybe a few hours. And I always question whether "identical" is really IDENTICAL when it comes to shooting conditions. Until the cameras are in the exact same spot with the exact same framing of the exact same scene, and the exact lighting conditions... you get the idea. Or for actual "live" video footage, the two cams need to be mounted side
by side...

All that aside, you can get "general impressions" that hopefully will help evaluate which cameras you want to seriously consider, so you know what meets your personal checklist, and what doesn't. Some people don't want to consider a camera without "p" options as even BEING a camera... not an issue IMO... so there's a bit of "bias" vor you!

That said, the German site is pretty good, CCI has its moments, and some other sites sometimes have something more than "fluff" reviews. Or you can ask around here on DVi... and get the straight answer <wink>.



I think the takeaway is that the CX550/XR550 can pull in some amazingly clean footage - a fact that those of us who have the camera or its immediate predecessors can easily confirm. My XR500 and CX500 have extremely clean and noise free images, and can "see" better than I can in bad light, while maintaining a good usable image. I expect the CX550 to continue that tradition.

I am REALLY skeptical that it would look anywhere near that much sharper than a properly shot EX1, but elsewhere on these boards, the feeling is the XR500 keeps up rather well with the NX5U - something which tends to show if you look at the stills on that site. From some samples I've seen, the NX5 looks a bit soft in comparison to the EX1, and I'd expect that, but to have the CX550V turning in very sharp, almost noise free image at a fraction of the cost is a bit counter-intuitive, but not impossible.

Because of the volume Manufacturers have to produce for the consumer space, and the speed with which they MUST come to the market with new products (new line EVERY YEAR - "pro" gear is a multi-year proposition!), one can see where the technology might just leapfrog in some respects...

The question is whether you can live with a small cam with relatively limited manual control/control surfaces (but the CX550 has a touch creen which brings its own advantages to the table)? Will you be able to live with a camera that looks pretty much like what uncle Bob brings to the party? Can you look past the "image" of shooting a small camera and keep the focus on the IMAGE that camera can give you to work with in post?

When you consider the relative cost, you can buy 2-3 of these for multicam coverage, add the T2i for "glamour shots", and still be out the door under the price of ONE EX1 class camera with room to start buying the needed accessories. For certain situations, that's not a bad setup. Might not "impress" the gearheads, but the final result of a good multicam mixdown WILL impress the viewer, and in my estimation in some markets that's more important than a fancy "big" camera...

I also have a XR500V which I am starting to use with my EX1. I, too, am amazed at the XR500v's low light capability. I just wonder why Sony hasn't implemented their "R" CMOS tech into their pro cameras.

When I was watching the video, I thought something was odd with #3 because it handled highlights far better than the others. Notice when the video moves up - the highlights are clipping far more in 1 & 2.

The most likely reason why the EX1r is soft is due to focus and much smaller depth of field.

Steve, how does your small Sony's footage look when edited with the EX1R?

The sony is 1080i, I believe.

Roger, I haven't used the XR500V alongside my EX1 yet but I will be this weekend.

Please report back. They are running an instant $150 rebate on the 550 thru the 8th and I'm considering buying one. Haven't pulled the trigger yet.

My last three show were with SR11, XR500 and NX5U. I can confirm that the XR500 is the cleanest of the three, NX5U next and then the SR11. The XR500 at 15db is cleaner than the NX5U at 9db by some margin. The NX5U is marginally better than the FX1 !!!! So far I like the NX5U but not startlingly impressed when compared to the XR500 that is almost 1/4 the price. A XR/CX type product with bigger lens and full manual control would be part of my kit in a heartbeat !!! We sometimes use an EX3 and the XR500 is cleaner than the EX3.

Ron Evans

The low light footage of this 550 is pretty amazing.

the chip is 1/2.88

Always wondered does that meant, is really a 11/25 image sensor, or is it .588???

Who in the known universe combines fractions and decimals??? its a .44 right?


Dale

If I understand the strange math used to size these chip, "2.88" would be ever so slightly less than "3", IOW, basically a 1/3" chip - always confuses the living daylights out of pretty much everyone, but there must be some reason for the strange way it's done...

I know the XR500/CX500 are very clean, and I've been trying to figure out why the samples I've seen appear to me to be cleaner than more expensive cams - I just presumed I was nuts... I do know that I like shooting these and the results I get more than any other cameras I've owned. I hate "creepy crawly" noise, as I know those large areas aren't really moving about like that, and these cameras just don't seem to have that in a noticeable way.

Roger, that current deal at Sony is pretty good, you probably wouldn't regret jumping on it. I've been waiting to pick up one secondhand myself, but that deal is tempting. The closeout prices on the 500/520 series were pretty good, but those are now bringing top $$ in the resale market, almost as much as when they were current.

My only reservation is the new lens range and whether you lose too much on the tele end, but I suppose you'd adjust or could add a Teleconverter if you need more long range. I won't miss carrying the .7 WA converter around...

Like Ron I'd like a bit more in the way of manual control surfaces/independent adjustments (not sure exactly how the 550 series handles the adjustments... it does have access to manual controls that weren't there in the 500/520 series). Ultimately I am rather liking the small size and resultant reduction in kit size, and may NEVER get another "big" camera, though I'd still like something along the lines of the TRV900 or FX7 in theory... but where would I get better image quality? The answer to that question just isn't out there, IMO, at least not at a price I might consider.

4-5x the price for what is marginally better (and apparently possibly worse) image quality simply doesn't add up to me - I've been shooting multicam with small Sonys for a while now, there's not a compelling argument to change.

These "R" CMOS chips seem to be pretty magical - I've been using the P&S Sony DSC-TX7, which has the "R" CMOS, and does AVCHD full HD... it's almost as good image wise (i'd call it an even money match for the SR11/12), and it slides into a pocket. For the price it's really scary.

Right, 2.88 is slightly smaller than three, so the fraction is a hair larger. More precisely, 1/2.88" is .3472", or just a little bit bigger than a 1/3" decimal equivalent of .3333"

Adam,

How did you do that math??

1/2 x .88 is .44 which would be a pretty big chip!!



dale

it is 1 divided by 2.88 = .3472222222222 I still get thrown by the odd approach - it LOOKS like a fraction, but it's not...

Well, it actually is a fraction. 1/2 is the same as 1 divided by 2, or .5. 1/3 is one divided by three, or .333. So 1/2.88 is one divided by 2.88, or .3472 of an inch.

The weirdness comes from a decimal in part of the "fraction" - somehow mentally I think we expect one or the other, i.e. .3472" pr 1/3", not a hybrid of the two...