Canon 50d sample pics.

Hello guys and gals,<p>

 

The 50d sample pics are up at dpreview. The iso 6400 have some banding noise, Is this normal on cameras with iso

6400 feature?

<p>

I think it is a little worst than I expected( at extreme hi iso). Even though it is still a prototype (I think) they tested, I

dont have any reason to believe that the production units are any better.<p>

 

<a href="http://www.dpreview.com/news/0809/08090102canoneos50dsamplegallery.asp" >50d sample pics</a>

<p>

 

Well, What do you think?

What's the deal? Every picture you click on takes you to the same picture.

Click on any picture and then click next/previous link to move to the next/previous picture...

<i>The iso 6400 have some banding noise</i><br><br>Which sample picture and where exactly..? The "plum, bananas & pears" looks OK, if slightly overexposed, to my eyes. Much nicer than 40D processed to ISOE 6400 IMO.

Still a fine pic. Not very vivid, it's ISO 6400 pretty amazing for a camera that level.

Yes , the pics with the fruits, The banding noise is noticable on the upper left, Even on a non 100% magnification. it looks so annoying. While the the iso6400 sample of the nikon d90 is much better to my eyes, filmlike noise, pleasant to look at.

 

I thought canon really have better sensor than nikon this time.

 

Anyway I have no plans to buy a nikon dslr. Just ranting :I

One has lots of colors the other don't.. don't think you can compare that.

Nice pictures, BUT... what I would really like to see is the exact same picture, taken with the exact same lens, compared to the 40D and XSI.

 

I realize that all of these bodies can not take all the photos at the same iso, due to specification limitations, but lets see how they do at their maximum iso with the same light and lens. Lets see the advantage of shooting at 6400 iso?

 

I'm tired of "evaluations" that include subjective comparisons. Why not show us what the camera body can do under the identical conditions so that we can make objective decisions on if the extra money is worth the difference in performance.

 

I do appreciate DPR's excellent reviews, I just would appreciate a comparison of real world photos, not test patterens, or studio props.

 

Thank you

I cannot believe Canon STILL hasn't addressed the banding issue in the darker/less-exposed areas of the images.

 

The 20D has it and it was released in 2004. The 30D has it and it was released over a year later. The 40D has it and it was

released another year later.

 

And now the 50D STILL has it, a full 4+ years after the 20D was first released.

 

Canon has really dropped the ball here. While it looks much better-controlled on the 50D (the 6400 ISO image with banding

is what my 1600 ISO usually looks like on the 30D), I just can't believe it's still in the image at all.

 

Noise is one thing, and is quite acceptable, especially when it's relatively low at a very high ISO. But that banding is just

awful.

"Even though it is still a prototype (I think) they tested, I dont have any reason to believe that the production units are any

better."

 

Why? A Canon engineer told you? If it didn't still have tweaks to be made, they wouldn't call it a prototype...

"I do appreciate DPR's excellent reviews, I just would appreciate a comparison of real world photos, not test patterens, or

studio props."

 

DPR takes pictures of the same studio props and same lighting so you can compare different camera models. You merely

have to use the search button (couple seconds) and open samples in separate tabs on your browser to compare them.

I don't think it's that bad.... It's not perfect... Unless you consider minimal noise and a touch of banding at ISO 6400 to be perfect (cough cough).

 

Either way, looks like it is a touch better than the 40D for sure... but hard to say exactly how much at ISO 1600...

I could accept noise, But I;m really annoyed by the banding noise, cause even after you pass the image through a noise reducing software, there would still be banding patterns. That image might be acceptable to some , But wait till you see an image with night sky, That banding would get more annoying .<p>

 

In comparison , the nikon d90's iso6400 sample images dont suffer from that issue. below is the link to nikon d90 sample mage, What I'm reffering to is the one with a news paper, the last one as of now.<p>

 

<a href="http://www.dpreview.com/gallery/nikond90_preview_samples/" >D90 sample images</a>

"Why? A Canon engineer told you? If it didn't still have tweaks to be made, they wouldn't call it a prototype..."

 

Everything can be change for sure, But I'm pessimistic about that banding noise being eleminated on the production unit. I just feel it.

Wait a minute, I made a mistake, It is a full production unit of the 50d, as quoted from the dprevview site.

 

"Just posted! We've now got a full production Canon EOS 50D and over the weekend produced a quick sample gallery to add to our preview of this 15 megapixel update to the EOS 40D."

 

That means, we get what we see, unless changes are made shorlty.

How does it look when you print out the image? I use that as my ultimate quality checker.

I mostly use ISO 160 on my 40D and IQ is impressive. But in low light I wouldn't hesitate in dialing up ISO 1250-1600 and

have been pleased with the results as long as I don't print larger than 8 x 12. The 50D samples at ISO 800 and 1600 look

cleaner than the 40D, and that's all I need to know. I've rarely needed to use anything higher than ISO 1600. Save the

minor banding in the mid-lows in the upper right of the fruit image, the 50D 6400 looks much better than 3200 on the 40D.

Once you dial into the H level of ISO you're always compromising IQ just to get an image. Canon makes that pretty clear.

Looking at the pictures of the fruits, all I can think is heavy noise reduction. The colors are muted, details and contrast are gone to the extent that the banana looks like a flat 2D object. I will have to see side-by-side comparison but I think that shooting at 1600/3200 and pushing the exposure in PP may yield better results.

well, ISO 6400 remains a 'surveillance' setting for this type of camera, but the ISO 1600 sample is far better than what the 40d can produce. 50D 1600 sample looks (probably) a tad smoother than 40D at 640. that's not bad

 

lower ISO samples have much more punch than 40D can produce. to me, the lower ISO images taken with fixed lenses hint at the look gotten from older 1D bodies. i've never thought that of a crop body image before

To me, the low ISO images look very soft and lack detail, the high ISO shots have heavy handed software noise reduction that makes them look like water-color paintings. I would still wait for CR2 samples and run them through DPP with no NR to see how good the sensor output is. Quality standards are very subjective, however IMHO, these shots are not accetable.

It could be the options chosen by the users e.g. heavy noise reduction. The images are mostly very poor examples, as the subjects lack sharp detail. Lots of architecture and landscapes would have been better.

yes, agreed on that. the images are not good samples. it's a little frustrating in both their subject/lighting, and the lenses they elected to use.

Well no biggie. The real deal will wash ashore soon enough and you'll have GBs of baby, dog and cat pics from early adopters to gush

over!

Ok. Can somebody who has ever shot film please let me know what film type existed that was 6400 speed and didn't look

terribly grainy and off-color. Seriously folks? the 6400 looks terrible? of course it does, it's 6400! What nobody is pointing

out is that 5 years ago 6400 didn't even exist on digital cameras, and 400 looked terrible and grainy. Digital picture quality

has grown exponentially in the past few years, and frankly, if i am shooting high ISO for the grainy effect (to try to replicate

my old fuji neopan 1600 days) i personally don't want grainless and smooth 6400 speed settings. does anyone use this

with any regularity who isn't a private investigator shooting covert photos? the 1600 though looks amazing, and that's a

much more useful film speed to use on a regular basis. Maybe that's the point, the 1600 used to look terrible when the 400

looked great, and now the 1600 looks great when the 6400 looks bad. maybe we will have to wait a few more years for the

51,200 ISO to look terrible so the 6400 looks great?

Hello Roger,

 

I agree with you, Having a 6400ISO that looks like that is an achievement by itself, But what made it worst is that Nikon d90 have better looking iso6400 noise pattern. And no banding noise. If only nikon have the same issue, I would'nt mind the banding noise at iso6400, I would have considered it as normal, But thats not the case. Meaning ,Canon could have done better or at least the same as nikon.