Canon EOS Rebel T3 image quality issue

Started Apr 15, 2011 | Discussions
maxnimo Junior Member • Posts: 44
Canon EOS Rebel T3 image quality issue

I just took a peek at the new Canon EOS Rebel T3 image sample comparison at ISO 3200, which is my sweet ISO spot for tough indoor shooting, and I have an issue: The T3 image quality at ISO 3200 is almost as good as the Nikon D3S! So what the hell is going on here? Am I the only one that sees this or is there some sort of conspiracy going on? Are all of you blind? Let me repeat... To my observations the T3 image quality at ISO 3200 is almost as good as the Nikon D3S! Now if there is some error in those review tests, that's not my fault, but my eyes don't lie and I have no vested interests here.

PaulRivers Veteran Member • Posts: 7,420
Re: Canon EOS Rebel T3 image quality issue

Assuming you're not trolling or just trying to be sensational, then maybe you messed up one of the iso boxes by accident or something.

The t3 has significant noise at iso3200. The d3s has pretty much no noise.

Click the "Original" link to see this full size (you may need to click on the image one more time after opening it to get the browser to unzoom it) -

Again, the d3s has effectively no noise, while the t3's images are very noticeably noisy.

sjwolfhope Regular Member • Posts: 435
Re: Canon EOS Rebel T3 image quality issue

Aren't you comparing the T3 at ISO 3200 vs the D3S at ISO 200? There don't appear to be any ISO 3200 samples for the D3S in the T3 review to compare to.

Steve W.

PaulRivers Veteran Member • Posts: 7,420
Re: Canon EOS Rebel T3 image quality issue

sjwolfhope wrote:

Aren't you comparing the T3 at ISO 3200 vs the D3S at ISO 200? There don't appear to be any ISO 3200 samples for the D3S in the T3 review to compare to.

Steve W.

Oh crap...you're totally right. I set it to iso3200 at the top, but evidently dpreview didn't feel that they wanted to take iso3200 shots with the d3s (why would they take high iso shots with a camera who's main features is high iso? rolls eyes) and the tool just leaves it at iso200. At least that's the way it is for RAW.

Well don't I feel like an idiot for chiding someone else now...looks like I was the one who had something messed up by accident...

Here's side-by-side's from the jpeg's -

AllOtherNamesTaken Veteran Member • Posts: 3,563
Re: Canon EOS Rebel T3 image quality issue

No Paul, you were correct. In the comparison screen, there is no ISO above 3200 for the D3s in RAW when you go to the T3 review and try compare it with a D3s.

And to the OP, if you look at a RAW image from the D3s vs the 1100D at higher ISOs and think they are even in the same universe, you must not be very picky at all :). There is no comparison there whatsoever, and that is largely due to physics, not because one camera is simply way better than the other.

Mark

AllOtherNamesTaken Veteran Member • Posts: 3,563
OOps I means 200 ISO (NT)

(NNT)

OP maxnimo Junior Member • Posts: 44
Re: Canon EOS Rebel T3 image quality issue

Okay, I'm at the jpeg comparison test chart for the fourth time and I'm looking at the Rebel T3 and Nikin D3S side by side, both set at iso 3200. Both samples look excellent, with the Rebel T3 having just a bit more noise, one f stop at the most. For a cheap camera with no metering and focusing problems and an image nearly as clean as the D3S, this should be seen as a miracle of the decade. So why are the reviews so negative? Something don't add up here.

007peter
007peter Forum Pro • Posts: 12,781
its POPULAR to bash 1100d, Gadget Geek need to feel superior?

maxnimo wrote:

I just took a peek at the new Canon EOS Rebel T3 image sample comparison at ISO 3200, which is my sweet ISO spot for tough indoor shooting, and I have an issue: The T3 image quality at ISO 3200 is almost as good as the Nikon D3S! So what the hell is going on here? Am I the only one that sees this or is there some sort of conspiracy going on? Are all of you blind? Let me repeat... To my observations the T3 image quality at ISO 3200 is almost as good as the Nikon D3S! Now if there is some error in those review tests, that's not my fault, but my eyes don't lie and I have no vested interests here.

Not only is image of a cheap Canon T3 look as good as Nikon D3s, at iso3200 jpeg, it easily beats out the Nikon d3100 for $200 less.

DPR upload only allow for JPEG, go to the review and compare the iso3200 yourself. Canon T3 has excellent detail retention, where as Pentax K-X is over sharpen to a point of jpeg artifacts, sony is ruined by overly strong NR, and Nikon d3100 has a terrible "banding problem"

Here is the screen shot of what I'm seeing:

 007peter's gear list:007peter's gear list
Panasonic Lumix DMC-GF6 Panasonic 20mm F1.7 II
viking79
viking79 Forum Pro • Posts: 14,156
Re: Canon EOS Rebel T3 image quality issue

maxnimo wrote:

Okay, I'm at the jpeg comparison test chart for the fourth time and I'm looking at the Rebel T3 and Nikin D3S side by side, both set at iso 3200. Both samples look excellent, with the Rebel T3 having just a bit more noise, one f stop at the most.

The D3s looks better at 6400 than the T3 does at 3200, so the difference is in excess of 1 f/stop. Full frame inherently has 1 1/3 f/stop advantage. Since those samples are JPG, a different source should be used.

Looking at DxOMark.com, the D3s has a 2 f/stop advantage in terms of RAW noise (it should only be 1 1/3 f/stops better, meaning the D3s has a sensor that does better than the full frame advantage, meaning the T3 has a so-so sensor). ISO 3200 on the T3 looks fairly similar to ISO 12800 on the D3s. This is a pre-production T3, but my experience has shown that it wont change much in that time (pre-production or not it will more than likely have the same sensor, processing of images might change though).

The question is, how good do you need the camera to perform? The T3 should still perform fine. The score on DxOMark.com puts it in the lower end of where APS-C sensor perform. Top APS-C camera is the Pentax K-5 and the bottom the D2h. Having owned both cameras at some point in time I agree with their assessment

Bottom line is comparing the T3 to the D3s is a silly thing to do in the first place.

Eric
--
I never saw an ugly thing in my life: for let the form of an object
be what it may - light, shade, and perspective will always make it
beautiful. - John Constable (quote)

See my Blog at: http://www.erphotoreview.com/ (bi-weekly)
Flickr Photostream: http://www.flickr.com/photos/28177041@N03/ (updated daily)

 viking79's gear list:viking79's gear list
Sony a7R Samsung NX1 Samsung NX 30mm F2 Pancake Samsung NX 85mm F1.4 ED SSA Samsung NX 60mm F2.8 Macro ED OIS SSA +5 more
Snaphaan Junior Member • Posts: 27
Re: Canon EOS Rebel T3 image quality issue

I'm confused...

So does the T3 have good IQ or not? I'm not a good pixel peeper so maybe some tech-savvy here can give it to me straight...

AllOtherNamesTaken Veteran Member • Posts: 3,563
Re: Canon EOS Rebel T3 image quality issue

Snaphaan wrote:

I'm confused...

So does the T3 have good IQ or not? I'm not a good pixel peeper so maybe some tech-savvy here can give it to me straight...

The T3 image quality is fine. Below it's peers (I.e. Pentax Kr and Nikon D3100) but still in the running.

As for this thread comparing it to a D3s that is a silly comparison to begin with and you can ignore it. They are not even close to the same type of sensor, price category, or performance world.

When comparing image quality you must look at the RAW image, otherwise you are just comparing in-camera JPEG processing engines, which is not a good representation of actual image quality because every company uses very different approaches to out-of-camera JPEG's, and the user can manipulate those values to make them all look basically identical. Some companies (i.e. Nikon) use far less default in-camera JPEG sharpening, but you can just turn it up if you want and it will look the same as everyone else.

If you want to look at unbiased, standardized, sensor comparisons, check out the following link from DXO and select the competition (Pentax Kr and Nikon D3100 or whatever else you may be considering):

http://ns365755.ovh.net/index.php/Camera-Sensor/Compare/Compare-sensors/%28appareil1%29/693 |0/%28appareil2%29/664|0/%28appareil3%29/672|0/%28onglet%29/0/%28brand%29/Canon/%28brand2%29/Nikon/%28brand3%29/Pentax

The 1100D sensor is fairly average given today's standards. Canon makes some great sensors but in this segment there seems to be some far better options if this is your main concern. If you're not a pixel peeper you'll probably never tell the difference. Just have a look at the DPreview RAW samples, and if the differences aren't big enough for you to care personally, then don't worry about it. It's best to buy cameras, especially your first camera, based on the entire system, current and future lens requirements, flash system, ergonomics, etc.

Hope that helps,

Mark

Snaphaan Junior Member • Posts: 27
Re: Canon EOS Rebel T3 image quality issue

'sigh' ... I've already purchased my 1100D. I'm reasonably happy and I've made my choice after carefully considering all of my choices (including higher tier cameras like the 550D). I just wish that with all the stuff going against the 1100D that at least they took some trouble with the sensor and IQ. Its disappointing really. Seems canon has been eating dust for a couple of years now.

But thanks for the advice in any case.

It was a VERY difficult decision for me. Yes, Pentax has a great camera but I feel more secure with Nikon and Canon. Nikon's D3100 is incredible but no exposure bracketing? Thats a pity. I like to shoot HDR.

The 1100D might be just a average run-of-the-mill entry-level DSLR with a average sensor and average IQ etc. But it has three things in its favor:

  • Exposure bracketing

  • Canon lenses

  • Price (compared to 550D) which means more cash for better lens investments.

And who knows. In two years time (maybe less) the 600D will price-drop to my level and then I'm off to DSLR fame and fortune. lol

tnnd Junior Member • Posts: 26
T3 high ISO quality is no where near D3S'

just checked out dpreview's article. For RAW performance, D3s is far far superior in high ISO performance, its 6400 is better than T3's 1600.

.

PaulRivers Veteran Member • Posts: 7,420
Re: Canon EOS Rebel T3 image quality issue

AllOtherNamesTaken wrote:

As for this thread comparing it to a D3s that is a silly comparison to begin with and you can ignore it. They are not even close to the same type of sensor, price category, or performance world.

And that is just a silly statement. "It must be better because it costs more and the camera companies say it's better" is not what dpreview is about - if it was, why even take shots and do reviews? We'll just assume that anything priced higher is better and be done with it.

When comparing image quality you must look at the RAW image, otherwise you are just comparing in-camera JPEG processing engines, which is not a good representation of actual image quality because every company uses very different approaches to out-of-camera JPEG's, and the user can manipulate those values to make them all look basically identical. Some companies (i.e. Nikon) use far less default in-camera JPEG sharpening, but you can just turn it up if you want and it will look the same as everyone else.

Yeah, it would definitely be better if dpreview had RAW shots at iso's other than just iso200.

If you want to look at unbiased, standardized, sensor comparisons, check out the following link from DXO and select the competition (Pentax Kr and Nikon D3100 or whatever else you may be considering):

http://ns365755.ovh.net/index.php/Camera-Sensor/Compare/Compare-sensors/%28appareil1%29/693 |0/%28appareil2%29/664|0/%28appareil3%29/672|0/%28onglet%29/0/%28brand%29/Canon/%28brand2%29/Nikon/%28brand3%29/Pentax

dx0 may or may not be biased, but there is certainly some debate about how relevant their results are to the real world picture you end up with. They have had some...well, suspect results that don't seem to make sense of their own.

PaulRivers Veteran Member • Posts: 7,420
Re: T3 high ISO quality is no where near D3S'

tnnd wrote:

just checked out dpreview's article. For RAW performance, D3s is far far superior in high ISO performance, its 6400 is better than T3's 1600.

.

I am definitely not saying you are wrong, this should be the case. But where is this comparison? Do you have a link? I ask because, as mentioned, dpreview didn't seem to take RAW photos above iso200...

tigrebleu Senior Member • Posts: 2,060
Re: Canon EOS Rebel T3 image quality issue

maxnimo wrote:

Okay, I'm at the jpeg comparison test chart for the fourth time and I'm looking at the Rebel T3 and Nikin D3S side by side, both set at iso 3200. Both samples look excellent, with the Rebel T3 having just a bit more noise, one f stop at the most. For a cheap camera with no metering and focusing problems and an image nearly as clean as the D3S, this should be seen as a miracle of the decade.

It's not only the noise, but also the level of detail. Look at the hair in the paint brush and at similar objects, and you'll see much more details in the D3s image at ISO 3200 than in the T3 image. Noise is annoying, but if the noise reduction kills the detail, then you're sacrifying detail to noise reduction. And that's where the D3s beats the T3 hands down: detail retention.

However, since the T3 is brand new, it takes advantage of the latest imaging technology (including on-sensor and/or in-camera noise reduction, depending on the manufacturer). The D3s is about a year old now, and since then, noise reduction has improved a lot.

The Rebel T3 is indeed excellent and very capable at high ISO, but it's not better than the D3100 or the K-5, so I wouldn't call that a miracle... And the full frame D3s still has the upper hand, thanks to its large photosites. In fact, the 5DII, D700 and D3 are still superior to APS-C DSLRs because of that. But the gap is getting smaller... for now!

Because when the next full frame is released, it will also benefit from the same improvements in noise reduction technology that the latest APS-C DSLRs took advantage of, and it will again take a clear lead over APS-C cameras. I'm sure the upcoming D4, D800 and 5D Mk III (or whatever name they have) will offer at least a full F-stop noise improvement over their predecessors, if not two F-stops.

But right now, some APS-C cameras are getting indeed excellent at high ISO, so much that the difference with FF cameras is almost impossible to tell on normal size prints (4x6 and 5x7) up to ISO 1600. Not bad, considering for instance that the T3 is about eight times less expensive than the D3s!

The recent Nikon D5100 and D7000 are amazing in the high ISO departement, pretty much like the Pentax K-5, which is using the same sensor. Same goes for the EOS 7D, 60D and T3i, which are similar to their Nikon and Pentax competitors in terms of high ISO performance. The T3 is in the same leagues, and behaves pretty much like the D3100.

These cameras almost match up the full frame DSLRs at high ISO when it comes to how noise is kept low, but the FF sensors still offer more detail retention than any APS-C camera. Period.

So why are the reviews so negative? Something don't add up here.

I think the bad reviews have more to do with the global "how good is it as a photographic tool" than with just the high ISO and overall image quality performance.

The T3 lacks some of the features that you can find on competitors' models with similar prices, which is a bit disappointing.

The video, for instance, is only 720p HD, while the D3100 offers 1080p full HD. Not a meaningful difference if you don't care about video, but if you do, the D3100 looks like a more interesting product right from the start. Why Canon didn't put 1080p video in the T3 is still a mystery to me... Maybe to make sure the T3 was set well apart from the T3i?

But mainly, I think the reviewers were expecting more from Canon with this new product, for they were used to that in the past (think about the 7D). It's not that the T3 isn't good, it's just that it's not as good as some were expecting. The same thing happened with the Pentax K-r, which was excellent but didn't offer much more than the already excellent K-x. The same goes for the T3 vs. the T2i and T1i. And yet, both the K-r and the T3 are still great cameras.

I'm a photo store manager and I have the chance to try many DSLRs for different brands, and I can honestly say the T3 is an excellent camera. It has faster AF than a D3100 and it has great image quality.

But besides the AF, I also thought it didn't offer much more than a D3100, which is a bit cheaper, has a better warranty, better features and similar high ISO image quality.

I think this is mainly why the reviews were a bit hard on the T3, even though the camera is actually very good and quite capable. It just offers less than the other cameras in this segment, even though the price is similar. Canon could have put a little more effort in the T3, and I think that's why it falls short of being a "highly recommended camera".

Just my two cents.

-- hide signature --

If photography can be considered like painting, then I'm still at the preschool "paint with your fingers" level.

 tigrebleu's gear list:tigrebleu's gear list
Olympus E-M1 Olympus 12-40mm F2.8 Pro Olympus M.Zuiko Digital ED 60mm F2.8 Macro Phase One Capture One Pro HDRsoft Photomatix Pro +3 more
ultimitsu
ultimitsu Veteran Member • Posts: 6,650
Re: Canon EOS Rebel T3 image quality issue

tigrebleu wrote:

The T3 lacks some of the features that you can find on competitors' models with similar prices, which is a bit disappointing....

.... It just offers less than the other cameras in this segment, even though the price is similar. Canon could have put a little more effort in the T3, and I think that's why it falls short of being a "highly recommended camera".

that is exactly right, and it is amazing how some people just dont get it.

Nikon had the exact same problem with D3000 and D5000, they were a whole year late compared to competition (1000D and 450D then 500D) and they were not better, they both had difficulty getting good reviews and subsequently couldnt get recommendations. D90 was still getting more sales and attention than either of them despite more than a year older.

But nikon had managed to turn things around with D3100 and it is now canon's turn with a lacklustre product that is getting a "meh" from reviews (and hobbyists who make recommendations to their friends and family).

tnnd Junior Member • Posts: 26
Re: T3 high ISO quality is no where near D3S'

PaulRivers wrote:

I am definitely not saying you are wrong, this should be the case. But where is this comparison? Do you have a link? I ask because, as mentioned, dpreview didn't seem to take RAW photos above iso200...

here is the screen shot:

tnnd Junior Member • Posts: 26
Re: T3 high ISO quality is no where near D3S'
AllOtherNamesTaken Veteran Member • Posts: 3,563
Re: Canon EOS Rebel T3 image quality issue

PaulRivers wrote:

And that is just a silly statement. "It must be better because it costs more and the camera companies say it's better" is not what dpreview is about - if it was, why even take shots and do reviews? We'll just assume that anything priced higher is better and be done with it.

I'm not sure how you got that from what I said, especially since you even quoted what I actually said directly above your response. You simply put words in my mouth and threw your own quotations around it. I did not list price as the main reason it's better. The sensor is much larger (which, coincidentally, is more expensive to produce), designed specifically for high ISO performance, and the laws of physics definitely come into play there too. It's better because it simply performs better, regardless of all other variables. Besides, typically, in the camera world, when some thing is ~$4,000/4X more expensive than something else, typically it is better, but that's not the point. Can you name me 2 current cameras where a several thousand dollar price difference between the two does not reflect a difference in performance in favor of the camera costing 4X the money? There are none I'm aware of, but if you know of some I am curious. The point is you put words in my mouth, then responded to your own comment that you made up. I am sorry if you misunderstood my original statement.

Yeah, it would definitely be better if dpreview had RAW shots at iso's other than just iso200.

Agreed, I don't know why it isn't in there considering the full review was done long ago.

dx0 may or may not be biased, but there is certainly some debate about how relevant their results are to the real world picture you end up with. They have had some...well, suspect results that don't seem to make sense of their own.

Real world or not, they provide a standardized, professional, means of comparison between sensors. A sensor that performs better than another by a large margin in the lab isn't likely to have a reversed outcome in real world scenarios. As for DXO being bias, I have never heard that from anyone except people who own cameras that didn't perform as well as they wanted them to (Not saying that is you, just a general comment).

I think we can all agree the T3 doesn't come close to D3s ISO performance, ALL other variables aside.

Mark

Keyboard shortcuts:
FForum MMy threads