TDP Canon EOS Rebel SL2 / 200D Noise Test Results

[mukul]

http://www.the-digital-picture.com/Reviews/Camera-Noise.aspx?Camera=1141

To me it seem have better/lesser noise @1600 compared to SL1

http://www.the-digital-picture.com/Reviews/Camera-Noise.aspx?Camera=1141&Test=0&ISO=1600&CameraComp=835&TestComp=0&ISOComp=1600

 

[Maximilian]

To me it seem have better/lesser noise @1600 compared to SL1

Thanks for sharing!

At least one good news about this camera.
But honestly I would have been as disappointed about the noise performance as I am already about the AF system.

Please don't forget that the sensor of the SL1 was already some times old as it was reused in this camera.
So we have a 6+ years old sensor compared to the latest APS-C sensor from Canon.
If there weren't any improvements visible I'd be at least puzzled.

 

[mukul]

It seems SL2 is at par in noise @1600 with both 80D and 800D

http://www.the-digital-picture.com/Reviews/Camera-Noise.aspx?Camera=1141&Test=0&ISO=1600&CameraComp=1111&TestComp=0&ISOComp=1600

http://www.the-digital-picture.com/Reviews/Camera-Noise.aspx?Camera=1141&Test=0&ISO=1600&CameraComp=1044&TestComp=0&ISOComp=1600

 

[Maximilian]

It seems SL2 is at par in noise @1600 with both 80D and 800D

Should be because it's the same sensor, AFAIK.
So RAW should be the same. Only in-body-JPEG-processing could be different.

 

[LonelyBoy]

It seems SL2 is at par in noise @1600 with both 80D and 800D

http://www.the-digital-picture.com/Reviews/Camera-Noise.aspx?Camera=1141&Test=0&ISO=1600&CameraComp=1111&TestComp=0&ISOComp=1600

http://www.the-digital-picture.com/Reviews/Camera-Noise.aspx?Camera=1141&Test=0&ISO=1600&CameraComp=1044&TestComp=0&ISOComp=1600

Going back and forth a bunch it actually looks cleaner to me - less blotchy under Centimeters and in the darker Green, Yellow, Red, and Magenta boxes.

 

[Sarpedon]

It seems SL2 is at par in noise @1600 with both 80D and 800D

http://www.the-digital-picture.com/Reviews/Camera-Noise.aspx?Camera=1141&Test=0&ISO=1600&CameraComp=1111&TestComp=0&ISOComp=1600

http://www.the-digital-picture.com/Reviews/Camera-Noise.aspx?Camera=1141&Test=0&ISO=1600&CameraComp=1044&TestComp=0&ISOComp=1600

Going back and forth a bunch it actually looks cleaner to me - less blotchy under Centimeters and in the darker Green, Yellow, Red, and Magenta boxes.

Yes, and it's actually pretty noticeable when you compare definition in the numbers on the chart.

Hey, look at that: Canon improved image quality on a cheaper camera in the same sensor format!

(Just messing with y'all, don't @ me!)

 

[mb66energy]

It seems SL2 is at par in noise @1600 with both 80D and 800D

Should be because it's the same sensor, AFAIK.
So RAW should be the same. Only in-body-JPEG-processing could be different.

Same sensor should give same results ... if the (electronic) enviroment is the same. My feeling is too that the SL2 / 200D results seem to be marginally better then the over bodys which is great news because I want to replace 40D and 600D with one light and small body. IQ is paramount so the 9 pt AF system will be sufficient ... especially because of DPAF which gives much more creative freedom to choose the subject which has to be in focus.

If the IQ of the SL2 / 200D is comparable to the 5D classic I will keep it (ordered one and will compare it) for ultrawide shots with the EF-S 10-22 and not buy the 16-35 4.0 EF lens - I need/like ultrawide sometimes but I am more the tele photographer.

The SL2 / 200D won against M5 because it has similar size with EF / EF-S lenses (shorty fourty, EF-S 60mm macro and has less power consumption. I really like distance scales on lenses so the native EF-M lenses aren't for me. I will keep the old EOS M for my FD lenses which are great for scientific photography (easy setting of distance and aperture in dark environments).

 

[Mt Spokane Photography]

I just received my Sl2 Wednesday evening. I took a snapshot in low light while configuring the options menu to get it setup the way I like. Because light was dim, it came out at ISO 20,000. I had to double check because the amount of noise was a lot lower than expected. I don't recall if I had yet changed it from jpeg to raw.

The measure I use to compare them is the retention of detail as well s colors. Other than a few walking around the yard photos to confirm it works, I have not had time to finish setting it up.

My hand aches while holding it, I have large hands as well as carpal tunnel, so that may be a factor.

It takes a long time between live view shots, I need to find out why I have to wait a few seconds.

 

[Sharlin]

It seems SL2 is at par in noise @1600 with both 80D and 800D

Should be because it's the same sensor, AFAIK.
So RAW should be the same. Only in-body-JPEG-processing could be different.

The 24MP DPAF sensors are all very similar but almost certainly not the exact same sensor. Tweaks and improvements in design and fabbing happen all the time. PTP.net and DxO data, for instance, indicate that the the 80D has about 1/3 stop advantage over the M5/M6 over the whole ISO range. For whatever reason.

 

[Sharlin]

Going back and forth a bunch it actually looks cleaner to me - less blotchy under Centimeters and in the darker Green, Yellow, Red, and Magenta boxes.

Yes, and it's actually pretty noticeable when you compare definition in the numbers on the chart.

Looks like there's less color noise specifically, at least at higher ISOs? Bryan captures the test images with zero in-camera noise reduction and uses DPP for RAW conversion, apparently adding neither luma nor chroma NR in post either. But in real life you basically always want some base amount of the latter - for example LR has 25 dialed in by default and that rarely needs to be changed.

 

[mb66energy]

It seems SL2 is at par in noise @1600 with both 80D and 800D

Should be because it's the same sensor, AFAIK.
So RAW should be the same. Only in-body-JPEG-processing could be different.

The 24MP DPAF sensors are all very similar but almost certainly not the exact same sensor. Tweaks and improvements in design and fabbing happen all the time. PTP.net and DxO data, for instance, indicate that the the 80D has about 1/3 stop advantage over the M5/M6 over the whole ISO range. For whatever reason.

I think just if they use the exactly same sensor and have no variation due to the production process you will have different behaviour in different cameras.
80D has much more space, maybe some step-up or step-down regulators are placed far away from the sensor while M5 and M6 do not allow such a distant spacing due to package size. Or it depends on the frequency of the step-up/down converters how much noise is generated in the sensor photosites. This is just an idea but: After optimizing the power supplies for my Hifi amps and DAC I had a lot oddities in voltage variations before I had the final set-up. Electronics is not only about using the right components - its a lot about placing them well on PCBs or inside a system.

Maybe with the SL2 they found an optimum solution for component choice and placement ...?!

 

[mb66energy]

I just received my Sl2 Wednesday evening. I took a snapshot in low light while configuring the options menu to get it setup the way I like. Because light was dim, it came out at ISO 20,000. I had to double check because the amount of noise was a lot lower than expected. I don't recall if I had yet changed it from jpeg to raw.

The measure I use to compare them is the retention of detail as well s colors. Other than a few walking around the yard photos to confirm it works, I have not had time to finish setting it up.

My hand aches while holding it, I have large hands as well as carpal tunnel, so that may be a factor.

It takes a long time between live view shots, I need to find out why I have to wait a few seconds.

red: I have to wait at least until monday ... but I am more and more interested in that little cam

blue: While I have no carpal tunnel syndrome I expect a similar thing with my moderately large hands. But I have a 3D printer to create some grip extension Maybe it's a matter of finding the right finger position - I like to have the little finger under the camera to gain more control.

pink: Maybe a matter of AF mode (just an idea) - a simple "no-tracking" and "no face recognition" mode should be faster.

 

[Sharlin]

80D has much more space, maybe some step-up or step-down regulators are placed far away from the sensor while M5 and M6 do not allow such a distant spacing due to package size. Or it depends on the frequency of the step-up/down converters how much noise is generated in the sensor photosites.

I agree that it could just be a question of RF attenuation or maybe more efficient thermal design afforded by the larger body. Or perhaps higher quality parts used. Canon has said that the sensors are not identical, but of course we don't know what exactly they mean.

 

[Mt Spokane Photography]

I just received my Sl2 Wednesday evening. I took a snapshot in low light while configuring the options menu to get it setup the way I like. Because light was dim, it came out at ISO 20,000. I had to double check because the amount of noise was a lot lower than expected. I don't recall if I had yet changed it from jpeg to raw.

The measure I use to compare them is the retention of detail as well s colors. Other than a few walking around the yard photos to confirm it works, I have not had time to finish setting it up.

My hand aches while holding it, I have large hands as well as carpal tunnel, so that may be a factor.

It takes a long time between live view shots, I need to find out why I have to wait a few seconds.

red: I have to wait at least until monday ... but I am more and more interested in that little cam

blue: While I have no carpal tunnel syndrome I expect a similar thing with my moderately large hands. But I have a 3D printer to create some grip extension Maybe it's a matter of finding the right finger position - I like to have the little finger under the camera to gain more control.

pink: Maybe a matter of AF mode (just an idea) - a simple "no-tracking" and "no face recognition" mode should be faster.

I found part of the answer, apparently, the 2 second review time causes the camera to wait for two seconds when using the touch shutter release. When using the shutter button, its not a issue, it takes shots pretty much as fast as I can press the shutter. I tried setting the review time to 4 seconds, it took much longer before I could tough the screen to take a photo. However, even with the review turned off, there is a delay. Solution - use touch to focus and the shutter button to capture. I had noticed the same issue with my new 5D MK IV but had not spent any time quantifying it, I thought it might be a memory card thing. A low level card format makes no difference.

I'll be checking this out further on both cameras.

 

[mb66energy]

80D has much more space, maybe some step-up or step-down regulators are placed far away from the sensor while M5 and M6 do not allow such a distant spacing due to package size. Or it depends on the frequency of the step-up/down converters how much noise is generated in the sensor photosites.

I agree that it could just be a question of RF attenuation or maybe more efficient thermal design afforded by the larger body. Or perhaps higher quality parts used. Canon has said that the sensors are not identical, but of course we don't know what exactly they mean.

Thanks for this info - I never read that! So it's maybe a combination of both factors.

 

[mb66energy]

[...]

I found part of the answer, apparently, the 2 second review time causes the camera to wait for two seconds when using the touch shutter release. When using the shutter button, its not a issue, it takes shots pretty much as fast as I can press the shutter. I tried setting the review time to 4 seconds, it took much longer before I could tough the screen to take a photo. However, even with the review turned off, there is a delay. Solution - use touch to focus and the shutter button to capture. I had noticed the same issue with my new 5D MK IV but had not spent any time quantifying it, I thought it might be a memory card thing. A low level card format makes no difference.

I'll be checking this out further on both cameras.

Thanks for your research! DPAF is one reason to buy a new camera (besides of more DR, compactness and vari angle disp). Hopefully it is useful for macro work in windy conditions with some AF tracking of macro subjects. A touch to focus/shutter release for off center subjects is maybe the only solution to get usable images ... getting more and more problems not thinking about my SL2 on the way to my hands

 

[Woody]

The most sensible way to compare is to download the RAW files and run them through ACR (or something equivalent). This method yields rather similar results from Canon 77D, 80D and M5.

 

[foo]

I found part of the answer, apparently, the 2 second review time causes the camera to wait for two seconds when using the touch shutter release. When using the shutter button, its not a issue, it takes shots pretty much as fast as I can press the shutter. I tried setting the review time to 4 seconds, it took much longer before I could tough the screen to take a photo. However, even with the review turned off, there is a delay. Solution - use touch to focus and the shutter button to capture. I had noticed the same issue with my new 5D MK IV but had not spent any time quantifying it, I thought it might be a memory card thing. A low level card format makes no difference.

I'll be checking this out further on both cameras.

On my 5d4 it appears to be a UI related thing, it's too busy mucking about with flashing the shutter count icons etc.. Regardless of the review time there seems to be a 1-2 second delay before the Q and touch shutter icons return when you use the touch shutter. Also if the drive mode is set to H, you still only get single shot from the touch shutter... all of which suggests the touch shutter takes a different path from the physical shutter button.

It's not an AF or IS thing either, I tried with MF and IS turned off as well. Actually.... I'm wondering if with touch shutter it's waiting until the buffer has cleared?

Seems like another one for Jacks caveats thread...

 

[mb66energy]

I see the same behaviour in my SL2 / 200d which arrived yesterday.

Maybe - while this is observed in 5d iv too - it is the delay between configuring the sensor between Live View Mode and Stills Capture Mode (this is my wording not Canon's so I have no internal information). In the live view mode the sensor bins berhaps several photosites together to reduce the resolution the digic processor has to manage for a 30Hz display with ... reduced resoultion. If the sensor has some on-board electronics, setting another mode needs some time to propagate. To have clean analog signals one has to wait until the sent digital signals effects on analog circuitry are close to zero.

While writing this I checked, if MF will shorten the lag - it doesn't so AF accuisition as main reason for the lag can be excluded.

[...]

I found part of the answer, apparently, the 2 second review time causes the camera to wait for two seconds when using the touch shutter release. When using the shutter button, its not a issue, it takes shots pretty much as fast as I can press the shutter. I tried setting the review time to 4 seconds, it took much longer before I could tough the screen to take a photo. However, even with the review turned off, there is a delay. Solution - use touch to focus and the shutter button to capture. I had noticed the same issue with my new 5D MK IV but had not spent any time quantifying it, I thought it might be a memory card thing. A low level card format makes no difference.

I'll be checking this out further on both cameras.

Thanks for your research! DPAF is one reason to buy a new camera (besides of more DR, compactness and vari angle disp). Hopefully it is useful for macro work in windy conditions with some AF tracking of macro subjects. A touch to focus/shutter release for off center subjects is maybe the only solution to get usable images ... getting more and more problems not thinking about my SL2 on the way to my hands

 

[mb66energy]

About the threads topic:

I observed that the grip of the SL2 / 200d is getting warm after a few minutes of use. My first thought was: Oh it sucks a lot of energy from the battery, if the battery gets that "hot".

But the battery isn't located inside the grip but in the body structure itself. Maybe they placed the digic vii and other electronics like the voltage converters inside the grip. If the sensor has ADCs on chip which can be assumed it is much easier to separate sensor and processor because only digital signals have to be transmitted which works lossless (information-wise).
The region behind the sensor (under the display) was not warm - opposite to my experience with the 600D / T3i(?) .

If that is the case: Kudos to the engineers of Canon ...

First checks with images show very precise tonal values and good capability to lift darker areas to overcome high contrast scenarios. I do that not just for fun but to regain my visual experience from the raw data of the camera.