Low Light P&S vs mu 4/3

[huashan]

My Canon A95 has a max aperture of 2.8. My Olympus kit lens has a max aperture of 3.5. Since (3.5/2.8) squared = 1.56 my canon lens is half stop faster than the olympus. Does this mean the point and shoot has better low light performance with a given ISO?

If the olympus iso 1600 has the same noise level as canon iso 400. Which means two stop better. And the olympus IBIS give you half stop advantage. Does this mean the olympus is 2 + 0.5 - 0.5 = 2 stops better in low light performance than the canon?

Obviously other factor also affect image quality but I just want to quantify the low light performance.

 

[~tc~]

Yes, the Canon will let in more light, especially considering the depth of field difference due to the small sensor it can be wide-open for more shots.

Unfortunately, you are trying to quantify what, in the end, is a completely subjective factor - image quality. This world seems obsessed with trying to put a "number" on everything, but it really comes down to what looks better to you. Witness all the disagreements about DXO scores, for example.

IMHO, the low light performance of the most current compacts is approaching, or even surpassing, some of the older, larger sensors.... But you will still never have any control over DOF with such a small sensor, and Therfore your creative options are more limited with a compact than they are with m43

 

[usayit]

Its not that simple.... even the scientific measurments of dxomark have a lot discrepancies. The best way to examine is through visual inspection of samples. don't forget that as iso settings climb your dynamic range is reduced.... this is especially true with tiny sensors in point and shoots and jpegs.

I have the high end point and shoot lx3 which up until recently set the bar for quality compacts. It too has a fast lens at f/2. Even pass iso 400 its images start to to fall apart.... while an aging old 6mp Epson Rd1 can easily produce better quality at iso 1600.

Ultimately I always end up examining samples..... knowledgable camera shops would even goes as far show printed samples to examine with customers. Some cameras simply print better as well.

 

[JoelSutherland]

Sensor Size

You're complicating this too much.

The sensor in a P&S is about 9x smaller than on a m43 camera. So you can't just consider aperture.

The P&S with the lower aperature does let in more light over the given sensor area, but there is 9x more sensor (and light) on an m43 camera. This is why it performs better. It is also by a m43 camera is outperformed by an FX Nikon.

 

[~tc~]

Exposure is independent of sensor size. Only field of view and depth of field are affected.

 

[996gt2]

IMHO, the low light performance of the most current compacts is approaching, or even surpassing, some of the older, larger sensors....

I can't agree with this statement. My 5D completely annihilates my E-PL1 at >400 ISO. By ISO 1600 the micro 4/3 camera is looking very much like a watercolor painting, while the 5D's images still look great even without any noise reduction applied. All this despite the fact that the PL1 has 5 years worth of technological advances over the 5D.

 

[~tc~]

They have really only started to focus intently on high ISO recently. Compare the latest G12 or LX7 vs your EPL1. The gap is closing quickly

 

[huashan]

Exposure is independent of sensor size. Only field of view and depth of field are affected.

So a f2.8 is always faster than a f3.5, regardless whether it's mounted on a P&S or DSLR. and the brightness of the picture only depends on aperture, shutter speed and iso. It has nothing to do with the sensor size.

 

[Luckypenguin]

So a f2.8 is always faster than a f3.5, regardless whether it's mounted on a P&S or DSLR. and the brightness of the picture only depends on aperture, shutter speed and iso. It has nothing to do with the sensor size.

Correct - exposure does not discriminate between formats.

 

[Ulfric M Douglas]

... the low light performance of the most current compacts is approaching, or even surpassing, some of the older, larger sensors...

I can't agree with this statement. My 5D completely annihilates my E-PL1 at >400 ISO. By ISO 1600 the micro 4/3 camera is looking very much like a watercolor painting, while the 5D's images still look great even without any noise reduction applied. All this despite the fact that the PL1 has 5 years worth of technological advances over the 5D.

Hey 996GT2, you making a strawman? You're talking about Canon 5D DSLR vs Olympus m4/3rds when tc was talking about current compacts. Your statement is irrelevant in tc's context since neither camera in your example is a current compact.

 

[996gt2]

Hey 996GT2, you making a strawman? You're talking about Canon 5D DSLR vs Olympus m4/3rds when tc was talking about current compacts. Your statement is irrelevant in tc's context since neither camera in your example is a current compact.

The same argument still holds with compacts. Look at the best compact (in terms of high ISO) from a few years back...the Fuji F30fd with the SuperCCD sensor. Its high ISO performance is comparable to the LX3 or S90, despite being much older. None of these small-sensor cameras, new or old, can compare to a DSLR or even a Micro 4/3 camera such as the PL1 in terms of high ISO ability. Look carefully at some images from each camera and you'll see that the output of the compacts has much of the fine details gone due to heavy noise reduction even at ISO 400.

 

[lenshoarder]

Don't underestimate the low light ability of modern P&S cameras. I carry a TX5 around as my travel camera. I'm constantly amazed by it's low light performance as well as the surprising quality of the shots. Here's a thread comparing a TX5 to a M43 camera in terms of low light.

The Sony TX5 is amazing. - Dyxum forums - Page 1

 

[Djarum]

Here's what I think. If someone is toying with the idea of either mFT or a high end compact, I recommend a high end compact if the user:

1. Doesn't care too much about DOF. Doing scenery and landscapes
2. Doesn't want to use different lenses or likely just to keep the kit lens on. With the kit lens on an mFT, I think the user will be using one stop less ISO on the P&S and that one stop less ISO is about the difference in image quality between the two.
3. Don't mind dealing with a stop or more less of dynamic range.

 

[usayit]

Don't underestimate the low light ability of modern P&S cameras. I carry a TX5 around as my travel camera. I'm constantly amazed by it's low light performance as well as the surprising quality of the shots. Here's a thread comparing a TX5 to a M43 camera in terms of low light.

The Sony TX5 is amazing. - Dyxum forums - Page 1

Don't think so.... samples below. They speak for themselves.

* ISO 800 E-PL1

Digital Cameras, Olympus E-PL1 Digital Camera Test Image

* ISO 800 Sony TX5

Digital Cameras, Sony Cyber-shot DSC-TX5 Digital Camera Test Image


High ISO Pixel peeping should be done in a controlled environment looking at details maintained close up. Small online images are a form of noise reduction in of themselves. Each camera has their own form of noise reduction... It depends on personal tastes, details versus noise reduction.

There are adv and disadv to all.... its the reason why my LX3 still sees much use even though I have other cameras that are more capable. Best of the P&S bunch is still in my mind the LX3/LX5 and S90/S95. Both of which still fall apart past ISO 400... most others fall apart sooner.

* TX5 @ ISO 400
http://www.imaging-resource.com/PRODS/TX5/FULLRES/TX5hSLI0400.HTM

As far as I'm concerned, the TX5 is about par with other P&S... with aggressive NR being performed even at ISO200.

 

[lenshoarder]

Don't think so.... samples below. They speak for themselves.

...

As far as I'm concerned, the TX5 is about par with other P&S... with aggressive NR being performed even at ISO200.

Those images don't speak at all to the topic. This thread is about low light performance, not well lit images. That thread I posted addresses this topic. The images in that thread are 100% crops. The TX5 image is head and shoulders above the E-P1 images. They speak volumes.

 

[usayit]

This thread is about low light performance, not well lit images.

That statement doesn't make any sense.....

The results of two images taken in ideal conditions from two different cameras at the same ISO setting isn't going to reverse themselves when the light is less then ideal.


I'm sorry... I find the TX5 images on par with consumer P&S.... muddy at high ISO and low light. I doubt the dynamic range is any better for the JPEG only P&S.

Seriously, glad you like it.. the samples in well lit and controlled environments indicate that it is no comparison.

 

[sLorenzi]

Twilight mode

I believe the reduced noise in the TX5 is because of the twilight mode, wich means that, in fact the camera takes three photos in a row and then combines them to reduce noise... At least it's what I've read about it so far.


Best regard

Sidinei

Don't think so.... samples below. They speak for themselves.

* ISO 800 E-PL1

Digital Cameras, Olympus E-PL1 Digital Camera Test Image

* ISO 800 Sony TX5

Digital Cameras, Sony Cyber-shot DSC-TX5 Digital Camera Test Image


High ISO Pixel peeping should be done in a controlled environment looking at details maintained close up. Small online images are a form of noise reduction in of themselves. Each camera has their own form of noise reduction... It depends on personal tastes, details versus noise reduction.

There are adv and disadv to all.... its the reason why my LX3 still sees much use even though I have other cameras that are more capable. Best of the P&S bunch is still in my mind the LX3/LX5 and S90/S95. Both of which still fall apart past ISO 400... most others fall apart sooner.

* TX5 @ ISO 400
Digital Cameras, Sony Cyber-shot DSC-TX5 Digital Camera Test Image

As far as I'm concerned, the TX5 is about par with other P&S... with aggressive NR being performed even at ISO200.

 

[lenshoarder]

That statement doesn't make any sense.....

The results of two images taken in ideal conditions from two different cameras at the same ISO setting isn't going to reverse themselves when the light is less then ideal.


I'm sorry... I find the TX5 images on par with consumer P&S.... muddy at high ISO and low light. I doubt the dynamic range is any better for the JPEG only P&S.

Seriously, glad you like it.. the samples in well lit and controlled environments indicate that it is no comparison.

It makes perfect sense. I find it strange that you are basing your opinion by extrapolating from one set of conditions to another instead of using the data from the actual test case. When there is data available I tend to use that instead guestimating. It's like you are saying that a human can out accelerate a car to 60 MPH just because a person can do so to 5 MPH. Once again, we are talking about low light performance. The samples from that indicate there is indeed no comparison. Unfortunately, they aren't inline with your conclusions.

 

[lenshoarder]

I believe the reduced noise in the TX5 is because of the twilight mode, wich means that, in fact the camera takes three photos in a row and then combines them to reduce noise... At least it's what I've read about it so far.


Best regard

Sidinei

It's actually 6 images. I tend to default to using HDR mode which takes 2 images. It averages out some noise and increases detail in dark areas. The down side is that it can make things look too bright at night.

There something more intelligent than simple averaging going on. Using twilight mode when there is motion, such as when someone is walking across the field, there should be some serious blurring. There's not as much as expected. It's like it identifies that there are moving objects in the scene and only uses one of the frames for that portion of the composite image.

 

[Amin]

They have really only started to focus intently on high ISO recently. Compare the latest G12 or LX7 vs your EPL1. The gap is closing quickly

Sensor size continues to make a big difference in high ISO performance. It isn't the only factor, but it is a real factor. The Canon S90 sensor is on par (at least) with the LX5 sensor based on my tests, and here is the S90 sensor compared with the G1 (same sensor as the E-PL1): ISO Shootout: Canon S90 vs Ricoh GR Digital III vs Panasonic G1

There are no current small (less than 1/1.6") sensor cameras which can approach the high ISO performance of any of the Micro 4/3 cameras. Some of the small sensor cameras make up for that by featuring lenses which are much faster than our kit zoom lenses.