Lumix noise comparisons with Fuji

There have been lots of comments around the noise "problems" of the Panasonic cameras, so I posed a few comparisons of Lumix cameras (TZ1 and FZ20) with a Fuji F20. Fuji's are well known for their low noise "SuperCCD" sensors. As you will see, there is NO COMPARISON in noise levels between these sensors.

The Fuji uses a 6 MPixel Super CCD (hexagonal, partially overlapping proprietary sensor pixel design).
The Lumix FZ20 uses a 5 MPixel traditional CCD (sensor pixels in rows/columns)

The Lumix TZ1 uses a 6 MPixel traditional CCD but crops only the center portion to produce a 5 MPixel image.

The first picture shows a comparison of the Panasonic TZ1 (left), Fuji F20 (middle), and the Panasonic FZ0 (right). All cameras on tripods, no flash, ISO 200. The Fuji is clearly less noisy and more clear than the Panasonics, even at ISO 200.



This picture shows the Fuji at ISO 1600 (staggering for an point-n-shoot) and the Lumix TZ1 at ISO200. The Fuji SuperCCD is delivering at least 3 f-stops better sensitivity than the Panasonic CCD. But what is also obvious is that the Panasonic has much higher chroma noise than the Fuji. Because of that, the Fuji more closely resembles film grain while the Panasonic looks very artificial. The Fuji may even have a similar amount of noise but its far less obvious since its mostly luminance noise.



This last picture is a 4-way comparison. Upper left is the Panasonic TZ1. Upper right is the Fuji F20. Lower left is a Nikon D70 thrown in for chuckles. Lower right is Panasonic FZ20. All cameras were on tripods, no flash, ISO 400. The Nikon and Fuji are clearly in an entirely different league here with both being completely clear and useable. The Fuji has a bit more noise, but not much more. The upper left TZ1 is the worst with details being blurred away by noise reduction. The FZ20 in the lower right is a bit "noisier" but with less noise reduction; I prefer it and it is clearer than the TZ1. I printed these at 4x6 on an Epson Photomate and the quality difference between the Panasonics and the others was visible.

--
David
[email protected]

Could you provide us the EXIF data? It would be nice to be able to compare the effective ISO.

Sure, it would be great if Panasonic used the Fuji sensor or came up with a better one on their own. But, fortunately for me, I do very little indoor low-light photography, which is really the only place the Fuji's sensor would offer much improvement for me.

To add my two cents to this thread, the fuji super CCD can also handle more lateral contrast than the Panasonic sensor, I have two Fujis, and one Panasonic so I know.
--
Steve Owen.

GregGory said:

Could you provide us the EXIF data? It would be nice to be able to
compare the effective ISO.

Click to expand...

Lemme see if I can find them- I actually made these comarisons a few weeks ago and then made images from the compared results. I saved the composite images with the results and noted the camera settings, but I may have subsequently deleted the original files. They are probalby lurking in iPhoto somewhere....or a backup.

Well, that does it. I've had enough. 500 threads about how bad my camera's sensor is have convinced me to just throw it out and buy a FUJI. Not.

No don't throw it out, its not the cameras fault!. Seriously I still use the Pany more than the Fujis, its just like in the days when I took 35mm slides. As long as you know your cameras limitations, you can always reposition yourself for better light, or just pass some really bad stuff by - regardless of how nice of a bad exposure it might be. Or just come back when the light is in your favor.
--
Steve Owen.

Thanks, comparisons are always interesting.

Whilst you commented on the the pixel arrangement, you forgot to
mention the much more important pixel size, i.e. sensor area/MP count.

The F20 has a MUCH larger sensor, but that's also why it has a DULL
zoom range and a SLOW tele end. You could use the mid-portion
of the F20 sensor in the TZ1, but then you'd have 2MP or something
like that. So you are comparing apples vs. oranges.

Bryanmsi said:

The first picture shows a comparison of the Panasonic TZ1 (left),
Fuji F20 (middle), and the Panasonic FZ0 (right). All cameras on
tripods, no flash, ISO 200. The Fuji is clearly less noisy and
more clear than the Panasonics, even at ISO 200.

Click to expand...

It's also a bit darker, less contrasty and softer, but granted, even if
we made it the same, 80% of the difference would probably remain.

Bryanmsi said:

The Fuji SuperCCD is delivering at least 3 f-stops better sensitivity
than the Panasonic CCD.

Click to expand...

No, you can't draw that conclusion from JPG. Fuji JPG processing is
usually better, so the sensor difference is smaller than you assumed.
And it's more a matter of pixel pitch (size) than it's CCD vs. SuperCCD.

And the Fuji shot is darker. If the Panny ISO 200 really is ISO 250,
that makes the difference smaller.

Here's a fair comparison of sensor tech, since the pixel pitch is the same
and we're looking at raw. From videozona.ru, FZ30 vs S9000.
Indoors:





Outdoors:





The Panny is also more sensitive since it uses a shorter shutter speed
to get an equally bright shot.

Just my two oere
Erik from Sweden

I think the Fuji has the edge in a little finer noise grain, if you want to call it that, for the lack of a better name. But the biggest difference is really a firmware one, contrast, and noise. Panny just needs to study Fuji tech manual.
--
Steve Owen.

I'd be happy to see Pansonic go to a better sensor, or at least do a firmware upgrade for the n.r. on some of their cameras. But really, for most of us, it's just not that big an issue. I don't shoot pictures in darkened rooms without a flash, for example. If that is the kind of photography I was into, i.e. indoors / available light-party-type, then I'd want a Fuji.

It is true that the Fuji sensor is larger than the Panasonic sensor in this example. You're right about that. The Fuji is 1/1.7 vs the 1/2.5 size Panasonic. If I do my math right (and I'm not promising I did) that's about a 32% size premium for the Fuji. I think Fuji get's some gain "for free" given the shape of its sensor pixels - less wasted space between pixels on the sensor itself.

However, the size difference isn't enough to explain the results entirely. Which is even more annoying in some ways, because it means Panasonic is choosing to use small sensors in order to offer small lenses or wide angles knowing it will compromise quality. If Panny used Fuji's sensors, it would proably only get a 4MPixel image out of the center of the sensor. And really, the TZ1 would be a fine 4MPixel camera.. I'd actually prefer a clear 4MPixel camera.

It would also appear Fuji is using dramatically more effective noise reduction, if I interpret your samples correctly.

Even if the difference is 1/3 sensor size, 1/3 better CCD performance, and 1/3 better noise reduction, the end result is the same. Panasonic needs to spend some more time focusing on image quality and less time chasing paper benchmarks like MPixels. As Phil noted in his TZ1 review, there is noise visible at ISO 80 and ISO200 plus is pretty well useless....

Bryanmsi said:

It is true that the Fuji sensor is larger than the Panasonic sensor
in this example. You're right about that. The Fuji is 1/1.7 vs
the 1/2.5 size Panasonic.

Click to expand...

The TZ1 only uses the middle 5MP out of a 6 MP chip, so it is effectively
a 1/2.8 size.

Bryanmsi said:

If I do my math right (and I'm not
promising I did) that's about a 32% size premium for the Fuji.

Click to expand...

Yes linearly vs the FZ20.
In area, it's 75% 121% bigger than the FZ20/TZ1's sensors.

http://www.dpreview.com/learn/?/Glossary/Camera_System/sensor_sizes_01.htm

1/1.7" area: 43.32 mm^2
1/2.5" area: 24.71 mm^2
1/2.8" area: 19.64 mm^2 (my own calculation based on focal length)

Using the F20 sensor:
6.1*24.71/43.32 = 3.5 MP (for the FZ20)
6.1*19.64/43.32 = 2.8 MP (for the TZ1)

Bryanmsi said:

I think Fuji get's some gain "for free" given the shape of its sensor
pixels - less wasted space between pixels on the sensor itself.

Click to expand...

It might have a shape better formed after the microlenses according to
some old Fuji marketing material. Don't know the validity of this.

It seems they have lower readout noise and more sensitivity to blue.
(Indoor light is very deprived of blue, that's why Panny has to
amplify the blue channel very much and it gets very noisy.)

Bryanmsi said:

However, the size difference isn't enough to explain the results
entirely.

Click to expand...

Absolutely not. But comparing to the F20 family isn't apples to apples.
People will get the wrong impression. A big sensor has its drawbacks to,
it's not for free. Looking at equal sized pixels there is still a difference
but not as dramatic.

Bryanmsi said:

Which is even more annoying in some ways, because it
means Panasonic is choosing to use small sensors in order to offer
small lenses or wide angles knowing it will compromise quality.

Click to expand...

What annoys me most is that there is no way to turn off their
clumsy noise reduction in JPG. Programs like Neat Image can do
a very good job with chroma noise without killing much detail if
they get the unsmeared noise to work from.

Bryanmsi said:

If Panny used Fuji's sensors, it would proably only get a 4MPixel
image out of the center of the sensor. And really, the TZ1 would
be a fine 4MPixel camera.. I'd actually prefer a clear 4MPixel
camera.

Click to expand...

As we saw above, it would be 2.8MP. It would be a very hard sell
in today's market. A 10x zoom is not very useful indoors anyway.
If it is used outdoors to shoot birds etc. more pixels are welcome
for cropping room. I think something in between the F20 and the
TZ1 pixel pitch would be suitable, your 4 MP, perhaps.

Bryanmsi said:

It would also appear Fuji is using dramatically more effective
noise reduction, if I interpret your samples correctly.

Click to expand...

Yes, dramatically. It is slower though, and they have no time to
remove the chromatic aberrations that Panny do.

Bryanmsi said:

Even if the difference is 1/3 sensor size, 1/3 better CCD
performance, and 1/3 better noise reduction, the end result is the
same. Panasonic needs to spend some more time focusing on image
quality and less time chasing paper benchmarks like MPixels.

Click to expand...

Unfortunately MPs sell better than per-pixel image quality.

Bryanmsi said:

As Phil noted in his TZ1 review, there is noise visible at ISO 80 and
ISO200 plus is pretty well useless....

Click to expand...

Yes Simon (sic) is right about that, I agree it's hard to get the full
5 MP's worth, and that's why I have resisted getting a TZ1 even if I
like compacts and tele reach a lot. I've seen many nice downsampled
shots on this forum, though.

Just my two oere
Erik from Sweden

As many on here knows I have been saying for months about the Fuji Sensor's and the Pana Sensors. I have said that there truly is not that much difference in the two sensors except for size, but as you see in the raw pics they are pretty close in noise levels, thusly I have said the in camera processing is better on the Fuji, it is not so much the sensor but that Fuji has the processing right, Pana only needs to see what Fuji does in the in camera processing and incorporate that into the Venus III engine, which is what Pana has said they have done, I don't mean copy the fuji but have made differences in how they handle the noise and resharpening. I will have to wait till we get some reviews on the new cameras to see if they finally got it right.

Panasonic has done it before with Venus Engine II (look at the detailed pictures of the FZ20 and FZ5). I don't know what happened to their engineers. Maybe Fuji bought them out, lol.

It seems that a lot of manufacturers are going towards "less noise, but acceptable" images that are not to be viewed full screen. Canon's new DIGIC engine on the SD800IS produces the same, watercolor effects, along with Casio and Samsung's offerings.

I think a major point that is missing out is the two approaches in photography between Fuji and Panasonic. Fuji relies on a clean, SuperCCD to take sharp photos at low light levels, while Panasonic relies on OIS.

It would be nice to see the EXIF data to see the shutter speeds and such..

But thanks for the comparison!

--
Best Regards,

Danny



My Albums
http://photobucket.com/albums/a44/yeeonly

Midwest said:

If
that is the kind of photography I was into, i.e. indoors /
available light-party-type, then I'd want a Fuji.

Click to expand...

No you wouldn't, they don't have image stabilization.

--
FZ7 + TC-E17ED

steve owen said:

I think the Fuji has the edge in a little finer noise grain, if you
want to call it that, for the lack of a better name.

Click to expand...

The noise of the FZ50 is quite fine; it's the JPEG engine and the converters that low-pass-filter the noise in such a way that it looks like it is in large clumps.

--
John

ay John, how comes you got a FZ50? Panasonic looking for advice?

--
nowhere - now here

Yawn . . .

Another noise discussion and people wishing Panasonic would put Fuji sensors in their cameras!

If you want a Fuji, they do make cameras, you know!

--
J. M. Daniels
Denver, Colorado
Panasonic FZ10, FZ50 & Fuji S602Z owner & operator