Canon 300d vs. Panasonic Lumix FZ20

[BryanMsi]

Hopefully nobody objects to the cross-post...I also posted this in the Panasonic area but I thought it would be of interest to both Canon and Panasonic owners...

I have been a amateur photographer for a few years, the last of which has been with the Canon Digital Rebel as my camera of choice. After reading all the reviews raving about the Canon quality I was starting to think I was crazy...the Rebel never really lived up to the hype for me. For one, its metering system is severely (and intentionally) limited by Canon and I find that it tends to underexpose pictures. For another, it rarely seems to produce the "tack sharp" photos I was expecting (I will admit to not owning any "L" lenses, but a Canon 28-135 IS, the 18-55 Kit lens, and a Sigma 28-300 APO zoom). But when it nails an image, the quality is great...and of course Photoshop works wonders on almost-but-not-quite-perfect shots.

So I bit the bullet earlier today and bought the Panasonic Lumix FZ20 – Circuit City had it for $539 plus a free bag and battery (worth $100). My thought was that the Lumix would serve as a "little brother" to the Canon for casual photos (vacations, museums, etc.) where I don't want to lug around the rather large dSLR.

For grins, I decided to compare the two cameras directly. Let me just say that I was shocked to see the results.

I took some sample photos for comparison and was hoping the quality of the Lumix would come close enough to the famous Canon sensor that I wouldn’t be disappointed with the Lumix. Instead, I found the Lumix consistently besting the Canon shots – both in exposure metering and in overall quality!



You can see these shots at the following address (I'll try to post full size images if anyone is interested):

http://homepage.mac.com/bmacdonald/P...toAlbum31.html

For starters, the Canon consistently got the exposure wrong – 2/3 of the shots I took were underexposed which is not all that unusual for this camera. The first shot (Canon vs. Lumix 1) shows both cameras at their exposure defaults and clearly the Canon shot is badly underexposed (the Canon is on the left). The Lumix consistently metered the shots correctly.

But, I adjusted the Canon using my handheld light meter so I could get comparable pictures.

Canon vs. Lumix 2-4 show shots taken with the cameras on tripods approx 2 inches away from each other with the same composition (or as close as I could come). I put them side-by-side in Photoshop with no adjustments. You be the judge…

Oh – image stabilization is off on the Lumix, and I was using a Sigma 28-300 lens on the Canon (which may be part of the problem, but that’s comparable to the lens on the Lumix, at least in terms of focal length).

That Leica lens really seems to rock! A Canon lens comparable to the f2.8 435mm image stabilized zoom built into the Lumix would cost thousands of dollars.

The details
Canon
Sigma 28-300 APO zoom
Aperture priority – 5.6
Exposure compensation +1.6 – 2.0 (except shot #1, that shot is +0)
Metering – not selectable (default)
ISO 100
White Balance - Cloudy
High quality JPG
Approx. zoom – 300mm (x 1.6 = 480)
Image Stabilization – N/A

Lumix
Leica 35-435mm (equivalent)
Aperture priority – 5.6
Exposure compensation – none
Metering – Center-weighted evaluative
ISO 80
White balance – Cloudy
High quality JPG
Approx zoom – 435mm
Image Stabilization - off

Surprising results, to say the least. I expected the Rebel (outside its rather weak metering) to blow away the Lumix and that just didn't happen. I think there are 2 things going on here. First, the Sigma 28-300 APO lens is simply no match for the Leica glass on the Lumix. Second, the Canon's metering doesn't do the sensor justice. Put both together and you have murky, soft images vs. properly exposed and tack sharp images.

If someone wants to send me a Canon L lens, I'd be happy to repeat the test!

-Bryan

[Rprincess]

I have been actually trying to decide which one of these cameras to buy. I have read tons of reviews. I think buying a car is much easier then buying a digitial camera! I finally found a display model to check out and I was impressed how fast it zoomed. However I told the store that I wanted to think about it tonight and make my decision on whether I was buying the FZ20 or the Rebel. You have helped my agonizing decision a lot easier.


Nice Photos, I love the comparisons that says it all! But I do love the landscape shot you took with your canon, that was beautiful.

[timmciglobal]

First let me say, the exposure thing is infuriating but can be helped. For one, I used the tool FECSet to set flash exposure comp to +2/3, helps alot. I also usually set exposure +1/3 to 2/3 depending where I'm shooting.

Now as far as the difference, keep in mind the quality at > ISO 100 in canon is far superior to the lumix, as is the ability to crop pictures and print with them.

Simply put the end result is the Lumix is a good point and shoot, and at worst the rebel is a poor professional SLR. The core issue with the lumix will always be image quality, something no level of automatic shooting downfall or greatness will overcome. +2/3 Exp won't "fix" noise at ISO 200 in lumix.

Tim

[BryanMsi]

Tim, you're quite right about the noise...at ISO 400 the Canon blows the Lumix away. But at ISO 80, the Lumix isn't much noiser than the Canon at 100 - both are perfectly useable at low ISOs.

For someone who shoots high ISOs, the Lumix (or any small-sensor digicam) simply isn't an option. I tend to shoot nature/landscapes and have the luxury of getting by with low ISOs.

Just playing around, I would say the Canon at ISO 1600 and the Panasonic at ISO 400 are comparable, noise-wise. And the fast 2.8 lens on the Lumix negates some of the need for higher ISOs.

But I still say at low ISOs...for these test shots...the Lumix is delivering better exposure and more detail than the Canon. Which should suggest 3 things:

1. A person could be pretty happy with a digicam under the right conditions (low ISO)

2. The Sigma 28-300 lens stinks

3. Canon needs to fix Rebel metering

[Mike Woods]

First let me say that I believe both of these are excellent cameras. I think, however your test is not altogether accurate and here is why. In most of the canon pictures, there is nothing in the image sharp, meanining that there is a problem here. A little camera shake comes to mind here, although I am not saying that is the problem. If there was one part of the picture that was sharp and the rest muddy looking, then I would agree with your test. When the entire picture looks this way, either the camera is not vey good, the lens is not very good, or there is some other outside cause. I have the Rebel and most of my pictures are very, very sharp. The kit lens produces a little softness in its images, but nothing like what you are showing. If my camera consistently produced images like these, I would throw it in the trash and try something else! I am not saying that you do not have some problems with yours, it could be the lens or the camera itself, but my results with the kit lens, the 50mm f 1.8 lens and the image stabilized 75-300 lens are different from yours.

[BryanMsi]

Mike - I think the softness is due in large part to the Sigma lens. The camera was on a tripod with a shutter speed of around 600 (it was fairly bright outside) and I used the remote trigger to eliminate all vibration. Short of mounting on bedrock, the camera couldn't be more stable.

I find my Canon 28-135 USM IS lens generally delivers sharper images than the Sigma, but not dramatically sharper. I have heard you really need an "L" lens on the Canon to deliver great images but I don't have one to verify.

[D70FAN]

Originally Posted by BryanMsi

Mike - I think the softness is due in large part to the Sigma lens. The camera was on a tripod with a shutter speed of around 600 (it was fairly bright outside) and I used the remote trigger to eliminate all vibration. Short of mounting on bedrock, the camera couldn't be more stable.

I find my Canon 28-135 USM IS lens generally delivers sharper images than the Sigma, but not dramatically sharper. I have heard you really need an "L" lens on the Canon to deliver great images but I don't have one to verify.

Not to put too fine an edge on things, but I think the lens in question was a Tamron 28-300, not a Sigma. I bought the Tamron for my D70 and returned it, as it just never seemed to produce consistant results. Hopefully the Tamron 18-200 due out in the spring will have fixed the probelms of the 28-300.

I now have the Sigma 18-125 DC and find it to be a very good day-to-day lens. I have officially retired the Nikkor 18-70 as I can see no difference compared to the Sigma.

Just my 2 cents worth.

[brad nichol]

I have been following this rebel vrs FZ20 thread and I would just like to make a couple of points that potential buyers and users could consider. First up let me say I do not own an FZ20, I do own a FZ3 and 300D and have used an FZ10. I caution all about making definitive judgements based on screen images and especially those compressed for efficient web loading etc, my comments are therefore not in relation to the images attached to this forum.

1) When making comparisons of noise and sharpness comparing images at identical apertures is not valid. For example if you set the Panasonic at f2.8 you will probably need an aperture of at least f8 on the 300D to get a similar depth of field. Now this brings up an interesting point regarding noise, because in real world use the FZ20 could be set on say 80iso and the 300D would need to be set on approx 800iso to maintain a similar shutter speed once depth of field is factored in.

2) The change in lens performance from one aperture to the next with most digicams is small but with regular consumer grade 35mm lens the gaps can be big enough to drive a truck through. Many zooms do not reach optimum performance until f11 or smaller, further negating the noise debate.

3) Never underestimate the lack of mirror slap in the scheme of things, that flapping mirror no matter how well damped can cause a fair bit of softness in even a tripod supported camera, by the way the 300D is no paragon in this area and to get around it you need to change the firmware using a 3rd party product, that allows you to actually lock the mirror up.

4) Image stabilization rocks and in reality a superzoom digicam without it is pretty useless in real world use, unless you choose to go the tripod mounted route.

5) Pixel wise DSLRs do not have a huge advantage. A 6 meg DSLR in real world use if you are thinking in terms of finished prints is roughly the equal of a 5 meg digicam. Most people print to 8 by 10 and 5 by 7 formats, this means that a considerable portion of the DSLRs frame is cropped off, which in fact means that only about 5 megs worth of the original image gets printed, in the case of the digicam for regular prints (assuming you framed properly in the first place) there is almost no cropping.

6) The Panasonic FZ series cameras take a very minimal approach to noise limiting which means that you get more detail and a bit more noise, all this is good especailly if you use some specific noise reduction software. I have found that Noise Ninja utterly fixes any noise issues, in fact the marriage of noise ninja and FZ series cameras is incredible.

The bottom line is this for many users a stabilized superzoom digicam is going to beat a 300D or most other DSLRs under many real world circumstances, especially at the telephoto end of things and all at a fraction of the price without having to lug a tripod around.

[timmciglobal]

Simply put, you're just wrong.

For starters, a 6.3 megapixel CMOS sensor in the rebel is not "equal" to a 5 megapixel digi cam, it's not "equal" to a 8 megapixel digi cam. I can get far greater quality due to image sharpness and lack of noise out of a rebel image then ANY P&S on the market today.

As far as ISO use, again, what are you comparing it to? Daytime shots? I'll show you some ISO 400 shots I've taken indoors in poorly lit banquet hall which for a fact there is no possible way the lumix could produce the same quality, low noise image.

Keep in mind also that you've got to consider lens. Take a canon 28>135 USM IS lens to the rebel and then compare the quality you're getting. Comparing the stock lens on the rebel vs a camera which clearly doesn't have the same versitility is on it's face wrong. The shots with my rebel are not significantly better then my friends kodak 6200 2 megapixel POS 100$ camera when printed at 4X6 or 5X7 full frame.

You take a lumix picture of a group, crop it in by 40% in anything else but perfect daylight ourdoor shooting then compare it to a rebel, that's where you'll see that there is truly no comparison. They are two totally different products and the key is determing which you want. For a point and shoot the lumix is a GREAT camera, but it is only a point and shoot. You don't want to crop, you don't want to use it in poor lighting situations or at night, you have limits. The rebel's only stock limit is if you don't buy another lens or you don't learn how to use the camera.

Tim

[timmciglobal]

This is the crop which I made great prints from, and the original scaled down to show.







Tim