E-330 review - better ISO than a Canon ???

[Paul Shields]

DPNow has a review of a production E-330 on their website. Interesting (and slightly odd) finding with regards to noise was that it bettered the Canon 350D at all ISOs, and even more odd was that the Canon came bottom on all tests.

Personally I find that very unusual as we all know Canon have good control of high ISO noise (and I'm not being sarcastic - I agree with most people's views that Canon lead in this area). Will be interesting to compare other reviews when they come out.



Quote from review: "Nevertheless, signal to noise ratio figures only tell half the story. The Canon EOS-350D/Rebel XT has a good reputation for visible noise suppression and yet its s/n results are the worst here. Meanwhile, the Nikon D50 is visibly the best."

[Mr. Pickles]

Very interesting. I would assume because of this, the DPReview has some more fodder,a nd that these guys are in a witness protection program for posting the numbers...

[Paul Shields]

Originally Posted by Mr. Pickles

Very interesting. I would assume because of this, the DPReview has some more fodder,a nd that these guys are in a witness protection program for posting the numbers...

True...

I'm sure it's just a misprint or something.

[Mr. Pickles]

Might not be. Why can't they be a bit better at engineering a chip than the Canon engineers? Usually the bigger the company, the more average the staff. Remember, big companies don't usually buy big companies. They buy the little guys, because they can do what the big ones can't, or don't want out on the street.

[BigB]

How is this possible? From what I've seen the Canon XT ISO 800 and 1600 blow everything else away (at least my E-300) in terms of visible noise. If s/n ratio is a culprit, then why does the article state that the Canon has best noise suppression, yet worst s/n ratings?

[Paul Shields]

Dunno... I do hear arguments that Oly has more noise but better detail at high ISO, so don't know if that s/n figure is measuring that.

[dougjgreen]

What's MOST striking about these numbers to me, is how close the scores are for ALL of these cameras. Essentially, in real life, there is no meaningful difference between ANY of them. Buy the one that has the features and the lensmount you need, because they are all effectively equivalent on these scores, as in EVERY case, there is no more than a 10% difference between the best of them and the worst of them.

[Raist3d]

I think these conclusions and others like imaging resource's prove to me beyond the shadow of a doubt that the DXO and Imagetest (or are they the same?) that imaging-resource , and these people are now uses, are near useless to truly measure what's important.
I don't care what the test says, I know when I see an ISO 1600 of a Canon and the Oly, I know who looks better.

These tests are starting to look to me like the Pentium 4 si-soft-sandra memory tests. Sure, the Pentium 4 shows a whopping 2x-4x memory bandwidth over the Athlons (forgot exactly how much) but what the memory bandwidth test is not telling you is how much memory the Pentium 4 is throwing away due to pipelines inefficiencies, making the Athlon architecture in true real world tests come right ahead.

Have you notice how imaging resources conclusions after posting the ImageTest charts always start with an apology of how the Nikon D70 in real life seems to have more dynamic range and how the e-300 shouldn't have as much as it seems to have considering its noise?

I honestly don't know why they bother if they are going to start apologizing for the results!

- Raist

[zoegy73]

Those data are Signal to noise ratio. E330 may or may not have low noise level but those data show that E330 have excellent signal gain capability. That comparison used starting package of each camera (body and lens kit), and we knew that Canon lenskit is ****. Maybe, that's why 350D had low Signal to noise ratio, because the lens could not deliver good amount of signal onto sensor surface. That's just my thought,


Just my Rp. 200,-