about DxOMark.com

[hurkmez]

Hi,

when i was reading forum about select camera. i found a link about compared camera www.DxOMark.com and then sign there select Canon 50D, Canon7D and 5Dmk II looked their result. 5Dmk II beat them and it's normal. But when i compared Canon 50D, Nikon D300 and Nikon D90 result is amazing. Nikon D90 beat them :S How is it? i can't understand.

http://www.dxomark.com/index.php/eng...(brand3)/Nikon

There's have a text like that too "Analyze performances of digital cameras and compare models
As digital cameras become more advanced, it is increasingly difficult for users to assess which model provides the best performance in real-world situations. In specially-designed tests, DxO Labs shows that digital camera RAW sensor performance in real-life situations varies widely and sometimes bears little relation to price."

is anybody know about how is posible. and how we can compare camera rightly,

thanks,

[Colin Southern]

You really need to look at the numbers in context. A Bugatti Veyron is faster than a Ferrari Enzo, but both are more than adequate for all tasks; same with the cameras. The dynamic range of all cameras today is more than we can display or print - the megapixel count is more than we need for all but the biggest of images - saturation & colour accuracy can be perfectly corrected for any camera model etc. So whereas I think DxO mark is great, you need to remember that it's about technical specs which NEVER equate to a linear improvement in image quality.

Have a read of this for a good practical example.

[hurkmez]

Hi Colin,

i read now it's amazing and then maybe i must to look composition, post processing and printing more than equipment

thanks a lot,

[Andy]

I don't think DxO mark is worth looking at personally. It says the 7D is rubbish(or so I hear) but the results say otherwise to me.

Trying to make sense of the rating would seem to me to be a waste of time as it has no practical use to me.

[Colin Southern]

Hi Colin,

i read now it's amazing and then maybe i must to look composition, post processing and printing more than equipment

thanks a lot,

No worries

At the end of the day, the image you end up with is the sum-total of the decisions you make along the way; camera decisions (capabilities, shutter speed, aperture, ISO), lens decisions (quality, focal length etc), composition decisions (a big one), post-processing decisions (saturation, sharpening, levels etc) etc etc etc. The "trick" is to make good decisions in all these areas; it's no good having a top-of-the-line camera if you make poor composition decisions just as it's not much good having only a phone camera if you have make excellent decisions in all other areas when you're trying to shoot something to hang on a museum wall.

Probably the biggest issue is that people get caught up too much in the specifications of the camera - it's a bit like buying a new car for everyday driving around town; the 195MPH Ferrari Enzo "sounds OK", but the 230MPH McLaren F1 has the better spec, and a 250MPH Bugatti Veyron is better still ... so looks like I need a Bugatti Veyron -- when of course a 70MPH Ford Focus would still be perfectly adequate. So in summary, if you have the money and you don't mind spending it then by all means go for top of the line gear - but don't necessarily expect vastly superior results from it if you're being held back by limitations in other areas.

[Colin Southern]

I don't think DxO mark is worth looking at personally. It says the 7D is rubbish(or so I hear) but the results say otherwise to me.

Trying to make sense of the rating would seem to me to be a waste of time as it has no practical use to me.

Hi Andy,

It's just a bunch of numbers, which I'm sure are accurate - but that only has the loosest of corelations with the results you might get. A radio controlled toy car would have rubbish specifications compare to a formula 1 car, but it would still win a race around the inside of my lounge; horses for courses

[hurkmez]

yep, i think so,

first of all i was looked equipments (lens) because i want to take portrait photos. and i haven't enough equipment. you know i asked a lens for this.

which lens,

now i have only canon 50mm prime lens. everybody advise to canon 70-200L. but i think if i have ff body it's right but i use 50D aps-c body. and i looking wider too.

and then i prepare 2 difrent set for select

1. with aps-c body 2270$
Canon 50D
Tamron 17-50 f2.8
Canon 50mm f1.8
Canon 70-200L f4

2. with ff body 2530$
Canon 5D
Canon 28mm f2.8 or canon 35mm f2
Canon 50mm f1.8
Canon 70-200L f4

but 5dmk2 so expencive for me. and then i looked 5d it's price good for me. But it's very old camera. i find DxOmark.com in the forum.

i have 2500$ and i can't decition now. :S

[Colin Southern]

Possibly you need to get the FF -v- Crop Factor decision made first. I've said this may times: "FF can be a blessing, a curse, or make no difference what-so-ever, depending (primarily) on your lens selection".

So first question must be "what would be the advantage in going to a FF camera for you"?

Next thing is Prime -v- Zoom. The old mantra is "primaes are sharper than zooms" but if you're talking L-Series then you'll find that the zooms are amazingly sharp and more than adequate, so that isn't a good reason. The other reason is that primes are often faster than zooms (a typical L-Series zoom is F2.8 whereas a couple of typical primaes in the same typical focal length are F1.2 / F1.4 (for L-Series), and F1.8 for non L-Series. Faster lenses give you an edge in low-light - but - at the expense of what can be a VERY tight depth-of-field. If you WANT a tight DoF then fine, but if you don't then it's no advantage and you're better of increasing the exposure by using a higher ISO or flash.

In terms of portraiture, it's going to depend on your envirunment - in a studio setup with strobes lack of light just isn't an issue (I typically shoot at F8 or F11) so zooms are more than adequate.

In your case, how would a 50D with a EF70-200 F4.0LIS USM & either a EF16-35mm F2.8L USM II or EF24-70mm F2,8L USM combo work?

[Shadowman]

Take a look at their ratings criteria and it would explain why the 50D scored lower.
Overview
This first tab shows product specifications and DxOMark Sensor scale and metrics results. The second tab, ISO Sensitivity, shows the differences between the manufacturer's published ISO sensitivity values and those measured by DxO Labs (following the ISO #12232 standard compliant method). As the ISO Sensitivity setting has a strong influence on measurement results, all subsequent Image Quality measurements are provided using the standards-compliant DxO Labs ISO sensitivity values for Signal-to-Noise Ratio at 18% mid-gray level (SNR 18%), Dynamic Range, Tonal Range, Color Sensitivity, the complete set of SNR curves (Full SNR), and (in the tab marked Full CS) the noise covariance ellipses used for assessing Color Sensitivity.