[joe.penn]

That is goddamn impressive.

That's more than impressive! That is one of the biggest advantages of FF - while other formats need 1.4 glass and massive PP (big noise reduction which equates to huge losses in DR, etc) just to [squeeze] and get by FF makes cake work of this with more readily available 2.8 glass.

[Pål Jensen]

Equals = FF resolves 67% Higher


I think ISO performance is more around a 3 to 4 stop advantage - I don't have hard numbers so not certain on that.

This cannot be the case. A 24mp sensor have twice the resolution of a 6mp sensor. Theres no way a 67% increase from 16MP APS to 23mp FF. This is proven by reality as the difference is hardly visible in a print. You need to quadruple the mp in order to make an obvious difference in a final print.

The image is noisy and soft and hardly useful for anything but web and newspaper publishing. I've seen equally "impessive" 6400 ISO images from the K-5...You can probably find them on this forums if you search...

[FrancisK7]

The image is noisy and soft and hardly useful for anything but web and newspaper publishing. I've seen equally "impessive" 6400 ISO images from the K-5...You can probably find them on this forums if you search...

Hardly. I'm on a calibrated U3011 and I can barely see noise... at ISO12800!

From a technical standpoint the bowling picture is a lot more impressive than what you posted.

[joe.penn]

Theres no way a 67% increase from 16MP APS to 23mp FF

Yep, there sure is - if you were to compare a 12mp FF vs a 12mp aps-c you would still see the huge difference in resolving power. An APS-C sensor is totally different than a FF sensor, and APS-C sensor is not a FF sensor cut down to size - your logic would state that a MF sensor is no different than a FF sensor or even a APS-C sensor when in fact they are all different no matter what the MP is.

Example: [imagine both of these rectangles are 12mp sensors]



Both being 12mp sensors, the output size would be the same so both would have the same amount of pixels - the difference is the information gathered per pixel which in return yields the higher resolution/resolving power. (no different than DR higher on FF because of size of sensor)

Here is a quick reference link to more of the science behind it -> True resolution - Learn - Snapsort (shows a couple illustrations)

True-Resolution is not a MP rating...

[ElJamoquio]

But the Canon is not a F:2.8 lens....

And of course you walk into an art gallery and point out the pictures that are shot with the F/2.8 lenses, right, Pål? I mean that's what this is about, what you can pick out in an art gallery?

What if I shot a scene in large format? Would I still need to use a f/2.8 lens? A point-and-shoot? f/2.8, again right?

You don't compare lenses between format from DOF wide open no more than you compare lenses within the same format from DOF wide open.

*YOU* don't compare lenses that way. **I** am concerned about comparing image quality, weight, and cost between two setups that will produce the same image, so I compare them at equivalent f-stops, instead of the same f-stops, and not necessarily wide open f-stops, as you continuously claim incorrectly. So **I** compare lens capability that way.

[Pål Jensen]

*YOU* don't compare lenses that way. **I** am concerned about comparing image quality, weight, and cost between two setups that will produce the same image, so I compare them at equivalent f-stops, instead of the same f-stops, and not necessarily wide open f-stops, as you continuously claim incorrectly. So **I** compare lens capability that way.

They will NOT produce the same image at "equivalent" F:stops. They will produce different DOF at the same F:stop but that is a property dictated by the format.

[Pål Jensen]

Here is a quick reference link to more of the science behind it -> True resolution - Learn - Snapsort (shows a couple illustrations)

True-Resolution is not a MP rating...

This is a misunderstanding of the link. They are comparing different sensors with vastly different pixel size. An 16mp APS camera and a 23mp FF sensor do not display vastly different pixel size to make a significant difference; the resolution difference is probably in the area of 12-15%. To double the resolution you need to quadruple the mp.
The Noise difference between an APS camera and a FF camera where the only difference is sensor size is one stop. However, if you, say, are shooting a bird image on APS with a certain DOF to get the whole bird in focus you not only need alonger lens on FF but you also need to shoot at one stop higher ISO to get the same exposure eating up the ISO advantage of FF. There's no free FF lunch...

---

Last edited by Pål Jensen; 08-07-2012 at 04:50 PM.

[Pål Jensen]

Hardly. I'm on a calibrated U3011 and I can barely see noise... at ISO12800!

From a technical standpoint the bowling picture is a lot more impressive than what you posted.

It is not a technically better image as it is both soft, grainy and noisy.

[ElJamoquio]

They will NOT produce the same image at "equivalent" F:stops.

In what meaningful way will they differ?

[ElJamoquio]

If you have the same sensor APS vs FF with the size the only difference the resolution difference is under 12-15%.

It's a gain of 50%, give or take. I'd like to see an example under 12%, please provide one.

[Pioneer]

I can't vote in this poll because none of the options really fit my situation. I am not "leaving" anybody since I have not married any camera brand. I like Pentax quite a bit and shoot several Pentax cameras, film and digital, but I shoot lots of other brand cameras as well. I am not worried whether Pentax brings out a full frame camera because I went ahead and bought a used Canon 5D, the first version. I actually really like that camera and shoot it fairly often when I need a digital image for something beside web posting or a quick shot to e-mail. I have held onto my K7 and a couple WR lenses but I rarely pull it out anymore. I do still use my *ist DL2 quite a bit because I like the image quality, and 6 megapixel is actually plenty for most of my digital imaging needs. If Pentax finally introduces a full frame I will be thrilled, but I seriously doubt that I will buy one now.

If you think about it, full frame is just 35mm, and I have a bunch of very nice 35mm cameras along with a scanner. I am actually more interested in the Q since that really looks like a top notch piece of gear. I would buy one now but I am afraid that Pentax will do with that like they did with the LX, and leave it hanging out there with no upgrades or additional lenses. If they do come out with a Q2, and a few additional lenses, I will be all over it. I may buy one anyway. IMHO that little camera is the real future of photography. Small, flexible, and really good images for such a small sensor. Obviously it can't compete with full frame images or even with the top of the line APS-C images. But in its class it is one of the finest. It can be a real groundbreaker, just like the Leica was in the 30s, if Pentax will only commit to the platform instead of letting it die. A few improvements, a couple more serious lenses, and I seriously believe that little camera will be on its way.

As for full frame? Lots to choose from and trying to catch up with Canon and Nikon from behind is a mistake in my mind. Instead Pentax needs to flank them. The Q and the K-01 are a great start, just keep those balls rolling. Pentax should keep aggressively developing that little Q and let everyone chase them for a change!!

[Pål Jensen]

It's a gain of 50%, give or take. I'd like to see an example under 12%, please provide one.

Sorry, that was a misprint. I was "thinking" noise. See edited text.

[Pål Jensen]

Yep, there sure is - if you were to compare a 12mp FF vs a 12mp aps-c you would still see the huge difference in resolving power. An APS-C sensor is totally different than a FF sensor, and APS-C sensor is not a FF sensor cut down to size - your logic would state that a MF sensor is no different than a FF sensor or even a APS-C sensor when in fact they are all different no matter what the MP is.
.

Of course there is difference between sensors. However, the difference between 16mp APS and a 23mp FF is not large by any definition and not easy to spot unless you do side by side comparison. The Nikon D800 is a totally diferent kettle of fish where FF start to make sense from a resolution point of view. A 24MP Pentax leaves me cold (I rather have a 24mp APS Pentax with one stop more noise at a third of the price).

[FrancisK7]

It is not a technically better image as it is both soft, grainy and noisy.

You must have a crappy monitor, because while it is soft in some spot, look at the shoe. It's quite sharp. This was obviously shot wide open. DoF is DoF. Your image is 42% smaller than the bowling image. If you resize it to the same resolution as the one you shared, noise is not visible.

Second, we're talking about a ISO12800 image. The K5 is renowned for its high iso performance, and even at 6400, I couldn't get a shot that looks like that.

You're obviously biased for whatever reason, and it shows in your posts.

[joe.penn]

Second, we're talking about a ISO12800 image. The K5 is renowned for its high iso performance, and even at 6400, I couldn't get a shot that looks like that.

Exactly - and you won't. The D3x has close to 2 stops (if not more) on the K5 in ISO performance. This is from DxOMark using their sports [Low ISO] ratings:

Sports Score is based on Low-Light ISO performance (values in ISO index). Low-Light ISO indicates the highest ISO sensitivity to which your camera can be set while maintaining a high quality, low-noise image (based on a Signal-to-Noise-Ratio [SNR] of 30dB, a dynamic range of 9EVs and a color depth of 18bits). As cameras improve, the highest ISO setting to produce 30dB, 9EVs, 18-bit images will continuously increase, making this scale open. Low-Light ISO performance is of primary importance in photojournalism, sports and action photography.

D3x -> 3253 ISO
K5 -> 1162 ISO

That's a pretty big difference going by the DxO rating. I recall seeing a chart from dpreview [I think] at one time that showed iso performance by camera model with the D3x somewhere around 3 stops better than the K5 (I could be mistaken though as this was quite some time ago). With all of that noted, this is really about FF vs APS-C [and not about brands and models], each has their strengths and weaknesses but one thing is certain, when you put all of the test data together, FF has the edge by a pretty big margin in both IQ and overall performance.