On 4/2/2010 11:12 AM, John Sessoms wrote:
From: P N Stenquist
On Apr 2, 2010, at 10:23 AM, Adam Maas wrote:
> That they would, but even better would be a K-7x with the K-x
> processing chain and either the K-x's 12MP sensor or Sony's closely
> related 14.6MP sensor. I'm seriously hoping we're going to see that
> soon (perhaps for Photokina?).
>
> I'd really love to have the K-x's IQ with the K-7's metering and AF
> (metering in particular is a weak point of the K-x).
>

I'm not so sure I'd want that. The K7 seems to have better detail
rendering than the Kx at low to moderate ISO. Good low light
performance is a nice plus, but for most work it's secondary.

So, how about a camera that combined the K7's metering, AF, and low to
moderate ISO performance with the K-x's high ISO performance?

And since we're off wishing in la-la land anyway, it'll be an 18 - 20 MP
full-frame sensor in 135 format as well for well under $2K. ;-D

Today's wishlist:
D700 performance
K7 ergonomics and features
Katzeye focusing screen
K-x size and price
Japanese K-x shell customizing options

I expect that it'll be available sometime around 2015.

Next year's wishlist will be something along the line of:
D3s low light performance
645D low ISO performance
etc.

For computer performance I use a metric I call "Craymarks" which is what year the most powerful computer in the world had that level of performance. I wouldn't be surprised if the K-x processor has a craymark of about 1980 or later.

For cars, I use "Vettemarks", which is in what year the top of the line Corvette had that level of performance. Note that a lot of econoboxes these days probably have vettemark performance in the 1980s.

I don't know what to call the equivalent metric for cameras, but note that the K-x will probably match the image quality of the best 135 lens format camera of not much more than 5 years ago. So there's probably about a 5 year lag in performance between top of the line bodies and entry level bodies.



--
PDML Pentax-Discuss Mail List
[email protected]
http://pdml.net/mailman/listinfo/pdml_pdml.net
to UNSUBSCRIBE from the PDML, please visit the link directly above and follow 
the directions.

Reply via email to