Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
09-25-2013, 06:24 AM   #61
Pentaxian
Fogel70's Avatar

Join Date: Sep 2009
Location: Stockholm, Sweden
Photos: Albums
Posts: 1,737
QuoteOriginally posted by Smeggypants Quote
how often would you need more than 16MP?
Not often, but as long as other IQ parameter than resolution increase with more MP, then I want higher pixel count on new cameras.

09-25-2013, 06:28 AM   #62
Pentaxian




Join Date: Sep 2011
Location: Nelson B.C.
Posts: 3,211
QuoteOriginally posted by Clavius Quote
But would it give current Pentax K-5, K-5II, K-5IIs owners any reason to upgrade? The sensor-shift gimmick would have finally been a reason to turn on SR.
A few things would make me upgrade, and I probably will be buying this fall. Better autofocus, USB3 support. I shoot K5 with SR off most of the time; it works on stationary objects ok, but any movement or the need to quickly focus and shoot or panning to follow causes problems. If it is fixed so that I can have it on all the time, so much the better. Higher resolution would be nice, AA less is also a plus. The K5iis is very tempting, have almost bought a couple of times over the last few weeks, but I'll wait to see what this version looks like. I'd really like a faster responding body as well. I'm getting the impression that there may be two bodies coming out, one pretty well an upgrade of the K5II, an incremental improvement, and something a bit higher end with more in the way of processing power. Maybe more resolution. A fully working movie mode would be very nice.

A friend upgraded from the D7000 to D7100. A worthy upgrade in resolution and improvements in autofocus but the buffer size just plain sucks. That part was a downgrade for him.

Someone on this site said that his D800 was never the problem when it came to his shots, it is so good. Hopefully we can say that about the next K*.
09-25-2013, 08:31 AM - 1 Like   #63
Pentaxian
thibs's Avatar

Join Date: Jun 2007
Location: Belgium
Photos: Albums
Posts: 5,159
One thing I'd really like, specially with a high res sensor is a sRAW-like function.
Flexibility or RAW but lower res and when we're at it, pixel binning.
09-25-2013, 08:42 AM   #64
Site Supporter
Eric Auer's Avatar

Join Date: Jul 2012
Photos: Albums
Posts: 861
QuoteOriginally posted by derekkite Quote

Someone on this site said that his D800 was never the problem when it came to his shots, it is so good. Hopefully we can say that about the next K*.
I can say that right now about my K-5 IIs.

Unfortunately for Pentax, I like mine so much I will not update easily.

A compact mirrorless with the same sensor as my IIs would have me adding another body to the herd tho.

09-25-2013, 09:27 AM   #65
Junior Member




Join Date: Dec 2012
Location: Trondheim
Posts: 27
I wonder whether we may be in for a surprise. There have been occasional indications that the K-3 will make Pentaxians happy (how is that possible?) and turn heads (even my K-01 does that). The specifications "leaked" point towards a very solid upgrade, but some have already described it as conservative. Could there be something more? The single head-turning feature rumoured so far would be the 40 MP trick, certainly stunning if well implemented. But I am fascinated by the "3 user modes" mentioned in the leak. Is it simply the usual user modes, and why would that be important enough to mention as a spec? Let us assume (optimistically) that this camera will really turn heads, that Pentax has been working on the technology for quite some time and that there is a 40 MP mode. Could the 3 user modes refer to 1) Standard, 2) Enhanced resolution, 3) Enhanced something (how about high ISO performance?). I believe there was a camera (Fuji?) some years back with such aspirations, but apparently not a great success. Could it be done better today? For me this is science fiction, but so is shake reduction and handheld HDR technology.

These thoughts are pure speculation, based on wild rumours, lofty dreams, and a bit of logical (?) reasoning.

For me personally, a well-made camera along the lines leaked is already a head-turner. I am not a likely upgrader though because I am very happy with my K-5ii, and super-happy with my K-01. Even when I bring both, I often prefer the K-01. In many situation it takes better pictures, not because the technology is better, but because it looks so innocent.
09-25-2013, 11:20 AM   #66
Site Supporter




Join Date: Apr 2010
Location: Rancho Cucamonga, CA
Photos: Albums
Posts: 472
QuoteOriginally posted by Tesfaye Quote
In many situation it takes better pictures, not because the technology is better, but because it looks so innocent.
That's so true. I was allowed into a big sporting event with my K-01 because it didn't look like a PRO DLR. I had a bag of lens and my 55-300. They examined it but still let me in.
09-26-2013, 02:26 AM   #67
Pentaxian
Clavius's Avatar

Join Date: Feb 2011
Location: De Klundert
Photos: Gallery
Posts: 4,115
I wonder if the 40mp-through-a-trick is a side-effect of the "software AA filter"? Normally, on single-shot images, such a software AA filter is impossible, because moiré is impossible to remove. (Unless you PP it out aftwerwards.) But for multi-shot images things are different. I don't believe it applies to multishot techniques that involve a sensor shift. Look at the results from Hasselblad H4D 200MS. It uses such a multi-shot sensor shift trick and is also without AA filter over the sensor.

QuoteQuote:
Moire is generated when the calculation/estimation of the correct pixel color based on the single-color R,G,B values delivered by the individual photosites of the bayer array fails. It can result in geometric patterns of incorrect colors that are visible to the naked eye because of the nature of the bayer sensor pattern. When you shift the sensor, even a half pixel, you mix up that Bayer pattern and make it much finer and more accurate. By bumping neighboring color photosites closer to the other respective colors, you're able to much more accurately calculate the correct colors of individual pixels because you're now getting information on that area from more than one photosite. So they may be calling it a software solution, but really what I think they would be doing is using a much more sophisticated demosaicing algorithm based on the additional information from the second or additional shots.
So it may be they stumbled upon an added advantage whilst searching for way to prevent moiré through a software AA?

[/whishfull thinking]
09-26-2013, 04:08 AM   #68
Senior Member




Join Date: Sep 2010
Location: Innsbruck
Posts: 283
that in fact would be a nice idea.

Now just imagine someone would think that through!

Pixel binning
super resolution
software AA

wohooo

09-26-2013, 04:56 AM   #69
Pentaxian
Uluru's Avatar

Join Date: May 2010
Location: Back to my Walkabout Creek
Photos: Gallery | Albums
Posts: 2,400
QuoteOriginally posted by Clavius Quote
The rumor says: "Pentax K3 with 40 MP via sensor shift". Not: "Pentax K3 with 40 MP via software shift". Please see the opening post.
Or,
"Shift Pentax via software K3 with 40MP".
Available for download for your Pentax K5II, K5IIs, K5, K-01 and Q.
09-27-2013, 02:00 AM   #70
Senior Member
troenaas's Avatar

Join Date: Apr 2009
Location: Norway
Photos: Gallery | Albums
Posts: 130
Original Poster
What about sensor shift combined with in camera AA software filter on/off setting?
09-27-2013, 02:12 AM   #71
Pentaxian
thibs's Avatar

Join Date: Jun 2007
Location: Belgium
Photos: Albums
Posts: 5,159
What about using a K5 (or K5-IIs ideally) and using composition tweaking feature as to get those images:
* normal untweaked sensor position (1)
* left and right (2)
* put the sensor a tad higher height and do the same, left, middle, right.
* same for down position.

You get 9 images you can blend. There should be some more resolution to get out of this...
09-27-2013, 05:02 AM   #72
Veteran Member
Louicio's Avatar

Join Date: Mar 2012
Location: Melbourne
Posts: 432
QuoteOriginally posted by vonBaloney Quote
That's not exactly what it (would be) doing -- it is just shifting a few pixels so the bayer array can hit its different color filters in the same spot of actual space, so it is more like turning your Bayer sensor into a Foveon sensor so that all 3 colors are captured at each point instead of only green, red, or blue at each point (where the rest is normally filled in by interpolation). So we are talking about increased colors and resolution with more MPs -- not creating a bigger image like when you stich together a panorama. There are already Hasselblads that do this -- see link earlier in thread.
Yes this. +1. Everyone else is carrying on about all sorts of random stuff in this thread. This would be awesome. Foveon clarity without the drawbacks. Shouldn't need a tripod either since the SR system is active, if the movement of the camera is within the capability of the SR it will just adjust. Prob a slight slower shutter speed maybe or high shutter speed limit? Would be awesome for all sorts of amazing clarity of images on a new AA-less 24MP sensor. Yes yes yes.

And then do it on a mirrorloess FF for a unique Ricoh/Pentax lol had to be said.
09-27-2013, 08:38 PM   #73
Pentaxian
Uluru's Avatar

Join Date: May 2010
Location: Back to my Walkabout Creek
Photos: Gallery | Albums
Posts: 2,400
QuoteQuote:
This would be awesome. Foveon clarity without the drawbacks.
This isn't as easy as it seems. Foveon-like technology is still worth pursuing, because that would be the ideal.

In Bayer sensor pattern, RGGB, there are twice as many green pixels as blues and reds. One or two sensor shifts won't help to establish a pure RGB readout in every pixel. Not at all. See the picture below. After two sensor shifts — one left, one up — the situation is like this:



There is still no true RGB pattern recorded in each pixel of the final image, but 50% of them do have it, and other 50% have recorded additional colour too. Which is outstanding, and I'd presume, at some 83% of the performance of the ideal Foveon-like sensor if it had hi-ISO performance of the current top of the line CMOS Bayers (current Foveon implementation can never come close to it).

So will this blow the socks out of current Foveon sensors? Almost for sure, especially after ISO 400. However, even without the complete RGB readout for each pixel, the sensor shift increases the resolution as well, and it could be said that a sensor-glide-APS-C camera with a Bayer sensor of 20MP can rival the resolution of the current MF 40MP Bayer sensors with no sensor glide tech.

Last edited by Uluru; 09-27-2013 at 09:47 PM.
09-27-2013, 09:17 PM   #74
Veteran Member




Join Date: Mar 2009
Posts: 336
QuoteOriginally posted by mamethot Quote
Has anyone here ever saw a raw file from a phone camera?

I have not. I searched the matter though and what I was able to find are mentions that the raw file isn't good at all and that your phone camera is only usable because of software correction.
At the risk of heading off topic, there's an app called 645 Pro that saves TIFF images, which many people regard as almost the same as RAW.
09-27-2013, 09:44 PM   #75
Veteran Member
scratchpaddy's Avatar

Join Date: Feb 2013
Location: Phoenix
Posts: 2,360
QuoteOriginally posted by Uluru Quote
There is still no true RGB pattern recorded in each pixel of the final image, but 50% of them do have it, and other 50% have recorded additional colour too. Which is outstanding, and I'd presume, at some 75% or more of the performance of the ideal Foveon-like sensor if it had hi-ISO performance of the current top of the line CMOS Bayers (current Foveon implementation can never come close to it).

So will this blow the socks out of current Foveon sensors? Almost for sure, especially after ISO 400. However, even without the complete RGB readout for each pixel, the sensor shift increases the resolution as well, and it could be said that a sensor-glide-APS-C camera with a Bayer sensor of 20MP can rival the resolution of the current MF 40MP Bayer sensors with no sensor glide tech.
What would be the point of high-ISO performance if you have to wait through several tripod-mounted exposures of a completely static scene? I would take a single shot over that any day, even restricted to ISO 400 or less.

However, I will say that the current 14.6 MP Foveon sensor puts out about as much detail as a 30 MP Bayered sensor, not the 46 MP figure Sigma puts on its spec sheets. It's still no 645D.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
k3, mp, pentax, pentax k3, pentax news, pentax rumors, sensor
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Which Pentax compacts have sensor shift SR? p38arover Pentax Compact Cameras 5 08-08-2013 06:13 AM
$2000 Superb Pentax 645D 40.0 MP with DFA AL IF SDM AW 55mm Sliver-Surfer Pentax Price Watch 23 07-12-2013 12:12 PM
Official Pentax K3 Full Frame 24 MP Den Pentax Full Frame 16 04-07-2013 02:09 AM
Dynamic Range of Pentax 16 mp sensor Steven Chain Photographic Technique 11 01-22-2013 04:20 PM
Native ISO of the 14.6 MP CMOS Sensor (with !shocking! attempt to lighten mood!) solar1 Pentax DSLR Discussion 22 01-27-2008 06:15 PM



All times are GMT -7. The time now is 01:56 AM. | See also: NikonForums.com, part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top