-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
K3 sensor toshiba...
- Thread starter Chris Mak
- Start date
Tan68
Veteran Member
Hey, wasn't What Digital Camera one of the sites that stopped reviewing Pentax because there was trouble getting samples ?
I may be misremembering...
However, if I remember correctly, there have been other changes if they are able to work with Pentax again. Or vice versa... I like WDC reviews.
I may be misremembering...
However, if I remember correctly, there have been other changes if they are able to work with Pentax again. Or vice versa... I like WDC reviews.
Received wisdom based on empirical evidence analysed by forum experts such as GordonBGood is that noise reduction in Pentax K-5 raw files kicks in at ISO 3200, not 2000. Furthermore, this noise reduction is relatively good at preserving detail.
Chris Mak
Senior Member
I found one more post by Asahi man. Put this together with the linked preview video's, ánd the term "brand new sensor" on the Ricoh website, and it really is looking like Pentax has put a new Sony sensor in the K3. I am going to disregard the "other forum administrator's" confidential information, and assume it is a Sony until someone finds out for sure. So time to start getting excited about the new AF, and how it will make the sigma 500/4.5 perform!!
Loading…
www.dpreview.com
Petroglyph
Veteran Member
- Messages
- 6,096
- Solutions
- 9
- Reaction score
- 1,384
Oh this isn't good news, unless Pentax can cook up some shadow noise magic. It looks like I'll have to crank up my copy of Topaz Denoise like I do for D7100 files. It makes about a 2X longer wokflow on shots above ISO400 (shadow noise and banding) but the results are fine. I had hoped they were wrong the Sony 24MP sensor is a really good starting point. Maybe Sony wants too much for the sensor?
Cheers.
paulkienitz
Veteran Member
According to DxOmark the Sony 24MP sensors are no better than the Toshiba, but those sensors are from 2011. Hopefully their current one is noticeably cleaner.
But it doesn't matter to me -- I need high megapixels more than I need low noise.
Petroglyph
Veteran Member
- Messages
- 6,096
- Solutions
- 9
- Reaction score
- 1,384
True. You can't create detail in post but noise can be fixed if it's at reasonable levels.
Tan68
Veteran Member
how do those tests respond to things like banding..?
even if overall noise level is the same, banding kills it. maybe the DXo accounts for it. i just don't know...
noise is one thing. banding scares me.
even if overall noise level is the same, banding kills it. maybe the DXo accounts for it. i just don't know...
noise is one thing. banding scares me.
Tan68
Veteran Member
RonakG
Forum Enthusiast
I'm not sure about that. But it's refreshing to see Pentax getting this much press coverage. Looks like Ricoh is doing all the right things in making sure most jurnos get a copy for review.
Nikon (correctly) chooses not to clobber the RAW files (or JPEGs for that matter if you turn off NR) with in-camera noise reduction. At least that's the case with the D7100.
The RAW files in particular preserve an amazing amount of detail and clean up beautifully in AfterShot Pro with NoiseNinja, right through ISO 6400.
I've had a D7100 for 3 months now and shot around 1,200 frames. I've only gotten it to band once, and that was at ISO6400 in ridiculously low light.
Here are some examples:
D7100 at ISO 6400
D7100 at ISO6400 again
D7100 at ISO6400 yet again
The last one is a little soft because it's slightly out of focus, but the noise cleaned up really well.
Alex Sarbu
Forum Pro
Let me understand. Noise reduction is bad if a very small amount of it it's done in the camera, above certain ISO values, but you'll be doing it anyway on the PC and then it's good?
The D7100 is an excellent camera, however I'm afraid the K-5IIs high ISO is slightly better - all factors considered.
Alex
torrent999
Active member
I suppose you might say that it's "bad form" ;-) on RAW files. After all they are supposed to be RAW, untouched, naked etc!Let me understand. Noise reduction is bad if a very small amount of it it's done in the camera, above certain ISO values, but you'll be doing it anyway on the PC and then it's good?
However, it's actually a bit moot because on the K5 the high ISOs past 800 are simply digitally brightened versions of the 800 ISO (approx) output. After 2000 ISO noise reduction is applied. So if you really want the absolute RAW you can always shoot at 800 (or at least below 2000) and brighten yourself. The NR that Pentax applies is not bad though, and it's not as if they are alone in this practice.
Joseph Tainter
Forum Pro
It's not clear which of the first two statements you consider to be "utter nonsense." If you are going to make such a statement, you should try to make sense yourself.
Your statement above is exactly what I said--manufacturers choose different levels of in-camera noise reduction. That is why comparison photos based on default settings are not a clear basis for deciding that a given sensor is or isn't good at high ISO settings.
Joe
britcam
Senior Member
If you take the measurements from DXOMark or Sensorgen and compute NSR and DR as per Joseph James writeup you'll see the that 7100 and K5 are pretty even in terms of noise, at low ISO and at high ISO There are a whole bunch of other factors that provide the look you like. For example Pentax's selection of filters on the array and how they work with them is bound to be different. This will be a factor in whatever "look" the K3 has. Another would be the in camera noise reduction used. Or the way that the DR is increased for JPEG's, tonal curve, shadow enhancement and the like. Some of that sleight of gives some of the noise you may perceive.
So with the K3 I would think that Pentax may do the same thing that they did with the K5 against the D7000 - slightly better raw performance by the use of unique hardware (filter array and the choice of processor like the new Milbeaut) and firmware choices. But the performance won't be markedly better and for most purposes pretty much the same. Which I'm happy with. The K3 looks like a product improved D7100 and thats a good thing as the D7100 is a very good camera.
As to sensor I would be happy with the Toshiba. If its a Sony I hope its not the same sensor that was used in the Nex7 but something more current and higher performing.
So with the K3 I would think that Pentax may do the same thing that they did with the K5 against the D7000 - slightly better raw performance by the use of unique hardware (filter array and the choice of processor like the new Milbeaut) and firmware choices. But the performance won't be markedly better and for most purposes pretty much the same. Which I'm happy with. The K3 looks like a product improved D7100 and thats a good thing as the D7100 is a very good camera.
As to sensor I would be happy with the Toshiba. If its a Sony I hope its not the same sensor that was used in the Nex7 but something more current and higher performing.
Leandros S
Senior Member
- Messages
- 2,165
- Solutions
- 1
- Reaction score
- 925
There were reasons Nikon chose the Toshiba version of the sensor, which afaik is entirely based on the earlier Sony design. The fact that it produced banding in the D7100 could be due to any number of causes. I'm fairly confident that Pentax would not have used the sensor if they hadn't solved or avoided the banding problem. As for a "filmy look", I never had that perception of the K-5 II over the D7100 - it must be subjective...
Sony has always been lousy in using its own sensors. See what Nikon and Pentax did to the same 16MP or 12MP Sony sensors and you will get the idea. DXO's D7100 sensor review shows that Toshiba's 24MP sensor technology is not quite there yet. Next generation maybe.
viking79
Forum Pro
Loading…
www.dpreview.com
The new comparison tool lets you select "print" in the upper right corner so you can compare the higher resolution camera on similar ground. Otherwise you are comparing pixel level to a camera with 50% more resolution. I set it to incandescent light as it is a little harsher, and ISO 1600 since Pentax smooths RAW at 3200.
Eric
Not sure about the D7100, because the images from that camera never impressed me as much as those from the D5200, which gets the highest DXO score ever: http://www.dxomark.com/index.php/Cameras/Camera-Sensor-Database/Nikon/D5200
Not sure the D7100 has the same sensor, but I have been blown away by the images I get from the D5200 with good glass. Here is the new Nikon 70-200 F4:
and here is the Nikon 24-85 (D600/D610 kit lens):
The dynamic range is awesome, and the images have very low noise in my experience, although I don't normally shoot in extremely low light. I agree with DXO that this Toshiba sensor is outstanding, and Pentax/Ricoh might make it work even better in the new K3
Not sure the D7100 has the same sensor, but I have been blown away by the images I get from the D5200 with good glass. Here is the new Nikon 70-200 F4:
and here is the Nikon 24-85 (D600/D610 kit lens):
The dynamic range is awesome, and the images have very low noise in my experience, although I don't normally shoot in extremely low light. I agree with DXO that this Toshiba sensor is outstanding, and Pentax/Ricoh might make it work even better in the new K3
Keyboard shortcuts
- f
- Forum