[Kunzite]

Well, I hope we won't see:
- 3" LCD, if that means moving the buttons (Canon-style). Higher resolution, yes (note that Pentax LCDs are quite good, but now everyone want the LCD from D300 - which could be quite expensive for a ~1000$ camera).
- Edge sharpening artifacts
- NR that destroy details, like Nikon (I don't care if it's luminance of chroma NR, the details are gone)

[mutley]

Well, I hope we won't see:

- 3" LCD, if that means moving the buttons (Canon-style). Higher resolution, yes (note that Pentax LCDs are quite good,

Obviously, the ergonomics need to be right, and buttons were not moved to the bottom on the D300, for example.

Actually Pentax chooses LCD screens that have been slightly lower in resolution such as in the K10D 2.5" 210,000 compared to others at 2.5" 230,000. Now, on the K20D the 2.7" is only 230,000.

but now everyone want the LCD from D300 - which could be quite expensive for a ~1000$ camera).

Samsung is a world leader in LCD's. As the K30D, like the K20D will most likely be $1300, and already on the similarly priced Sony A700 as well, I'm sure Samsung can provide themselves & Pentax with a killer deal on a 920,000 pixel 3" screen.

- Edge sharpening artifacts

Obviously this should be settable to different levels so you don't get those artifacts just like with the current only-choice textural sharpening. The point is it should be an alternate choice.

- NR that destroy details, like Nikon (I don't care if it's luminance of chroma NR, the details are gone)

Of course, the noise reduction I am talking about is the kind already on the chip for RAW files - just with a little more importance placed on minimizing chroma noise, not at the user level of destroying details.

[Kunzite]

There is no room on the K10D/K20D body for a 3" screen (without moving some buttons), the 2.7" one is at the limit. Of course, they could make a new, bigger body - but why fix it if it's not broken? IMO, the difference between 2.7 and 3" is mostly marketing.
That's the K20D list price and the A700 street price after several months on the market. The K20D price will/must go down - probably under 1000$ - while still making profit.

[mutley]

IMO, the difference between 2.7 and 3" is mostly marketing.

Its over 1/3" larger. For the size of the screen it is noticeable. The huge difference is in resolution. This should become the same (unless Samsung has the proprietary ability to make it better, as they might be good at making LCD's).

[lol101]

Of course, the noise reduction I am talking about is the kind already on the chip for RAW files - just with a little more importance placed on minimizing chroma noise, not at the user level of destroying details.

What's the point of doing chroma noise reduction on the RAW format?

Wouldn't it be better to leave NR out of raw and do stronger chroma noise reduction on jpegs only?

I figure that if you're going to use raw, you'd use a raw converter anyway and would thus have all latitude to set the chroma noise reduction to your liking.

Of course, any kind of luma noise reduction should be prohibited on raw format.

[Kunzite]

Mutley, of course 0.3" will be noticeable if you put the cameras one next to another - but IMO it's not such a big deal. Since I like the K10D body (and putting a larger screen means either a new - larger - body or moving some buttons) I won't ask for a 3" screen - the quality matter much, much more.
I agree that (at least on the rumored more upmarket model, hope on the K30D too) Pentax should start using hi-res LCDs.

[mutley]

What's the point of doing chroma noise reduction on the RAW format?

Wouldn't it be better to leave NR out of raw

There already is some necessary degree of NR on RAW files on all cameras. In CMOS it is applied at or near the chip.

All I was saying, is that if this is to be the case, then minimizing chroma noise should be given a little more importance as part of whatever level they are automatically using.

I agree also about better chroma noise reduction for the various in-camera NR settings since Pentax thought it to be a good idea to use more luminance NR in the settings than chroma, yet luminance is the one that loses details.

As difficult as NR algorithms are supposed to be, I still wonder if Pentax realizes how much more important minimizing chroma NR is compared to smeary luminance NR. I'm assuming they just don't get it.

[mattdm]

As difficult as NR algorithms are supposed to be, I still wonder if Pentax realizes how much more important minimizing chroma NR is compared to smeary luminance NR. I'm assuming they just don't get it.

I think it's pretty safe to assume that they hire engineers with at least a basic understanding of their field.

My guess is: when doing noise reduction at the CMOS level, there isn't the time nor computing resources to transform to a color space (of any sort) in order to do luminance NR; instead, it's just applied directly to the R, G, B, G photosite data (since that's all there is in raw form).

[ogl]

PopPhoto admit using K10D AF speed data in K20D review: Pentax SLR Talk Forum: Digital Photography Review
Did you see it????
"we decided to incorporate the AF results from the K10D to minimize confusion"???

how can we trust this site?

[ogl]

Why don't they make test with K20D with SDM lenses? Pentax's kit has no SDM.
Canon and Nikon have. I understand that SDM is not miracle and not helper in low light, but it's more honestly.
It will be more interested to compare Pentax AF speed with Canon/Nikon/Olympus.
'Cos I don't care AF of 18-55 with 14.6 Mp camera.
K20D are shiped in body version to the shops. No kit version.

[RiceHigh]

Why don't they make test with K20D with SDM lenses? Pentax's kit has no SDM.
Canon and Nikon have. I understand that SDM is not miracle and not helper in low light, but it's more honestly.
It will be more interested to compare Pentax AF speed with Canon/Nikon/Olympus.
'Cos I don't care AF of 18-55 with 14.6 Mp camera.
K20D are shiped in body version to the shops. No kit version.

Yes, I think this is the very first time I'm totally agreed with you.

Since they have got the DA*16-50 and reviewed it in the same issue. They should measure the AF speed with this lens on the K20D! How difficult is it to do this measurement??

Their excuse on since K20D and K10D share the same AF system (as officially informed by Pentax) don't seem anything convincing neither.

[ogl]

Yes, I think this is the very first time I'm totally agreed with you.

Since they have got the DA*16-50 and reviewed it in the same issue. They should measure the AF speed with this lens on the K20D! How difficult is it to do this measurement??

Their excuse on since K20D and K10D share the same AF system (as officially informed by Pentax) don't seem anything convincing neither.

Rice, my friends who have Canon 5D said (when they saw even beta-version photos from K20D till ISO1600) that Pentax/Samsung made the great job for APS-C sensor - resolution, dynamic range, noise reduction and details are on very-very good level. What do you think?
Can you say about your first impressions? Are you agreed?
Can you give us unprejudiced answer?

[mutley]

My guess is: when doing noise reduction at the CMOS level, there isn't the time nor computing resources to transform to a color space (of any sort) in order to do luminance NR; instead, it's just applied directly to the R, G, B, G photosite data (since that's all there is in raw form).

Yet, Nikon, to people's visual pleasure, has made it a point to figure this out while retaining details. Do their cameras have more computing power?

[thibs]

Yet, Nikon, to people's visual pleasure, has made it a point to figure this out while retaining details. Do their cameras have more computing power?

In fact you should read some DPR posts, Mutley, it seems that Pentax did indeed NR the chroma noise quite a bit but as usual you *know* don't you?

[mattdm]

Yet, Nikon, to people's visual pleasure, has made it a point to figure this out while retaining details. Do their cameras have more computing power?

Uh, yes. That's why they're faster and more expensive.