Ask for Pentax DSLR internal craft improvement (II)

The box is probably a telephone line or some other cable junction box. It looks to be metal and should be grounded. Not a likely RF source anyway.

The repeat of this test is a good indication that your 'pod is acting as part of an antenna system. It may not make a difference whether it is grounded or not that it makes interference to the camera depending on the frequency, especially if higher, of several hundred MHz.

Most evidence points to the change from the nucore ADC and possible improvements on on-sensor black point sampling which was the probable cause of the banding......

The tradeoff for the nucore was the banding vs the k10d having a large dynamic range.
--
360 minutes from the prime meridian. (-5375min, 3.55sec) 1093' above sea level.

'The exposure meter is calibrated to some clearly defined standards and the user needs to adjust his working method and his subject matter to these values. It does not help to suppose all kinds of assumptions that do not exist.'
Erwin Puts

gazonk said:

But K10D is nevertheless LESS noisy than the D200 and most other
cameras that use the same 10mp Sony sensor, including all (I think)
Sony and Nikon 10mp models.
--

Click to expand...

it is really:



Did you really think this is merely a firmware's level question and not other?

Gamma=2.0
--
Regards
PeterZheng

ISO=800

ISO 800

PeterZheng said:

it is really:



Did you really think this is merely a firmware's level question and
not other?

Click to expand...

Please note that you are comparing processed jpegs here! Please read what dpreview itself writes about this: "At ISO 800 the D80 looks cleaner, a glance at the graph below tells you why, it's when Nikon's noise reduction kicks in properly and that also has some effect on detail."

For RAW output, the K10D has slightly better S/N ratio and significantly better DR at ISO 800 than the D80, just look here:

http://www.dxomark.com/index.php/eng/Image-Quality-Database/Compare-cameras/ (appareil1) 212|0 (appareil2) 201|0 (onglet) 0 (brand) Pentax (brand2) Nikon

Here's a revealing comparision, the salt grinder of the imaging resource ISO 800 test shot. The D80 shot is on the left, of course. Can you tell if it contains milk or salt? From the K10D shot, it's obvious, but not quite so from the D80 shot...



--
Espen

PeterZheng said:

Kursun said:

You say there is no rfi shielding on the K10D. But a large and thick
shielding is clearly visible on the back cover in your picture.

Click to expand...

It is not enough, this shield is for its own LCD and so on. you
have seen the EOS400D its back cover?

Pentax did not have any metallic shield craft onto PCB as its EMC/EMI
processing.

Click to expand...

True, but don't forget that the PCB is mounted on a heavy stainless-steel internal frame, so perhaps Pentax determined that it was sufficient to shield most electronic emissions. The EOS400D is all plastic. Not to say that more shielding couldn't help, it is a demonstrated fact that electronic emissions will affect the image, but I'd like to see a shielded vs non-shielded K10D field test to demonstrate the real impact on noise for this particular camera given its internal frame. The noise seems pretty random and as shown by others, electronic interference seems to be creating pattern noise rather.

--
Roger

rogerstpierre said:

True, but don't forget that the PCB is mounted on a heavy
stainless-steel internal frame, so perhaps Pentax determined that it
was sufficient to shield most electronic emissions. The EOS400D is
all plastic.

Click to expand...

please don't forget a right fact, The EOS400D has a larger main metallic framework!

gazonk said:

Please note that you are comparing processed jpegs here! Please read
what dpreview itself writes about this: "At ISO 800 the D80 looks
cleaner, a glance at the graph below tells you why, it's when Nikon's
noise reduction kicks in properly and that also has some effect on
detail."

Click to expand...

yes, it is jpegs. but it is also reflect the noise, Did you really think this is merely a Nikon's noise reduction better and not other?

PeterZheng said:

yes, it is jpegs. but it is also reflect the noise, Did you really
think this is merely a Nikon's noise reduction better and not other?

Click to expand...

Of course it's the noise reduction! The chroma noise reduction is Nikon's speciality. But it can reduce detail, especially in highlights it seems (which Nikon has a tendency to blow anyway) - please look once again at the salt grinder and tell me honestly if you can tell if it's milk or salt!

And have a look at the dxomark.com site, where they measure D80 noise at ISO 800 to be higher than on the K10D.

--
Espen

K10d metallic framework is not a full length, perhaps It will cause the serious damage.

look here:
http://docs.google.com/Doc?id=df2k6qz2_10c2wnq2fm

Eos 400d has a full length metallic framework:





--
(I am sorry for my poor English to could hinder your reading)
my article is also here:
Ask for Pentax DSLR internal craft improvement (II)
http://docs.google.com/Doc?id=df2k6qz2_89db3xzcdq
Ask for Pentax DSLR internal craft improvement (I)
http://docs.google.com/Doc?id=df2k6qz2_10c2wnq2fm
FSViewer 3.5 as well as UMPC
http://docs.google.com/Doc?id=df2k6qz2_76cfc4kwgr

(If you live in China, then your browsing will only depend on Internet's proxy into some website.)

gazonk said:

Of course it's the noise reduction!

Click to expand...

Let me tell you, it is not only a noise reduction! Your milk or salt are in a software!

Laurens said:

So all those armchair electronic engineers can say what they like about those pictures, but without some schematics and the layout gerbers then it'll mostly be BS.

Click to expand...

i rember reading that the grip on a K10/20 can cause interference/banding...
for this reason, i am worried about cheap radio triggers.
can they cause trouble ?

if i use them, i think i would not mount the emitter on the camera. i would mount it on the flash bracket. is this a reasonable (or even necessary) precaution ?

Peter, I believe you know more than me about electronics, internals of cameras and how they work. I know a little I have been into tech including electronics since I was 13. My first computer I bought was the VIC 20 then traded for the 64 from paper route money. I use to alter (or add) the crossover in my speakers in hopes of making (cheap) my speakers better. But its a hobby, my degree is in computer science (programming). I believe it was EMI/RF coming into the camera from the tripod, it connected with the cameras cold ground, the frame another poster was speaking of, forming an antenna, this radiated inside the camera to the NuCores, CCD or the wires (that you show) connecting the CCD and Nucores. It probably did not happen after the NuCores because the signal was digital then. So it almost proves you 100%. But we are not 100% sure. It could happen to any number of other cameras and I will try to get a Canon and test it at that site. What I am saying is my area is worst case, and it does not prove Pentax was neglect in its design. I slipped in through the backdoor, the tripod mount if you will.

In the Pic you can see the antenna to the right, its real close? But what you don't see is behind me is another antenna even closer! and to my left about 100m away there is a cell phone antenna. But I think I have proved cell phone antennas don't make noise in the K10ds output. And their are about 5 or so more antennas all within 5 miles surrounding me. I live in the middle of the Metro Detroit area. This is worst case. And you and me both know that building is reflecting signals making it even worse. And when I don't use the tripod there is 'no' problem. There is no problem using the tripod with plastic head. The K10 alone had enough shielding in this worst case to prevent noise. Its only by me introducing a 5' antenna on it bottom that it happened.

I had the Sony A300 it uses I believe the same CCD as the K10? It went to iso3200, but in jpeg it can not touch the K10 it smeared away detail. The K10s designer left it alone for you to PP. Remember the noise you see in the K10 is the floor, the base of S/N and Pentax did not mask it. With my new GX10 working the way it was supposed to work I see the VPN and noise is not that bad and the Pentax K10 does OK at iso1600l, and allowed us the end user the most from its CCD. The noise or VPN on my camera is not a bad thing, its just the cameras natural limit or noise floor. Others may have less noise but wont show the detail. I think others have proved this. Of course the less external or internal noise the better.

Peter because of the slight language barrier I want to make clear. I write this kindly, I am not angry or trying to make you look wrong or bad. Just writing my IMO, I know you know your stuff.

--
jamesm007,
http://s195.photobucket.com/albums/z77/jamesm700/

Tan68 said:

rember reading that the grip on a K10/20 can cause interference/banding...
for this reason

Click to expand...

Interesting I did not know this. The grip uses the tripod mount...
--
jamesm007,
http://s195.photobucket.com/albums/z77/jamesm700/

jamesm007 said:

So it almost proves you 100%. But we are not 100% sure.

Click to expand...

I did not any proves 100%. I had only said that:

A metallic shielding way it is not the only panacea or it has abused, but without any metallic shield’s way would be in no way a better way.

I request PENTAX to improve their DSLR, that is included its internal electronic crafts prettifies.

you had seen them each picture, k10d PCB is full of messy cables and plaster, and it looks like a student's experiment and not a product.

PeterZheng said:

gazonk said:

Of course it's the noise reduction!

Click to expand...

Let me tell you, it is not only a noise reduction! Your milk or salt
are in a software!

Click to expand...

I'm sorry, but you cannot prove your point with those stamp shots, since they are not directly comparable as long as they've done with and without NR. If you want to prove anything about the noise of the camera + sensor system, you need to compare raw format files. And dxomark aims at doing just this, and according to them, k10d is less noisy than all the nikons with the same sensor.

--
Espen

gazonk said:

dxomark aims at doing just this, and according to them, k10d is less
noisy than all the nikons with the same sensor.

Click to expand...

you had proved a software NR.

1. DxO did not stop NR.

2. If Nikon performance was insufficient, and then attempts with it to prove Pentax was a lofty master and K10D should be not any metallic shielding onto PCB?

Did you think that, if nikon did not have any metallic shielding, then its EMC as well as the image quality would be better?

I think firmly that if Pentax is earnest to process their DSLR EMI/EMC, then their opportunity is better, but without any metallic shield’s way would be in no way a better way.

A simple fact, if the Pentax like to stubbornly keep its K10D's electronic craft continues, then more person or me could give up the Pentax camera.

I completely forgot about this. A news stations weather radar (a big ball on top of another tall building) is about 200m away from me (at the site I took the pics). Could radar signals reach the ground, the ball is in direct eye site, but I know its looking up, but???? That's crazy how much transmissions of every kind is in that area. The radar might even be bouncing off the building toward me. Because that building (in the pic) is taller than the building the radar is on.

But I grant your IMO, the K10 should have a full solid frame to prevent flexing, I noticed this myself, and other improvements. That I would not know myself, but you have proven this. I would be interested in the internals of Sony and Oly? Also I believe a lot of what you see versus Canon and Nikon is because they don't have internal shake reduction, the Pentax had to be built in an entirely different way because the CCD moves a lot, without SR they may have been able to do things different. The Pentax engineers had to face some obstacles that the non-sealed non-shake reduction Canon and Nikons don't. This was a state of the art camera, offering a lot. Lets see if latter generations don't improve.

However I must say the Pentax K10 K20 have been dependable very good performing cameras. I don't read of too many problems? I don't make things up most of the time, and tell you the failure rate on this forum is low. So flaws or not they get the job done. Now I am not saying its good enough. But if Pentax fixed those things you pointed out they would be near bullet proof IMO.

You are demanding improvements but I don't understand why you believe Pentax fails more than Nikon or Canon?

You should see the inside of my home built computer, now that I have gone SATA its a bit better - but still it looks like a rats nest, but its 100% reliable. Compare that to my neighbors DELL wonderfully constructed but no more reliable if as reliable as mine. Its pretty, but... There's a line between just being neat and necessary for reliability.
--
jamesm007,
http://s195.photobucket.com/albums/z77/jamesm700/