Has anyone verified Qwntm's test for K-3?

Started 10 months ago | Discussions
Roland Karlsson
Forum ProPosts: 20,910Gear list
Like?
Has anyone verified Qwntm's test for K-3?
10 months ago

Hi there,

Lots of threads on the issues of K-3. I have to admit I am now lost. I cannot remember any post showing a verification of Qwantum's assertion that the problem is HIGH NR. I.e. a test where he did turn HIGH NR on and off and compared the result.

Some says they cannot repeat it - but shows no images. At least what I have seen.

But has anyone succeeded in repeating it?

 Roland Karlsson's gear list:Roland Karlsson's gear list
Sigma DP3 Merrill Sigma DP2 Quattro Sony Cyber-shot DSC-RX100 III Pentax K-3 Pentax smc DA 15mm F4 ED AL Limited +8 more
Pentax K-3
If you believe there are incorrect tags, please send us this post using our feedback form.
KL Matt
Veteran MemberPosts: 5,173
Like?
Re: Has anyone verified Qwntm's test for K-3?
In reply to Roland Karlsson, 10 months ago

Roland Karlsson wrote:

Hi there,

Lots of threads on the issues of K-3. I have to admit I am now lost. I cannot remember any post showing a verification of Qwantum's assertion that the problem is HIGH NR. I.e. a test where he did turn HIGH NR on and off and compared the result.

Some says they cannot repeat it - but shows no images. At least what I have seen.

But has anyone succeeded in repeating it?

Should be easy enough... all you have to do is focus poorly, shoot, turn off the high ISO noise reduction, refocus accurately, and Bob's your uncle!

Reply   Reply with quote   Complain
Tim A2
Senior MemberPosts: 1,099
Like?
Re: Has anyone verified Qwntm's test for K-3?
In reply to Roland Karlsson, 10 months ago

Roland Karlsson wrote:

Hi there,

Lots of threads on the issues of K-3. I have to admit I am now lost. I cannot remember any post showing a verification of Qwantum's assertion that the problem is HIGH NR. I.e. a test where he did turn HIGH NR on and off and compared the result.

If I understand correctly, Qwntm's test uses auto NR and not high NR. He says that auto turns on high NR in only a certain kind of light, so when somebody does not verify his result, he says it is because they did not have the right kind of light.

Some says they cannot repeat it - but shows no images. At least what I have seen.

But has anyone succeeded in repeating it?

Reply   Reply with quote   Complain
Roland Karlsson
Forum ProPosts: 20,910Gear list
Like?
Re: I am serious
In reply to KL Matt, 10 months ago

KL Matt wrote:
Should be easy enough... all you have to do is focus poorly, shoot, turn off the high ISO noise reduction, refocus accurately, and Bob's your uncle!

Please, I asked this seriously. I really want to know. And I do not plan to buy any K-3 right now just to do the test myself.

 Roland Karlsson's gear list:Roland Karlsson's gear list
Sigma DP3 Merrill Sigma DP2 Quattro Sony Cyber-shot DSC-RX100 III Pentax K-3 Pentax smc DA 15mm F4 ED AL Limited +8 more
Reply   Reply with quote   Complain
Roland Karlsson
Forum ProPosts: 20,910Gear list
Like?
Re: Has anyone verified Qwntm's test for K-3?
In reply to Tim A2, 10 months ago

Tim A2 wrote:
If I understand correctly, Qwntm's test uses auto NR and not high NR. He says that auto turns on high NR in only a certain kind of light, so when somebody does not verify his result, he says it is because they did not have the right kind of light.

No - I do not think so. I think he used NR HIGH and NR OFF.

 Roland Karlsson's gear list:Roland Karlsson's gear list
Sigma DP3 Merrill Sigma DP2 Quattro Sony Cyber-shot DSC-RX100 III Pentax K-3 Pentax smc DA 15mm F4 ED AL Limited +8 more
Reply   Reply with quote   Complain
zakaria
Senior MemberPosts: 1,618Gear list
Like?
Re: Has anyone verified Qwntm's test for K-3?
In reply to Roland Karlsson, 10 months ago

I am also interested in but you can find a prove that the k3 is amazing tool if you go the other forum there are many amazing photos taken with this camera.

-- hide signature --

pentaxian .

 zakaria's gear list:zakaria's gear list
Pentax K-5 IIs Pentax K-3
Reply   Reply with quote   Complain
KL Matt
Veteran MemberPosts: 5,173
Like?
Re: I am serious
In reply to Roland Karlsson, 10 months ago

Roland Karlsson wrote:

KL Matt wrote:
Should be easy enough... all you have to do is focus poorly, shoot, turn off the high ISO noise reduction, refocus accurately, and Bob's your uncle!

Please, I asked this seriously. I really want to know. And I do not plan to buy any K-3 right now just to do the test myself.

If you have a camera shop that carries Pentax near you, you could do the test yourself, shouldn't be too hard.

But may I ask why you really want to know? There is something really fishy about the vehemence with which he's pursuing this. It's like a crusade. One wonders what the point is. Perhaps there is some motive. I wouldn't ascribe any significance to it at all, personally.Maybe, just maybe, he got a bad one. But I doubt it.

We should invite some kind people from the d800 etc. forum to come over and have a look at this mess and see what they think just for kicks. Bet they'd all have flashbacks. It's a very sharp sensor, people are seeing ghosts.

Matt

Reply   Reply with quote   Complain
Rosember
Senior MemberPosts: 1,575Gear list
Like?
Re: Has anyone verified Qwntm's test for K-3?
In reply to Roland Karlsson, 10 months ago

I tried to but could not reproduce his findings. He changed or added new setting to the description and new theories from time to time. In the end I gave up. I cannot confirm his findings but I think he might be right that there is an issue under certain settings (which I don't use). All his findings only concern jpegs, not Raws. My K-3 seems to work flawlessly.. I am very impressed by the K-3, especially  by the low light AF, the SR, the DR and colour accuracy at high ISO, and the newly designed build-in grip which helps to stabilize the cam. A very fine upgrade from my K20D.

 Rosember's gear list:Rosember's gear list
Pentax Q Pentax K-3 Pentax smc DA* 55mm F1.4 SDM Pentax smc DA 35mm F2.8 Macro Limited Pentax smc DA 18-135mm F3.5-5.6ED AL [IF] DC WR +3 more
Reply   Reply with quote   Complain
Tim A2
Senior MemberPosts: 1,099
Like?
Re: Has anyone verified Qwntm's test for K-3?
In reply to Roland Karlsson, 10 months ago

Roland Karlsson wrote:

Tim A2 wrote:
If I understand correctly, Qwntm's test uses auto NR and not high NR. He says that auto turns on high NR in only a certain kind of light, so when somebody does not verify his result, he says it is because they did not have the right kind of light.

No - I do not think so. I think he used NR HIGH and NR OFF.

"To reproduce what I've found, set your lens to 135mm, ISO 400-800 or higher, shoot something at 1/500th of a second with High ISO noise reduction on AUTO (default) and then with High ISO noise reduction set to off, and watch the blur magically go away at 100% peeping...." / Qwntm

"I have observed that in flat light conditions, the AUTO setting will employ VERY HIGH ISO noise reduction even at ISO's as low as 400, resulting in completely blurry images when viewed at 100%. Yet the same scene photographed in more contrasty light will receive little High ISO noise reduction applied by the AUTO setting. When High ISO Noise Reduction is set to “OFF” image detail is preserved and the camera produces results one would expect, very very sharp." / Qwntm

Your point that no one has reproduced his test is well taken.

There is a thread devoted to shots taken with NR set to high.

http://www.dpreview.com/forums/thread/3590631

Reply   Reply with quote   Complain
Roland Karlsson
Forum ProPosts: 20,910Gear list
Like?
Re: Has anyone verified Qwntm's test for K-3?
In reply to Tim A2, 10 months ago

Tim A2 wrote:

"To reproduce what I've found, set your lens to 135mm, ISO 400-800 or higher, shoot something at 1/500th of a second with High ISO noise reduction on AUTO (default) and then with High ISO noise reduction set to off, and watch the blur magically go away at 100% peeping...." / Qwntm

"I have observed that in flat light conditions, the AUTO setting will employ VERY HIGH ISO noise reduction even at ISO's as low as 400, resulting in completely blurry images when viewed at 100%. Yet the same scene photographed in more contrasty light will receive little High ISO noise reduction applied by the AUTO setting. When High ISO Noise Reduction is set to “OFF” image detail is preserved and the camera produces results one would expect, very very sharp." / Qwntm

Hmmmm ... seems I have been reading his posts to fast.

It says nothing about setting the NR to HIGH. This can be interpreted in two ways - either he means that AUTO may use more NR than HIGH, or that he has not bothered to test HIGH.

Your point that no one has reproduced his test is well taken.

There is a thread devoted to shots taken with NR set to high.

http://www.dpreview.com/forums/thread/3590631

Thanx - I shall read that.

 Roland Karlsson's gear list:Roland Karlsson's gear list
Sigma DP3 Merrill Sigma DP2 Quattro Sony Cyber-shot DSC-RX100 III Pentax K-3 Pentax smc DA 15mm F4 ED AL Limited +8 more
Reply   Reply with quote   Complain
Roland Karlsson
Forum ProPosts: 20,910Gear list
Like?
Re: Has anyone verified Qwntm's test for K-3?
In reply to Roland Karlsson, 10 months ago

Roland Karlsson wrote:

Tim A2 wrote:

"To reproduce what I've found, set your lens to 135mm, ISO 400-800 or higher, shoot something at 1/500th of a second with High ISO noise reduction on AUTO (default) and then with High ISO noise reduction set to off, and watch the blur magically go away at 100% peeping...." / Qwntm

"I have observed that in flat light conditions, the AUTO setting will employ VERY HIGH ISO noise reduction even at ISO's as low as 400, resulting in completely blurry images when viewed at 100%. Yet the same scene photographed in more contrasty light will receive little High ISO noise reduction applied by the AUTO setting. When High ISO Noise Reduction is set to “OFF” image detail is preserved and the camera produces results one would expect, very very sharp." / Qwntm

Hmmmm ... seems I have been reading his posts to fast.

It says nothing about setting the NR to HIGH. This can be interpreted in two ways - either he means that AUTO may use more NR than HIGH, or that he has not bothered to test HIGH.

Your point that no one has reproduced his test is well taken.

There is a thread devoted to shots taken with NR set to high.

http://www.dpreview.com/forums/thread/3590631

Thanx - I shall read that.

Yes! That is the thread I remembered. There he sets it to HIGH and OFF and compares. So - if that is correct - then he has a point.

So - what was your point?

 Roland Karlsson's gear list:Roland Karlsson's gear list
Sigma DP3 Merrill Sigma DP2 Quattro Sony Cyber-shot DSC-RX100 III Pentax K-3 Pentax smc DA 15mm F4 ED AL Limited +8 more
Reply   Reply with quote   Complain
Roland Karlsson
Forum ProPosts: 20,910Gear list
Like?
Re: I am serious
In reply to KL Matt, 10 months ago

KL Matt wrote: If you have a camera shop that carries Pentax near you, you could do the test yourself, shouldn't be too hard.

well ...

  1. I dont have any shop with a K-3 nearby
  2. Making tests is not all that easy

But may I ask why you really want to know?

  1. I am just curious.
  2. I sometimes use JPEG. I which case I seldom change default settings.

There is something really fishy about the vehemence with which he's pursuing this. It's like a crusade. One wonders what the point is. Perhaps there is some motive. I wouldn't ascribe any significance to it at all, personally.Maybe, just maybe, he got a bad one. But I doubt it.

He seems to be a person that gets deeply involved and emotional.

We should invite some kind people from the d800 etc. forum to come over and have a look at this mess and see what they think just for kicks. Bet they'd all have flashbacks. It's a very sharp sensor, people are seeing ghosts.

This is not about sharp sensors. Its about in camera software.

Now, of course, if I get that K-3 (which I may or may not this spring) I will of course shoot some tests comparing NR OFF and NR HIGH and LOW. Just to see what one suits me best.

 Roland Karlsson's gear list:Roland Karlsson's gear list
Sigma DP3 Merrill Sigma DP2 Quattro Sony Cyber-shot DSC-RX100 III Pentax K-3 Pentax smc DA 15mm F4 ED AL Limited +8 more
Reply   Reply with quote   Complain
awaldram
Forum ProPosts: 10,776Gear list
Like?
Re: I am serious
In reply to Roland Karlsson, 10 months ago

Well 4000 images in my observations

There many ways to get blurred Images the vast majority caused by lack of understanding how features work.

1 panning to un-smoothly will reset SR during shot

2 Expanding AF will fail to select AF point during snap shooting (read 1st actions)

3 Hold Af if mis-set will stop Auto focus achieving sharp images

4 High NR will soften image (well Der !! )

There are many others but all the above have caught me out and some (4 especially ) are just obvious and not worth the forums time being wasted, If you don't like Pentax's NR levels in auto then use custom and select what you consider appropriate.

There's no smoking gun just a well though out camera that is obviously to much camera for some.!

Shock horror I was able to give  my correctly configured K3 (in P) total strangers and they happily shot off sharp shot after sharp shot.

The K3 as delivered (default) will happily shoot sharp shots but seems some like to customize when they don't know what settings alter then complain, making out its  anybodies fault but their own !!

 awaldram's gear list:awaldram's gear list
Pentax Q Pentax K-3 Pentax smc DA* 55mm F1.4 SDM Pentax smc DA* 16-50mm F2.8 ED AL (IF) SDM Pentax smc DA* 50-135mm F2.8 ED (IF) SDM +11 more
Reply   Reply with quote   Complain
Bob Corson
Senior MemberPosts: 2,246Gear list
Like?
Re: Has anyone verified Qwntm's test for K-3?
In reply to Roland Karlsson, 10 months ago

Roland Karlsson wrote:

Tim A2 wrote:

"To reproduce what I've found, set your lens to 135mm, ISO 400-800 or higher, shoot something at 1/500th of a second with High ISO noise reduction on AUTO (default) and then with High ISO noise reduction set to off, and watch the blur magically go away at 100% peeping...." / Qwntm

"I have observed that in flat light conditions, the AUTO setting will employ VERY HIGH ISO noise reduction even at ISO's as low as 400, resulting in completely blurry images when viewed at 100%. Yet the same scene photographed in more contrasty light will receive little High ISO noise reduction applied by the AUTO setting. When High ISO Noise Reduction is set to “OFF” image detail is preserved and the camera produces results one would expect, very very sharp." / Qwntm

Hmmmm ... seems I have been reading his posts to fast.

It says nothing about setting the NR to HIGH. This can be interpreted in two ways - either he means that AUTO may use more NR than HIGH, or that he has not bothered to test HIGH.

Your point that no one has reproduced his test is well taken.

There is a thread devoted to shots taken with NR set to high.

http://www.dpreview.com/forums/thread/3590631

Thanx - I shall read that.

It is in that post where he talks of setting NR to High. That is part of the problem test shots were not all shot the same way.

-- hide signature --

Bob Corson
To truly see one must open their mind as well as their eyes

 Bob Corson's gear list:Bob Corson's gear list
Pentax K-30 Sigma 10-20mm F4-5.6 EX DC HSM Pentax smc DA* 60-250mm F4.0 ED (IF) SDM Tamron SP AF 17-50mm F/2.8 XR Di II LD Aspherical (IF) Tamron SP AF 90mm F/2.8 Di Macro +5 more
Reply   Reply with quote   Complain
Roland Karlsson
Forum ProPosts: 20,910Gear list
Like?
Re: I am serious
In reply to awaldram, 10 months ago

awaldram wrote:

Well 4000 images in my observations

There many ways to get blurred Images the vast majority caused by lack of understanding how features work.

1 panning to un-smoothly will reset SR during shot

2 Expanding AF will fail to select AF point during snap shooting (read 1st actions)

3 Hold Af if mis-set will stop Auto focus achieving sharp images

4 High NR will soften image (well Der !! )

There are many others but all the above have caught me out and some (4 especially ) are just obvious and not worth the forums time being wasted, If you don't like Pentax's NR levels in auto then use custom and select what you consider appropriate.

There's no smoking gun just a well though out camera that is obviously to much camera for some.!

Shock horror I was able to give a correctly my configured K3 (in P) total strangers and they happily shot of sharp shot after sharp shot.

In the old days it was much easier.

There was no SR, so you HAD to hold on steady yourself.

There was no AF, so you had to see to it that you focussed yourself.

There was no software, so the film you had in the camera was the limit.

 Roland Karlsson's gear list:Roland Karlsson's gear list
Sigma DP3 Merrill Sigma DP2 Quattro Sony Cyber-shot DSC-RX100 III Pentax K-3 Pentax smc DA 15mm F4 ED AL Limited +8 more
Reply   Reply with quote   Complain
awaldram
Forum ProPosts: 10,776Gear list
Like?
Re: I am serious
In reply to Roland Karlsson, 10 months ago

Roland Karlsson wrote:

awaldram wrote:

Well 4000 images in my observations

There many ways to get blurred Images the vast majority caused by lack of understanding how features work.

1 panning to un-smoothly will reset SR during shot

2 Expanding AF will fail to select AF point during snap shooting (read 1st actions)

3 Hold Af if mis-set will stop Auto focus achieving sharp images

4 High NR will soften image (well Der !! )

There are many others but all the above have caught me out and some (4 especially ) are just obvious and not worth the forums time being wasted, If you don't like Pentax's NR levels in auto then use custom and select what you consider appropriate.

There's no smoking gun just a well though out camera that is obviously to much camera for some.!

Shock horror I was able to give a correctly my configured K3 (in P) total strangers and they happily shot of sharp shot after sharp shot.

In the old days it was much easier.

There was no SR, so you HAD to hold on steady yourself.

There was no AF, so you had to see to it that you focussed yourself.

There was no software, so the film you had in the camera was the limit.

You should see the scans from my eos100 I was shocked to find only 1-2 in 10 where critically sharp , seems our expectations are tougher as well as the ability to see error.

With the K3 I can see focus error nose to eye from 30mm F2.8 shot 10-15 ft away its insane

 awaldram's gear list:awaldram's gear list
Pentax Q Pentax K-3 Pentax smc DA* 55mm F1.4 SDM Pentax smc DA* 16-50mm F2.8 ED AL (IF) SDM Pentax smc DA* 50-135mm F2.8 ED (IF) SDM +11 more
Reply   Reply with quote   Complain
Tim A2
Senior MemberPosts: 1,099
Like?
Re: Has anyone verified Qwntm's test for K-3?
In reply to Roland Karlsson, 10 months ago

Roland Karlsson wrote:

Roland Karlsson wrote:

Tim A2 wrote:

"To reproduce what I've found, set your lens to 135mm, ISO 400-800 or higher, shoot something at 1/500th of a second with High ISO noise reduction on AUTO (default) and then with High ISO noise reduction set to off, and watch the blur magically go away at 100% peeping...." / Qwntm

"I have observed that in flat light conditions, the AUTO setting will employ VERY HIGH ISO noise reduction even at ISO's as low as 400, resulting in completely blurry images when viewed at 100%. Yet the same scene photographed in more contrasty light will receive little High ISO noise reduction applied by the AUTO setting. When High ISO Noise Reduction is set to “OFF” image detail is preserved and the camera produces results one would expect, very very sharp." / Qwntm

Hmmmm ... seems I have been reading his posts to fast.

It says nothing about setting the NR to HIGH. This can be interpreted in two ways - either he means that AUTO may use more NR than HIGH, or that he has not bothered to test HIGH.

Your point that no one has reproduced his test is well taken.

There is a thread devoted to shots taken with NR set to high.

http://www.dpreview.com/forums/thread/3590631

Thanx - I shall read that.

Yes! That is the thread I remembered. There he sets it to HIGH and OFF and compares. So - if that is correct - then he has a point.

But the test (more of a demonstration) in that thread, had a different purpose. It was simply to show how destructive high NR and hence auto mode is when High NR is enabled by the auto mode. here is his conclusion.

"THUS, logically we can conclude that if HIGH is a possibility for the AUTO mode to choose, and we have our camera set to AUTO, then some of our shots could experience random obliteration of the detail." /  Qwntm

So - what was your point?

That is clear now.

Reply   Reply with quote   Complain
Roland Karlsson
Forum ProPosts: 20,910Gear list
Like?
Re: Has anyone verified Qwntm's test for K-3?
In reply to Tim A2, 10 months ago

Tim A2 wrote:

That is clear now.

No - sorry - I still don't get what your point was.

In my thread I asked if anyone had verified his finding the high NR had that kind of destructive effect on details that he claimed.

Now, have you verified that or have you a test that shows him wrong?

 Roland Karlsson's gear list:Roland Karlsson's gear list
Sigma DP3 Merrill Sigma DP2 Quattro Sony Cyber-shot DSC-RX100 III Pentax K-3 Pentax smc DA 15mm F4 ED AL Limited +8 more
Reply   Reply with quote   Complain
Tim A2
Senior MemberPosts: 1,099
Like?
Re: Has anyone verified Qwntm's test for K-3?
In reply to Roland Karlsson, 10 months ago

Roland Karlsson wrote:

Tim A2 wrote:

That is clear now.

No - sorry - I still don't get what your point was.

In my thread I asked if anyone had verified his finding the high NR had that kind of destructive effect on details that he claimed.

Now, have you verified that or have you a test that shows him wrong?

I see the misunderstanding now. I had no idea you were talking about his demonstration of how destructive high NR is.  My impression is that no one thought it needed to be verified. High NR will certainly degrade the sharpness.  The degree of destruction would depend on the scene and lighting. To reproduce his demonstration one would first have to reproduce the scene he shot including the lighting and that  is not practical.

Reply   Reply with quote   Complain
il_alexk
Senior MemberPosts: 1,977Gear list
Like?
Yes, the problem was confirmed in this review
In reply to Roland Karlsson, 10 months ago

Roland Karlsson wrote:

But has anyone succeeded in repeating it?

Of course, see this review by Neocamera:

"The Pentax K-3 matches the K-5 IIs in terms of image-quality at low ISO which is quite impressive considering its 50% extra megapixels. At ISO 400, the K-3 begins to show noise-reduction softness but its output remains equally usable thanks to the added resolution. At ISO 800+ though, the K-3 starts falling behind, giving a full-stop lead to the K-5 IIs by ISO 6400."

http://www.neocamera.com/camera/pentax/k3/samples/indoor

-- hide signature --
 il_alexk's gear list:il_alexk's gear list
Pentax K-5 Olympus PEN E-PM1 Olympus OM-D E-M10 Panasonic Lumix G 20mm F1.7 ASPH Pentax smc FA 50mm F1.4 +7 more
Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads