Super curious about the KP's autofocus

ogl · 03-05-2017, 09:50 PM

Originally posted by MarkJerling

If you're not talking about hardware improvements, or software improvements, then what are you talking about?

Is it a joke? I don't see hardware and software in camera, but I can see how AF works in reality.

I talk about work of AF. How AF works - accuracy, speed, responsivity, AF-C - how effective.

It's strange to expect from KP any progress - AF module is 2013 year made. Magic software algorithms?
Good marketing slogan for users - "We improved software algorithms"
...Ohhh...Yes. I feel it in my camera. It's really works. I'm happy at last.

MarkJerling · 03-06-2017, 02:37 AM

Originally posted by ogl

Is it a joke? I don't see hardware and software in camera, but I can see how AF works in reality.

I talk about work of AF. How AF works - accuracy, speed, responsivity, AF-C - how effective.

It's strange to expect from KP any progress - AF module is 2013 year made. Magic software algorithms?
Good marketing slogan for users - "We improved software algorithms"
...Ohhh...Yes. I feel it in my camera. It's really works. I'm happy at last.

No, it's not a joke. Technological improvements in terms of hardware, or software, or both, will result in improvements in AF. How AF works in reality depends on those parameters. You are dealing with a machine that can be improved in terms of hardware or software. While the outcomes is what you refer to as "reality" that "reality" is subjective.

The test you reference is hardly conclusive. I'd like to see more scientific tests before we shoot down the KP's abilities.

**BigMackCam** · 03-06-2017, 03:06 AM - **2 Likes**

Originally posted by ogl

It's strange to expect from KP any progress - AF module is 2013 year made. Magic software algorithms?
Good marketing slogan for users - "We improved software algorithms"

With all due respect, that makes no sense to me

Computer operating systems and applications software are improved all the time without requiring changes to the underlying hardware. My PC, which I bought - coincidentally - in 2013, was running Windows 8 when I bought it. Now it's running Windows 10. In the last year or two, updates to my applications software have improved performance and features without requiring any change to the hardware.

Our cameras are merely computers with sensors, electro-mechanical mechanisms and user-interface hardware. Arguably the most important part - and you'll know this if you've ever done any software development work - is the software / firmware, without which the hardware is useless. The software can be written in an almost-limitless number of ways to do the same basic functions, but some approaches will yield better results than others.

To put it another way - and in complete contrast to what you've stated - it would be strange NOT to expect progress from the KP... New software algorithms don't need to be magical - they just need to be better than the algorithms they're replacing

Now, it's certainly possible that brand new AF sensor design combined with new algorithms could have provided even greater improvements in AF; but you don't necessarily need new hardware to gain significant improvements.

ogl · 03-06-2017, 03:49 AM

Originally posted by MarkJerling

No, it's not a joke. Technological improvements in terms of hardware, or software, or both, will result in improvements in AF. How AF works in reality depends on those parameters. You are dealing with a machine that can be improved in terms of hardware or software. While the outcomes is what you refer to as "reality" that "reality" is subjective.

The improvements in AF could not be seen by users, because of small difference between the cameras of different generations.
But if you had problems with old camera, your new camera without old illnesses seems to be much better for you. It's psychology.

**BigMackCam** · 03-06-2017, 04:19 AM

Originally posted by ogl

The improvements in AF could not be seen by users, because of small difference between the cameras of different generations.
But if you had problems with old camera, your new camera without old illnesses seems to be much better for you. It's psychology.

I think we need to wait and see some documented A/B tests between the K-3 and KP, where the tests are consistent and repeatable. Those tests need to include more than one lens, and cover a range of AF.S and AF.C scenarios. Without this, every opinion we hear or read will be highly subjective and - whilst interesting to some extent - of limited value, since it's not based on hard evidence. We certainly can't conclude that the KP does or doesn't offer significant AF improvement without that kind of testing. If your comrades' testing is of that nature, it would be great to see the documented results, and very helpful for potential buyers.

howieb101 · 03-06-2017, 04:34 AM

Originally posted by BigMackCam

I think we need to wait and see some documented A/B tests between the K-3 and KP, where the tests are consistent and repeatable. Those tests need to include more than one lens, and cover a range of AF.S and AF.C scenarios. Without this, every opinion we hear or read will be highly subjective and - whilst interesting to some extent - of limited value, since it's not based on hard evidence. We certainly can't conclude that the KP does or doesn't offer significant AF improvement without that kind of testing. If your comrades' testing is of that nature, it would be great to see the documented results, and very helpful for potential buyers.

Sorry, its not just psychology. I have a K5 and a K3. The AF in the K3 is far superior to the K5.

I can't vouch for the KP but we will know soon enough once enough users have bought KP's.

Howie Be

lennyl · 03-06-2017, 04:40 AM

the SAFOX AF modul is the same crap from time K-3, K-1 to KP.
why? is unreliable. all the time proper is focus. and suddenly have a photo of the front focus.

1. GOOD DAY

my test AF-C, Pentax K-1 fw1.30 + slow Pentax SMC DA*200/2,8 SDM.
Settings:
-TAv, AutoISO,
-AF-C, SEL 2
-1st Frame Action in AF.C - Focus-priority, Action in AF.C Cont. - FPS-priority
- Continuous Shooting (H)
- Hold AF Status - OFF, Shake Reduction - OFF, Menu C3 Meter Operating Time - Continue

13 OK, 2 NOK

19 OK, 3 NOK

13 OK, 2 NOK

14 OK, 4 NOK

16 OK, 4 NOK

4 OK

19 OK, 3 NOK

13 OK, 2 NOK

2. BAD DAY

AF-S, central point, K-1 + FA43/1,9 Limited.

ALL 8 pics got 20-30cm front focus, wtf?

next photos from D-FA HD70-200/2,8 was absolutely awesome.

why?

because SAFOX.

MarkJerling · 03-06-2017, 04:41 AM

Originally posted by ogl

The improvements in AF could not be seen by users, because of small difference between the cameras of different generations.
But if you had problems with old camera, your new camera without old illnesses seems to be much better for you. It's psychology.

No, we're talking about data. Test results and expectations of improvement based upon those (as yet unpublished) test results. We're not talking psychology. The camera will either be proved to be better at AF or not - it's not psychology.

**BigMackCam** · 03-06-2017, 04:57 AM

Originally posted by lennyl

the SAFOX AF modul is the same crap from time K-3, K-1 to KP.
why? is unreliable. all the time proper is focus. and suddenly have a photo of the front focus.

Sigh...

You obviously have much worse results than I do. I've been shooting with Pentax since the K-7 came out. I've had good AF results with the K-5, and very good results with the K-3, both with AF.S and AF.C in a variety of situations. Could it be better? You bet. Things can always be improved upon. All DSLRs can suffer from front or back-focusing with a particular lens. Mirrorless cameras can suffer from it too.

I'll believe the KP is the same, better or worse when I see some properly controlled A/B tests between it and a K-3

ogl · 03-06-2017, 10:49 AM

Originally posted by howieb101

Sorry, its not just psychology. I have a K5 and a K3. The AF in the K3 is far superior to the K5.

Blessed are those who believe

Far superior is AF of Nikon D500. K-3 is just a bit better. Taste the difference.

ogl · 03-06-2017, 11:02 AM

Originally posted by BigMackCam

Sigh...
I've had good AF results with the K-5, and very good results with the K-3, :

K-5's AF was the worst for me. Just a bit better than *ist DS.

If the different users have different opinions about the same camera, it means SAFOX system is not stable.

Let's say the truth - even last SAFOX 12 is one of the weakest AF module between the cameras of other brands.
Even Nikon D5600 has real tracking AF and has 39 AF points. I has the feeling that cross-type AF sensors of Pentax hardly help to camera.

The old Multi-Cam of Nikon with 51 AF points with only 15 cross-type sensors is far better than K-1's AF.
Pentax cameras have no good level tracking AF system. It's true.

If we can make photos in AF-C mode doesn't mean that this system is really good. We just the users with high level of self-adaptive and it's all.
We have already got accustomed to such work of SAFOX and nothing else.

**BigMackCam** · 03-06-2017, 11:52 AM - **3 Likes**

Originally posted by ogl

If the different users have different opinions about the same camera, it means SAFOX system is not stable.

I think it's more the case that different users have different requirements, and different levels of ability in using what they're given. We have plenty of members on these forums getting great results with recent Pentax models in a wide variety of situations - birds in flight, motorsport, ball sports, low light, etc. Their success is a result of learning how the AF system works, and developing their techniques accordingly.

I think Pentax AF - especially AF.C - needs to improve, and I prefer the implementation on my Sony / Sony-based bodies - but I'm more-or-less happy with the K-3's AF and hopeful that the KP offers another small step in the right direction.

reh321 · 03-06-2017, 12:15 PM - **2 Likes**

Originally posted by ogl

K-5's AF was the worst for me. Just a bit better than *ist DS.

If the different users have different opinions about the same camera, it means SAFOX system is not stable.

Let's say the truth - even last SAFOX 12 is one of the weakest AF module between the cameras of other brands.
Even Nikon D5600 has real tracking AF and has 39 AF points. I has the feeling that cross-type AF sensors of Pentax hardly help to camera.

The old Multi-Cam of Nikon with 51 AF points with only 15 cross-type sensors is far better than K-1's AF.
Pentax cameras have no good level tracking AF system. It's true.

If we can make photos in AF-C mode doesn't mean that this system is really good. We just the users with high level of self-adaptive and it's all.
We have already got accustomed to such work of SAFOX and nothing else.

So, which of these do you actually use??

MarkJerling · 03-06-2017, 01:30 PM

Originally posted by ogl

K-5's AF was the worst for me. Just a bit better than *ist DS.

If the different users have different opinions about the same camera, it means SAFOX system is not stable.

Let's say the truth - even last SAFOX 12 is one of the weakest AF module between the cameras of other brands.
Even Nikon D5600 has real tracking AF and has 39 AF points. I has the feeling that cross-type AF sensors of Pentax hardly help to camera.

The old Multi-Cam of Nikon with 51 AF points with only 15 cross-type sensors is far better than K-1's AF.
Pentax cameras have no good level tracking AF system. It's true.

If we can make photos in AF-C mode doesn't mean that this system is really good. We just the users with high level of self-adaptive and it's all.
We have already got accustomed to such work of SAFOX and nothing else.

I can't compare with other brands because I don't shoot other brands. I presume you do?

All I can say is that I have no problem shooting fast moving sports such as football or cycling with my K-5. Clearly my K-5 has an AF module that works!

ogl · 03-06-2017, 09:13 PM

Originally posted by MarkJerling

I can't compare with other brands because I don't shoot other brands. I presume you do?

All I can say is that I have no problem shooting fast moving sports such as football or cycling with my K-5. Clearly my K-5 has an AF module that works!

to simply work and to work very effective are not the same.

03-06-2017, 03:06 AM - 2 Likes	#93
BigMackCam Digitiser of Film Join Date: Mar 2010 Location: North East of England Posts: 20,657	Originally posted by ogl It's strange to expect from KP any progress - AF module is 2013 year made. Magic software algorithms? Good marketing slogan for users - "We improved software algorithms" With all due respect, that makes no sense to me Computer operating systems and applications software are improved all the time without requiring changes to the underlying hardware. My PC, which I bought - coincidentally - in 2013, was running Windows 8 when I bought it. Now it's running Windows 10. In the last year or two, updates to my applications software have improved performance and features without requiring any change to the hardware. Our cameras are merely computers with sensors, electro-mechanical mechanisms and user-interface hardware. Arguably the most important part - and you'll know this if you've ever done any software development work - is the software / firmware, without which the hardware is useless. The software can be written in an almost-limitless number of ways to do the same basic functions, but some approaches will yield better results than others. To put it another way - and in complete contrast to what you've stated - it would be strange NOT to expect progress from the KP... New software algorithms don't need to be magical - they just need to be better than the algorithms they're replacing Now, it's certainly possible that brand new AF sensor design combined with new algorithms could have provided even greater improvements in AF; but you don't necessarily need new hardware to gain significant improvements. Last edited by BigMackCam; 03-06-2017 at 03:25 AM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Curious about the 28-45, I rented one...	iCrop	Pentax Medium Format	32	02-24-2015 09:39 AM
What's so Super About the Super Program	dubiousone	Film SLRs and Compact Film Cameras	28	04-04-2014 02:59 PM
Kinda curious about Older Manual lenses and the new K-5.	Abstract	Pentax SLR Lens Discussion	8	11-07-2010 10:43 AM
Curious about the white K-x kit - have you seen it?	switters	Pentax DSLR Discussion	7	11-27-2009 02:24 PM
curious about the rave reviews of the FA 50mm f1.4	roverlr3	Pentax SLR Lens Discussion	27	09-29-2008 12:02 AM

03-05-2017, 09:50 PM	#91
ogl Banned Join Date: Dec 2007 Location: Sankt Peterburg Photos: Gallery \| Albums Posts: 8,382	Originally posted by MarkJerling If you're not talking about hardware improvements, or software improvements, then what are you talking about? Is it a joke? I don't see hardware and software in camera, but I can see how AF works in reality. I talk about work of AF. How AF works - accuracy, speed, responsivity, AF-C - how effective. It's strange to expect from KP any progress - AF module is 2013 year made. Magic software algorithms? Good marketing slogan for users - "We improved software algorithms" ...Ohhh...Yes. I feel it in my camera. It's really works. I'm happy at last.

03-06-2017, 02:37 AM	#92
MarkJerling Moderator Site Supporter Join Date: May 2012 Location: Wairarapa, New Zealand Photos: Gallery \| Albums Posts: 20,403	Originally posted by ogl Is it a joke? I don't see hardware and software in camera, but I can see how AF works in reality. I talk about work of AF. How AF works - accuracy, speed, responsivity, AF-C - how effective. It's strange to expect from KP any progress - AF module is 2013 year made. Magic software algorithms? Good marketing slogan for users - "We improved software algorithms" ...Ohhh...Yes. I feel it in my camera. It's really works. I'm happy at last. No, it's not a joke. Technological improvements in terms of hardware, or software, or both, will result in improvements in AF. How AF works in reality depends on those parameters. You are dealing with a machine that can be improved in terms of hardware or software. While the outcomes is what you refer to as "reality" that "reality" is subjective. The test you reference is hardly conclusive. I'd like to see more scientific tests before we shoot down the KP's abilities.

03-06-2017, 03:49 AM	#94
ogl Banned Join Date: Dec 2007 Location: Sankt Peterburg Photos: Gallery \| Albums Posts: 8,382	Originally posted by MarkJerling No, it's not a joke. Technological improvements in terms of hardware, or software, or both, will result in improvements in AF. How AF works in reality depends on those parameters. You are dealing with a machine that can be improved in terms of hardware or software. While the outcomes is what you refer to as "reality" that "reality" is subjective. The improvements in AF could not be seen by users, because of small difference between the cameras of different generations. But if you had problems with old camera, your new camera without old illnesses seems to be much better for you. It's psychology.

03-06-2017, 04:19 AM	#95
BigMackCam Digitiser of Film Join Date: Mar 2010 Location: North East of England Posts: 20,657	Originally posted by ogl The improvements in AF could not be seen by users, because of small difference between the cameras of different generations. But if you had problems with old camera, your new camera without old illnesses seems to be much better for you. It's psychology. I think we need to wait and see some documented A/B tests between the K-3 and KP, where the tests are consistent and repeatable. Those tests need to include more than one lens, and cover a range of AF.S and AF.C scenarios. Without this, every opinion we hear or read will be highly subjective and - whilst interesting to some extent - of limited value, since it's not based on hard evidence. We certainly can't conclude that the KP does or doesn't offer significant AF improvement without that kind of testing. If your comrades' testing is of that nature, it would be great to see the documented results, and very helpful for potential buyers.

03-06-2017, 04:34 AM	#96
howieb101 Veteran Member Join Date: Nov 2008 Location: Melbourne, Australia Photos: Gallery \| Albums Posts: 594	Originally posted by BigMackCam I think we need to wait and see some documented A/B tests between the K-3 and KP, where the tests are consistent and repeatable. Those tests need to include more than one lens, and cover a range of AF.S and AF.C scenarios. Without this, every opinion we hear or read will be highly subjective and - whilst interesting to some extent - of limited value, since it's not based on hard evidence. We certainly can't conclude that the KP does or doesn't offer significant AF improvement without that kind of testing. If your comrades' testing is of that nature, it would be great to see the documented results, and very helpful for potential buyers. Sorry, its not just psychology. I have a K5 and a K3. The AF in the K3 is far superior to the K5. I can't vouch for the KP but we will know soon enough once enough users have bought KP's. Howie Be

03-06-2017, 04:40 AM	#97
lennyl Forum Member Join Date: Mar 2013 Location: Slovakia Posts: 63	the SAFOX AF modul is the same crap from time K-3, K-1 to KP. why? is unreliable. all the time proper is focus. and suddenly have a photo of the front focus. 1. GOOD DAY my test AF-C, Pentax K-1 fw1.30 + slow Pentax SMC DA200/2,8 SDM. Settings: -TAv, AutoISO, -AF-C, SEL 2 -1st Frame Action in AF.C - Focus-priority, Action in AF.C Cont. - FPS-priority - Continuous Shooting (H) - Hold AF Status - OFF, Shake Reduction - OFF, Menu C3 Meter Operating Time - Continue 13 OK, 2 NOK 19 OK, 3 NOK 13 OK, 2 NOK 14 OK, 4 NOK 16 OK, 4 NOK 4 OK 19 OK, 3 NOK 13 OK, 2 NOK 2. BAD DAY* AF-S, central point, K-1 + FA43/1,9 Limited. ALL 8 pics got 20-30cm front focus, wtf? next photos from D-FA HD70-200/2,8 was absolutely awesome. why? because SAFOX.

03-06-2017, 04:41 AM	#98
MarkJerling Moderator Site Supporter Join Date: May 2012 Location: Wairarapa, New Zealand Photos: Gallery \| Albums Posts: 20,403	Originally posted by ogl The improvements in AF could not be seen by users, because of small difference between the cameras of different generations. But if you had problems with old camera, your new camera without old illnesses seems to be much better for you. It's psychology. No, we're talking about data. Test results and expectations of improvement based upon those (as yet unpublished) test results. We're not talking psychology. The camera will either be proved to be better at AF or not - it's not psychology.

03-06-2017, 04:57 AM	#99
BigMackCam Digitiser of Film Join Date: Mar 2010 Location: North East of England Posts: 20,657	Originally posted by lennyl the SAFOX AF modul is the same crap from time K-3, K-1 to KP. why? is unreliable. all the time proper is focus. and suddenly have a photo of the front focus. Sigh... You obviously have much worse results than I do. I've been shooting with Pentax since the K-7 came out. I've had good AF results with the K-5, and very good results with the K-3, both with AF.S and AF.C in a variety of situations. Could it be better? You bet. Things can always be improved upon. All DSLRs can suffer from front or back-focusing with a particular lens. Mirrorless cameras can suffer from it too. I'll believe the KP is the same, better or worse when I see some properly controlled A/B tests between it and a K-3

03-06-2017, 10:49 AM	#100
ogl Banned Join Date: Dec 2007 Location: Sankt Peterburg Photos: Gallery \| Albums Posts: 8,382	Originally posted by howieb101 Sorry, its not just psychology. I have a K5 and a K3. The AF in the K3 is far superior to the K5. Blessed are those who believe Far superior is AF of Nikon D500. K-3 is just a bit better. Taste the difference.

03-06-2017, 11:02 AM	#101
ogl Banned Join Date: Dec 2007 Location: Sankt Peterburg Photos: Gallery \| Albums Posts: 8,382	Originally posted by BigMackCam Sigh... I've had good AF results with the K-5, and very good results with the K-3, : K-5's AF was the worst for me. Just a bit better than *ist DS. If the different users have different opinions about the same camera, it means SAFOX system is not stable. Let's say the truth - even last SAFOX 12 is one of the weakest AF module between the cameras of other brands. Even Nikon D5600 has real tracking AF and has 39 AF points. I has the feeling that cross-type AF sensors of Pentax hardly help to camera. The old Multi-Cam of Nikon with 51 AF points with only 15 cross-type sensors is far better than K-1's AF. Pentax cameras have no good level tracking AF system. It's true. If we can make photos in AF-C mode doesn't mean that this system is really good. We just the users with high level of self-adaptive and it's all. We have already got accustomed to such work of SAFOX and nothing else.

Thread Tools	Search this Thread
	Search this Thread: Advanced Search

03-06-2017, 11:52 AM - 3 Likes	#102
BigMackCam Digitiser of Film Join Date: Mar 2010 Location: North East of England Posts: 20,657	Originally posted by ogl If the different users have different opinions about the same camera, it means SAFOX system is not stable. I think it's more the case that different users have different requirements, and different levels of ability in using what they're given. We have plenty of members on these forums getting great results with recent Pentax models in a wide variety of situations - birds in flight, motorsport, ball sports, low light, etc. Their success is a result of learning how the AF system works, and developing their techniques accordingly. I think Pentax AF - especially AF.C - needs to improve, and I prefer the implementation on my Sony / Sony-based bodies - but I'm more-or-less happy with the K-3's AF and hopeful that the KP offers another small step in the right direction.

03-06-2017, 12:15 PM - 2 Likes	#103
reh321 Pentaxian Join Date: Nov 2014 Location: South Bend, IN, USA Photos: Gallery \| Albums Posts: 23,177	Originally posted by ogl K-5's AF was the worst for me. Just a bit better than *ist DS. If the different users have different opinions about the same camera, it means SAFOX system is not stable. Let's say the truth - even last SAFOX 12 is one of the weakest AF module between the cameras of other brands. Even Nikon D5600 has real tracking AF and has 39 AF points. I has the feeling that cross-type AF sensors of Pentax hardly help to camera. The old Multi-Cam of Nikon with 51 AF points with only 15 cross-type sensors is far better than K-1's AF. Pentax cameras have no good level tracking AF system. It's true. If we can make photos in AF-C mode doesn't mean that this system is really good. We just the users with high level of self-adaptive and it's all. We have already got accustomed to such work of SAFOX and nothing else. So, which of these do you actually use??

03-06-2017, 01:30 PM	#104
MarkJerling Moderator Site Supporter Join Date: May 2012 Location: Wairarapa, New Zealand Photos: Gallery \| Albums Posts: 20,403	Originally posted by ogl K-5's AF was the worst for me. Just a bit better than *ist DS. If the different users have different opinions about the same camera, it means SAFOX system is not stable. Let's say the truth - even last SAFOX 12 is one of the weakest AF module between the cameras of other brands. Even Nikon D5600 has real tracking AF and has 39 AF points. I has the feeling that cross-type AF sensors of Pentax hardly help to camera. The old Multi-Cam of Nikon with 51 AF points with only 15 cross-type sensors is far better than K-1's AF. Pentax cameras have no good level tracking AF system. It's true. If we can make photos in AF-C mode doesn't mean that this system is really good. We just the users with high level of self-adaptive and it's all. We have already got accustomed to such work of SAFOX and nothing else. I can't compare with other brands because I don't shoot other brands. I presume you do? All I can say is that I have no problem shooting fast moving sports such as football or cycling with my K-5. Clearly my K-5 has an AF module that works!

03-06-2017, 09:13 PM	#105
ogl Banned Join Date: Dec 2007 Location: Sankt Peterburg Photos: Gallery \| Albums Posts: 8,382	Originally posted by MarkJerling I can't compare with other brands because I don't shoot other brands. I presume you do? All I can say is that I have no problem shooting fast moving sports such as football or cycling with my K-5. Clearly my K-5 has an AF module that works! to simply work and to work very effective are not the same.