-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
When will we get a solid report on EM1 wPDAF&CDAF w/micro4/3's lenses?
- Thread starter jl123
- Start date
DLBlack
Forum Pro
There was one reviewer that was at the Olympus event that used an 70/1.8@ 1.8 to follow the lady on horseback coming out of the lake. It did really well. So there are examples out there.
vincentnyc
Senior Member
This is a great point, btw, jl123. Kudos for being specific and pointing it out.
Like you, I also have been scratching my head regarding their announced capabilities vis-a-vis the EM1's autofocus capabilities SPECIFICALLY as it pertains to Micro 43rds lenses. To me, the Micro43rds system was designed sans legacy concerns, and, as such, it has the freedom to have motors which function as they should during this decade. things like focusing internally, faster, quieter, are all de rigueur nowadays, whereas using legacy systems' lenses, they were not optimized for the Micro 43rds bodies' use.
As a result, why wouldn't a flagship model be optimized to use ALL the benefits available to it from the dedicated Micro43rds lenses? It is such a pity that the entire and sole purpose of the on-sensor PDAF pixels are strictly for legacy purposes. It's great for folks with legacy 43rds lenses, mind you, but it doesn't speak positively of a forwarding looking, flagship model. For any flagship models, it needs to be optimized to do the best with the dedicated accessories!
So, let's hope some knowledgeable folks reply to this thread!
Thanks for putting it up and giving it the attention it deserves. Kudos!
RealPancho
Senior Member
At first, I was under the impression that this could be manually selected. Then, from reading other threads, I gathered that it was automatically selected dependent upon whether a 4/3 or m4/3 lens is mounted. I can't find anywhere in the manual where this is addressed.
--
Frank
--
Frank
Brian Wadie
Forum Pro
My reading of the manual is that:
If you put a 4/3rds lens on the front it automatically selects PDAF (it can't use CDAF apparently)
If you put a m4/3rds lens on the front it automatically selects CDAF, unless you switch to CAF when it switches to the composite PDAF / CDAF focus system
Bob Greenberg
Well-known member
As I uderstand it, pdaf works effectively with m43 lenses if you use continuous af, so it does significantly help with tracking. That means we do gain benefits as m43 users!
Not many people have delivery yet (including me!) so we should learn more in the next few weeks.
Bob
Not many people have delivery yet (including me!) so we should learn more in the next few weeks.
Bob
Godfrey
Forum Pro
Don't get the point of your angst. The E-M1 is optimized for mFT lenses, of course, which are primarily designed for CDAF. In C-AF and C-AF tracking, PDAF supplements the CDAF system with these lenses to provide better tracking capabilities.
CDAF is not the best solution for any focus tracking system due to how it is designed to drive the focusing mount.
CDAF is not the best solution for any focus tracking system due to how it is designed to drive the focusing mount.
Godfrey
Forum Pro
This is correct. The focusing system is designed to choose the best focusing protocol for the lens automatically.
Timur Born
Senior Member
I spoke to Toshi Terada today and got this answer:
"For targets that are within the phase-detection (PD) areas there is no official statement (from Mr. Terada) as to whether AF-C with MFT lenses uses pure PD or a mixture of PD plus contrast detection (CD). But once a tracked target leaves the PD areas the E-M1 will automatically switch to CD only. I assume the latter is also true when you choose an AF frame outside the PD areas, but my quick tests during the event didn't show a clear difference and the PD areas are not displayed on screen with MFT lenses."
The "no official statement" part (likely) means that Olympus don't want to tell how it all works in detail, so it's a kind of trade secret.
"For targets that are within the phase-detection (PD) areas there is no official statement (from Mr. Terada) as to whether AF-C with MFT lenses uses pure PD or a mixture of PD plus contrast detection (CD). But once a tracked target leaves the PD areas the E-M1 will automatically switch to CD only. I assume the latter is also true when you choose an AF frame outside the PD areas, but my quick tests during the event didn't show a clear difference and the PD areas are not displayed on screen with MFT lenses."
The "no official statement" part (likely) means that Olympus don't want to tell how it all works in detail, so it's a kind of trade secret.
In my experience with em1 the hit rate is much bhigher than the em5 we using caf and micro 4/3 rods lenses . There is no on screen indicator to show you how pdaf is performing so its hard to get a real feel, but if hit rate is any indication it's much much better.
vincentnyc
Senior Member
Oh my goodness. I'm not optimistic.Timur Born wrote:
I spoke to Toshi Terada today and got this answer:
"For targets that are within the phase-detection (PD) areas there is no official statement (from Mr. Terada) as to whether AF-C with MFT lenses uses pure PD or a mixture of PD plus contrast detection (CD). But once a tracked target leaves the PD areas the E-M1 will automatically switch to CD only. I assume the latter is also true when you choose an AF frame outside the PD areas, but my quick tests during the event didn't show a clear difference and the PD areas are not displayed on screen with MFT lenses."
The "no official statement" part (likely) means that Olympus don't want to tell how it all works in detail, so it's a kind of trade secret.
To me, the evasiveness of Mr. Terada and, essentially, everyone associated with Olympus, after the initial press release, does not bode well.
It really feels like Olympus exaggerated its claims about what the EM1 actually does with PDAF when MICRO 43rds lenses are used. I think they were concerned that their customers would be unhappy that ALL Micro 43rds lenses would not benefit whatsoever from their "on sensor PDAF", and that this integration of PDAF on the sensor itself is merely an effort to appease legacy 43rds lens owners.
This would have no benefit to draw any de novo customers to Olympus EM1 as all de novo customers are likely to ONLY purchase strictly Micro 43rds lenses.
Quite disappointing. They exaggerated, and now, do not wish to disclose this to the public.
Does anyone else feel this way?
I certainly 'dont' feel this way.
1) I dont really feel that Olympus has actually made extravagant claims over the effectiveness of using PDAF with CDAF continuous focusing with M43 lenses. I recall, in particular, when the 'leaked' video of the E-M1 came out, that PDAF's abilities with M43 lenses wasnt even mentioned as a benefit of the camera.
2) On sensor PDAF for mirrorless cameras is not new - in particular it has been on the Nex 6 for around about a year now (I doubt Olympus's implementation is much different). I have never ever seen it written on this forum that someone was switching to the Nex6 for PDAF. In fact we have heard very little about it - implying that its implementation is 'incremental' benefits rather than 'transforming' performance.
3) I feel it is likely that this will be even more the case with the E-M1. The E-M1's s-af is much faster than the Nex 6. So users can get 'reasonable' focusing performance on some moving objects already by using s-af. Over that performance c-af needs to be quite a bit better to improve on existing s-af performance.
The jury is still out (as far as I can tell) about how effective the PDAF actually is for M43 lenses - so we still have to see. Personally I dont feel we have been promised much more than a step in the right direction and that is what I am expecting and hoping for.
vincentnyc
Senior Member
Adaf:
officially, certainly, they're currently claiming that with C-AF, a combination of PDAF and CDAF is used. this is entirely clear.
i'm not sure your post addressed this specifically.
no one is doubting that CDAF is rather good on the EM1 currently. check, understood, agreed.
the issue is that they specifically that in C-AF, that a HYBRID autofocus system is used. then, as the EM1 is rolled out, they certainly have seemed to go silent on this - totally silent. zilch. only via the initial press announcement.
i'm wondering if this is merely a minority's perception or whether a good number of other folks also feel this way.
officially, certainly, they're currently claiming that with C-AF, a combination of PDAF and CDAF is used. this is entirely clear.
i'm not sure your post addressed this specifically.
no one is doubting that CDAF is rather good on the EM1 currently. check, understood, agreed.
the issue is that they specifically that in C-AF, that a HYBRID autofocus system is used. then, as the EM1 is rolled out, they certainly have seemed to go silent on this - totally silent. zilch. only via the initial press announcement.
i'm wondering if this is merely a minority's perception or whether a good number of other folks also feel this way.
I'm not sure I understand how the entire Olympus organization has exaggerated and been evasive.
M4/3 lenses, designed from the ground up to be focused CDAF, focus in CDAF on the E-M1 when in AF-S mode, which has been fast and not problematic. This info was public before the camera shipped.
M4/3 lenses do take advantage of the PDAF sensors in AF-C mode, to allow better tracking of moving subjects, which has been a weakness of M4/3 CDAF. This info was also made public before the camera shipped. Initial user reports seem to confirm that AF-C focus speed has improved over the E-M5.
So the new sensor scheme does improve M4/3 AF performance.
Where's the problem?
M4/3 lenses, designed from the ground up to be focused CDAF, focus in CDAF on the E-M1 when in AF-S mode, which has been fast and not problematic. This info was public before the camera shipped.
M4/3 lenses do take advantage of the PDAF sensors in AF-C mode, to allow better tracking of moving subjects, which has been a weakness of M4/3 CDAF. This info was also made public before the camera shipped. Initial user reports seem to confirm that AF-C focus speed has improved over the E-M5.
So the new sensor scheme does improve M4/3 AF performance.
Where's the problem?
No,vincentnyc wrote:
Oh my goodness. I'm not optimistic.Timur Born wrote:
I spoke to Toshi Terada today and got this answer:
"For targets that are within the phase-detection (PD) areas there is no official statement (from Mr. Terada) as to whether AF-C with MFT lenses uses pure PD or a mixture of PD plus contrast detection (CD). But once a tracked target leaves the PD areas the E-M1 will automatically switch to CD only. I assume the latter is also true when you choose an AF frame outside the PD areas, but my quick tests during the event didn't show a clear difference and the PD areas are not displayed on screen with MFT lenses."
The "no official statement" part (likely) means that Olympus don't want to tell how it all works in detail, so it's a kind of trade secret.
To me, the evasiveness of Mr. Terada and, essentially, everyone associated with Olympus, after the initial press release, does not bode well.
It really feels like Olympus exaggerated its claims about what the EM1 actually does with PDAF when MICRO 43rds lenses are used. I think they were concerned that their customers would be unhappy that ALL Micro 43rds lenses would not benefit whatsoever from their "on sensor PDAF", and that this integration of PDAF on the sensor itself is merely an effort to appease legacy 43rds lens owners.
This would have no benefit to draw any de novo customers to Olympus EM1 as all de novo customers are likely to ONLY purchase strictly Micro 43rds lenses.
Quite disappointing. They exaggerated, and now, do not wish to disclose this to the public.
Does anyone else feel this way?
Have you contacted "Wiki Leaks", they may have the answer!
M
MPA1
Guest
Would have been nice if there was an EVF overlay to show the area, wouldn't it?
Firmware update, Olympus?!
M
MPA1
Guest
I agree.
And to return to the OP, where are the tests done by reviewers (outside the horses in the water thing I am getting bored of seeing!) testing this to see how well it works?
Optical1
Senior Member
I think that you need to read up on the difference between PDAF lenses and lenses that are CDAF optimized. The reason that they are talking so specifically about legacy 4/3 glass is that the older non-optimized lenses had focussing elements that were large and did not move fast enough to occilate through the focus point for CDAF to focus quickly. PDAF improves focus speed greatly because the camera knows where to focus.vincentnyc wrote:
Oh my goodness. I'm not optimistic.Timur Born wrote:
I spoke to Toshi Terada today and got this answer:
"For targets that are within the phase-detection (PD) areas there is no official statement (from Mr. Terada) as to whether AF-C with MFT lenses uses pure PD or a mixture of PD plus contrast detection (CD). But once a tracked target leaves the PD areas the E-M1 will automatically switch to CD only. I assume the latter is also true when you choose an AF frame outside the PD areas, but my quick tests during the event didn't show a clear difference and the PD areas are not displayed on screen with MFT lenses."
The "no official statement" part (likely) means that Olympus don't want to tell how it all works in detail, so it's a kind of trade secret.
To me, the evasiveness of Mr. Terada and, essentially, everyone associated with Olympus, after the initial press release, does not bode well.
It really feels like Olympus exaggerated its claims about what the EM1 actually does with PDAF when MICRO 43rds lenses are used. I think they were concerned that their customers would be unhappy that ALL Micro 43rds lenses would not benefit whatsoever from their "on sensor PDAF", and that this integration of PDAF on the sensor itself is merely an effort to appease legacy 43rds lens owners.
This would have no benefit to draw any de novo customers to Olympus EM1 as all de novo customers are likely to ONLY purchase strictly Micro 43rds lenses.
Quite disappointing. They exaggerated, and now, do not wish to disclose this to the public.
Does anyone else feel this way?
When the CDAF capable mk 2 lenses were released, they redesigned the focusing elements to be smaller, and improved the focus motor to allow for fast focusing with CDAF. The key here is that CDAF enhancements do nothing to reduce the effectiveness of PDAF - if nothing else it actually speeds up the already sufficiently speedy PDAF.
Now, since all m43 lenses are designed with CDAF in mind, they are all quite capable of fast AF. There is nothing technically limiting m43 lenses from performing as well as 4/3 glass when mounted to an E-M1. So, Olympus would have to intentionally cripple the E-M1 when using m43 glass if the AF doesn't meet or beat 4/3 glass.
This conjecture that the m43 glass is crippled, or will be ineffective using PDAF on an E-M1 is nothing more than conspiracy theory...
Keyboard shortcuts
- f
- Forum