p.14 #1 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
The Sony 50/1.2 has a lot of glass to move. Still it worked absolutely great even with the old A7III. I could easily shoot bursts with it with incoming kids and have many of the pictures in focus (I didn't do any official 'test shoots', but it was so good I could use it to get a shot I wanted). Regarding the new 85 GM II, I would just wait and see what the folks will report.. I'm sure it will be pretty fine, though I would not be surprised if Sony would introduce a f/1.2 version a little later on..
p.14 #3 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
I'm always skeptical of AF reports until I try the lens in hand. Many reviewers simply confuse the wrong shutter speed with poor AF performance. This is especially true when using a high MP body as you are essentially using a much longer focal length when you zoom in. There are shots in this thread that are clearly motion blur from a bad shutter speed. My experience with the 85mm DG DN was that it struggled keeping up for sports. The newer GM lenses that I use have no problem keeping up (135mm GM and 50mm 1.2 most often) as long as the body is not the cause of the issues (even with the a9iii, there are times where the camera is the cause of missed focus because it jumps off subject or becomes indecisive).
p.14 #4 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
The 50mm f/1.2 GM has two pairs of focus motors - two bigger ones on one focus group, two smaller ones on another focus group.
The 135 GM also have two pairs of focus motors.
The 85mm GM II has one pair of focus motors, reportedly bigger than usual to speed the AF in this lens.
Just received my GM II. Taking it out to shoot tomorrow.
Comparing the GM and GM II, the new lens is narrower near the mount. I recall that a few people complained about clearance between the original GM and the grip on Sony cameras, so the new will make a few people happier.
p.14 #5 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
Tony Ross wrote:
The 50mm f/1.2 GM has two pairs of focus motors - two bigger ones on one focus group, two smaller ones on another focus group.
The 135 GM also have two pairs of focus motors.
The 85mm GM II has one pair of focus motors, reportedly bigger than usual to speed the AF in this lens.
Just received my GM II. Taking it out to shoot tomorrow.
Comparing the GM and GM II, the new lens is narrower near the mount. I recall that a few people complained about clearance between the original GM and the grip on Sony cameras, so the new will make a few people happier....Show more →
Wait, so it's not 4x AF motors?
Sep 20, 2024 at 05:31 AM
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.14 #6 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
j4nu wrote:
Wait, so it's not 4x AF motors?
It is only 2 XD motors. I thought I saw reports of 4 motors, but this is directly from the Sony site:
"2 XD (extreme dynamic) linear motors and high-speed control give the FE 85mm F1.4 GM II AF speed that is up to 3x faster than previous model. Moving subject tracking is improved by as much as 7x, allowing reliable capture even when using extremely narrow depth of field at F1.4. The 120 fps capability of the α9 III with its global shutter is fully supported."
p.14 #7 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
Steve Spencer wrote:
It is only 2 XD motors. I thought I saw reports of 4 motors, but this is directly from the Sony site:
"2 XD (extreme dynamic) linear motors and high-speed control give the FE 85mm F1.4 GM II AF speed that is up to 3x faster than previous model. Moving subject tracking is improved by as much as 7x, allowing reliable capture even when using extremely narrow depth of field at F1.4. The 120 fps capability of the α9 III with its global shutter is fully supported."
p.14 #8 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
j4nu wrote:
Not to be too picky, but doesn't 135/1.8 have to move even more glass?
F1.4 vs f1.8. Plus 135MM primes tend to have more robust AF motors and are priced accordingly. Fast 85MM tend to be designed and optized for portraits if you look at how they are marketed so less emphasis on AF
p.14 #9 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
I just picked up the lens from my local shop, took a few test shots and looked through the EVF to review the photos. I will take a proper look in LR later. My very early initial impressions on my A7RIV;
- Subjects pop away from the background a lot more than the original.
- Its MUCH sharper wide open and does 61mp justice which in my opinion the MK1 did not.
- The MKII has MUCH faster and silent AF.
- Much lighter and better balanced on the camera.
p.14 #10 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
Mystik wrote:
F1.4 vs f1.8. Plus 135MM primes tend to have more robust AF motors and are priced accordingly. Fast 85MM tend to be designed and optized for portraits if you look at how they are marketed so less emphasis on AF
Yes, it seems Sony went the old route by including 2 AF motors...
Sep 20, 2024 at 08:14 AM
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.14 #11 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
j4nu wrote:
Thanks, so it might be the same case as with 50/1.4GM where the tracking is not as great as one would expect...
I am not sure the numbers of XD motors is going to be a great predictor of how well the AF works. The size of the motors and the size of the focussing group(s) they are moving is going to matter at least as much as the number of motors. For the record of the second generation GM lenses here is the number of motors each has:
All else being equal more motors is probably a good thing and especially so if it means more focussing groups and keeping those focussing groups smaller, but all else is not necessarily equal. In my view we will have to wait until we see lots of real world performance to see how the AF actually performs and in my view that is how it always is with AF.
p.14 #12 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
Steve Spencer wrote:
I am not sure the numbers of XD motors is going to be a great predictor of how well the AF works. The size of the motors and the size of the focussing group(s) they are moving is going to matter at least as much as the number of motors. For the record of the second generation GM lenses here is the number of motors each has:
All else being equal more motors is probably a good thing and especially so if it means more focussing groups and keeping those focussing groups smaller, but all else is not necessarily equal. In my view we will have to wait until we see lots of real world performance to see how the AF actually performs and in my view that is how it always is with AF....Show more →
Sure, let's wait and see.
Maybe 85GM II will be the 2-motors exception...
Sep 20, 2024 at 08:56 AM
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.14 #13 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
j4nu wrote:
Sure, let's wait and see.
Maybe 85GM II will be the 2-motors exception...
I don't think there is any sort of exception as it is a basic matters of physics. The speed of the AF will be a function of the power of motors (summed across 2 or 4) by the weight of the elements that are being moved. The problem is we don't know either the power of the motors or weight of the elements being moved so we have to wait to see how it works in practice. By the way, the 35 f/1.4, IMO, has excellent and fast AF with no problems keeping up with even demanding AF needs, so I don't think all 2 motor lenses will be deficient, and even the tests reported recently with the 50 f/1.4 GM although not optimal showed pretty decent AF in a very demanding situation. I think it is one of the consistent features of this new generation of GM lenses with XD motors that they all have excellent AF capabilities.
p.14 #14 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
Steve Spencer wrote:
I don't think there is any sort of exception as it is a basic matters of physics. The speed of the AF will be a function of the power of motors (summed across 2 or 4) by the weight of the elements that are being moved. The problem is we don't know either the power of the motors or weight of the elements being moved so we have to wait to see how it works in practice. By the way, the 35 f/1.4, IMO, has excellent and fast AF with no problems keeping up with even demanding AF needs, so I don't think all 2 motor lenses will be deficient, and even the tests reported recently with the 50 f/1.4 GM although not optimal showed pretty decent AF in a very demanding situation. I think it is one of the consistent features of this new generation of GM lenses with XD motors that they all have excellent AF capabilities....Show more →
Exception to the list you provided, where it's easy to pick the best focusing / tracking lenses and their common element ...
Sep 20, 2024 at 11:41 AM
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.14 #15 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
j4nu wrote:
Exception to the list you provided, where it's easy to pick the best focusing / tracking lenses and their common element ...
And with such a small number of lenses (we are talking 5 here) it is also easy to think that one common element is affecting performance when it is really is a different characteristic altogether that is affecting performance . Like in this case it might be cumulative power of motors and not the number of motors that is affecting performance and make a wrong inference. Five of anything just isn't enough to make a sound inference about whether the presence or absence of any feature is related to performance. It is super easy to see false associations with such a small number of instances (i.e., lenses in this example). Basic statistics tells you that with only 5 examples any association you see would occur simply by chance fairly often. That is why we need tests for AF and not merely look for associations that may or may not be real.
p.14 #16 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
Steve Spencer wrote:
And with such a small number of lenses (we are talking 5 here) it is also easy to think that one common element is affecting performance when it is really is a different characteristic altogether that is affecting performance . Like in this case it might be cumulative power of motors and not the number of motors that is affecting performance and make a wrong inference. Five of anything just isn't enough to make a sound inference about whether the presence or absence of any feature is related to performance. It is super easy to see false associations with such a small number of instances (i.e., lenses in this example). Basic statistics tells you that with only 5 examples any association you see would occur simply by chance fairly often. That is why we need tests for AF and not merely look for associations that may or may not be real....Show more →
Yes, but these are not random 5 lenses, but as you said it yourself, 5 GM 2nd gen. lenses (which means they share some common tech, design capabilities, etc.) .
Anyways, no matter how we look at it, if 85/1.4GM II proves to be the fastest focusing lens of them all, it will still be a 2-motors outlier ...
Sep 20, 2024 at 12:29 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.14 #17 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
j4nu wrote:
Yes, but these are not random 5 lenses, but as you said it yourself, 5 GM 2nd gen. lenses (which means they share some common tech, design capabilities, etc.) .
Anyways, no matter how we look at it, if 85/1.4GM II proves to be the fastest focusing lens of them all, it will still be a 2-motors outlier ...
I think my point is with 5 instances you can't have an outlier and yes these aren't 5 random lenses, but that does not change the fact that you should not draw any inferences about the relation of the presence or absence of any characteristic with any sort of performance from just 5 instances. That is simple basic statistics.
p.14 #18 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
Steve Spencer wrote:
I think my point is with 5 instances you can't have an outlier and yes these aren't 5 random lenses, but that does not change the fact that you should not draw any inferences about the relation of the presence or absence of any characteristic with any sort of performance from just 5 instances. That is simple basic statistics.
That's the thing, it's not "simple basic statistics" because these are not independent cases, but related lenses. 5 lens release of a new (2nd) gen, which span from 04.2019 to 09.2024, is a lot.
You can throw in 70-200GM II as well (edit: I'll counter myself by saying 300GM actually uses *2* motors, as I've just checked).
Anyways, the trend now is visible and it is that 4-motor lenses are better/faster at AF than their same-gen 2-motor counterparts.
Can 85/1.4GM II change that? Sure, but it will be an outlier .
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.14 #19 · Sony 85mm f/1.4 GM II Announced—Pre-Orders Now Open!
j4nu wrote:
That's the thing, it's not "simple basic statistics" because these are not independent cases, but related lenses. 5 lens release of a new (2nd) gen, which span from 04.2019 to 09.2024, is a lot.
You can throw in 70-200GM II as well.
Anyways, the trend now is visible and it is that 4 motor lenses are better/faster at AF than their same-gen 2-motors counterparts.
Can 85/1.4GM II change that? Sure, but it will be an outlier .
You seem to believe that simple basic statistics require independent cases. They do not and in fact in the real world independent cases are exceedingly rare and yet you can and should analyze real world occurrences with basic statistics. In fact, the point of the relevant basic statistics here, the chi square test, is to specifically tell you when it is reasonable to make an inference that the presence or absence of a characteristic (here 2 or 4 motors) is associated with an outcome (here AF performance). The statistic needs more than 5 instances. This is a matter of math and you are wrong about making an inference in this case just as you would be wrong is you said 2 + 2 = 5. I am sure you won't admit to that, but it is the simple facts of evaluating these lenses based on the number of motors as the only consideration.
...
. Like in this case it might be cumulative power of motors and not the number of motors that is affecting performance and make a wrong inference. Five of anything just isn't enough to make a sound inference about whether the presence or absence of any feature is related to performance. It is super easy to see false associations with such a small number of instances (i.e., lenses in this example). Basic statistics tells you that with only 5 examples any association you see would occur simply by chance fairly often. That is why we need tests for AF and not merely look for associations that may or may not be real.