photozone.de said the same thing as well, for the 24-70 SSM but not any other SSM. Seems like not all SSM are born equal.
Is Sony still the camera with the fastes AF?
- Thread starter shaoken
- Start date
- Status
photozone.de said the same thing as well, for the 24-70 SSM but not any other SSM. Seems like not all SSM are born equal.
Thanks 
I'm a little confused though, not disputing the test but I'm seeing results that don't tally that I hope someone can clarify.
Taking the D3 and the 1D for example:
The D3 apparently has a lower percentage of out of focus shots compared to the 1DIII, in other words it has a higher percentage of in focus shots compared to the 1DIII. So the D3 should produce more in focus results than the 1DIII.
What's confusing is how the 1DIII ends up with almost 50% more in focus shots than the D3. I know the 1DIII has a higher frame rate than the D3 but that's only about 10% higher, and the test says it's (slightly) less accurate than the D3 to boot.
And I've checked about the Olympus claim too... "world's fastest autofocus... Among interchangeable lenses for D-SLR cameras. In combination with the new Olympus ZUIKO DIGITAL ED 12-60mm, 1:2.8-4.0 SWD and E-3, as of October 17 2007, according to a survey by Olympus." I know the lens is different this is just FYI
I'm a little confused though, not disputing the test but I'm seeing results that don't tally that I hope someone can clarify.
Taking the D3 and the 1D for example:
The D3 apparently has a lower percentage of out of focus shots compared to the 1DIII, in other words it has a higher percentage of in focus shots compared to the 1DIII. So the D3 should produce more in focus results than the 1DIII.
What's confusing is how the 1DIII ends up with almost 50% more in focus shots than the D3. I know the 1DIII has a higher frame rate than the D3 but that's only about 10% higher, and the test says it's (slightly) less accurate than the D3 to boot.
And I've checked about the Olympus claim too... "world's fastest autofocus... Among interchangeable lenses for D-SLR cameras. In combination with the new Olympus ZUIKO DIGITAL ED 12-60mm, 1:2.8-4.0 SWD and E-3, as of October 17 2007, according to a survey by Olympus." I know the lens is different this is just FYI
GENO
Active Member
U must really do a full test BY yrself..not reading up some forum telling their babababa..
@900 with 24-70 is fast due to ssm build i agree, seriously, i can put another pc of lens at 1/4 price of the CZ and still get the same speed..:bsmilie:
For yr interest, i own the sony..tested the rest..its the slowest...but i still use it..WHY??? too many minolta lenses..:sweat:
Hi Everyone....
On the subject of focusing, its a balance of focus resolution and focus speed of the lens design itself. A lens that have by reference - number of turns or distance of turns of the focus ring, should be taken into consideration. Lens that have very short focus travel tends to result in focus errors as slight change in focusing rotation moves translates to a great distance in the focus plane.
You can have a lens that focuses fast with only a little bit of resolution for manual focus.
On the other, several other real life factors needs to be included in the consideration.
Among those factors that results in AF failures are - bad selection of focus point. Especially on moving targets, selection of focus point is important. Which again depends greatly on situation - on a case by case basis. Factors such as lighting condition, subject brightness, contrast etc. Its then coupled with the right DOF achieved by knowledge of which aperture to work on will help you get shots that are in focus and focusing on the right thing. All these really depends on how much experience you have with your gear to achieve your desired shoot.
You can have the fastest and greatest AF but if there is no handling touch to it then it would certainly fail. I've always put this in the back of my mind that even those days where AFs were like snails....photographers can even take the greatest photographs...How do they do it? More so for the manual focus.
Cheers...
rgds,
Sulhan
On the subject of focusing, its a balance of focus resolution and focus speed of the lens design itself. A lens that have by reference - number of turns or distance of turns of the focus ring, should be taken into consideration. Lens that have very short focus travel tends to result in focus errors as slight change in focusing rotation moves translates to a great distance in the focus plane.
You can have a lens that focuses fast with only a little bit of resolution for manual focus.
On the other, several other real life factors needs to be included in the consideration.
Among those factors that results in AF failures are - bad selection of focus point. Especially on moving targets, selection of focus point is important. Which again depends greatly on situation - on a case by case basis. Factors such as lighting condition, subject brightness, contrast etc. Its then coupled with the right DOF achieved by knowledge of which aperture to work on will help you get shots that are in focus and focusing on the right thing. All these really depends on how much experience you have with your gear to achieve your desired shoot.
You can have the fastest and greatest AF but if there is no handling touch to it then it would certainly fail. I've always put this in the back of my mind that even those days where AFs were like snails....photographers can even take the greatest photographs...How do they do it? More so for the manual focus.
Cheers...
rgds,
Sulhan
Galdor
Senior Member
ricleo
Senior Member
quite interested to see some links to prove the above comment
although i will be the first to admit my 5d2 doesn't focus that fast hehe.....
Hi, apparently you have some reading disorder because i just said i have tested all 3 SIDE BY SIDE MYSELF. Although i don't own all 3, i have friends who do, and all 3 of us have concluded that out of the 3 24-70s, the CZ is the fastest. If you don't believe, well that's not my problem.
I think there are some misunderstanding here, Geno talking about Sony/Minolta mount lenses, CZ 24-70mm lens speed is not the fastest compare to other lenses, while Dominum talking about 3 different systems, where as CZ 24-70mm on A900 still the fatest.
Do I get it right? :sweat:
tchoonyong
New Member
To compare by brand specify the war will never end at all.
Every company have their way to build their own AF.
It does not mean having more AF point means faster. From my own experience, sony AF is faster in the centre point as they had dual centre cross AF which not all DSLR had.
And to emphasize, i do not agree that there are world best AF Camera.... As all of these are just a marketing method to attract more customers to buy their products.
Every company have their way to build their own AF.
It does not mean having more AF point means faster. From my own experience, sony AF is faster in the centre point as they had dual centre cross AF which not all DSLR had.
And to emphasize, i do not agree that there are world best AF Camera.... As all of these are just a marketing method to attract more customers to buy their products.
... Didn't Overture posted a link at page 1???
Let me paste for you guys to read for those who are lazy to click: ;p
===================================================================
Source: fotoMAGAZIN (Germany) (fotoMagazin), Nr. 5/2008, p30ff.
blocked URL
Results
Gear:
* 40D: Canon 40D + EF 70-200 f/2.8 L IS USM (3230,- German list price)
* 1D: Canon 1D-mkIII + EF 70-200 f/2.8 L IS USM (6390,- German list price)
* 1Ds: Canon 1Ds-mkIII + EF 70-200 f/2.8 L IS USM (10290,- German list price)
* D80: Nikon D80 + Nikkor 70-200 f/2.8 VR G IF-ED (3260,- German list price)
* D300: Nikon D300 + Nikkor 70-200 f/2.8 VR G IF-ED (4120,- German list price)
* D3: Nikon D3 + Nikkor 70-200 f/2.8 VR G IF-ED (7140,- German list price)
* E3: Olympus E3 + Zuiko 50-200 f/2.8-3.5 ED SWD (3100,- German list price)
* K20: Pentax K20D + DA* 50-135mm f/2.8 ED IF SDM (2180,- German list price)
* A700: Sony Alpha 700 + 70-200 f/2.8 G SSM (3800,- German list price)
(Order and prices as chosen by original print, K20 has best price in the field due to affordable DA* lens.)
Result (Test #4 -- Relative ratio of out-of-focus shots, central AF):
1. A700: 2.1 %
2. 40D: 6.4 %
3. D3: 9.2 %
4. D300: 10.0 %
5. 1D: 11.7 %
6. 1Ds: 14.3 %
7. E3: 17.8 %
8. D80: 23.2 % (but 10.7 % with multi-field AF)
9. K20: 25.6 % (and 43.6 % with multi AF)
Result (Test #3 -- Absolute number of in-focus shots, central AF):
1. 1D: 30.2 images
2. D3: 21.6 images
3. 40D: 20.6 images
4. A700: 15.3 images
5. 1Ds: 15.0 images
6. D300: 13.5 images
7. D80: 8.6 images (10.7 with multi AF)
8. E3: 7.4 images
9. K20: 5.8 images
3.7 images (with multi-field AF, other multi-field AF results ommitted here)
Method:
* Continous shots (burst, continous AF aka AF.C) of an approaching red Chrysler PT cruiser, tripod mounted and panned centering onto radiator grill. Photograph besides road. Road straight. Test done on a closed military airport.
* Controlled speed: 100 km/h (62 mph).
Starting photo burst at 100m distance (109 yards, 328 ft). Stopping when passing by.
Time to drive through 100m: 3.6s.
* Focal length: 35mm equivalent of 200mm (135mm for K20).
* Aperture: f/2.8 or widest (Olympus).
* Shutter: 1/320 or shorter.
* Light: Daylight overcast with grayness. Constant conditions.
* Storage: JPEGs onto Sandisk Extreme III.
* Statistics: 6 series, averaged.
Short verdict (mine):
While the K20 comes out last, it did do its job surprisingly well after reading all the complaints around. We all don't position the K20 as the sport photographer's machine gun and rightly so with only 20% of the available peak performance on the market (1D).
However, it is more than sufficient for everyday photography with ~75% of images in focus.
If I may make my personal scale:
Outstanding: 95-100% (relative ratio of in-focus shots)
Excellent: 85-95%
Good: 75-85%
Average: 25-75%
Poor: 0-25%
Then I get:
* Outstanding: A700
* Excellent: 40D, D3, D300, 1D, 1Ds
* Good: E3, D80, K20
====================================================================
Let me paste for you guys to read for those who are lazy to click: ;p
===================================================================
Source: fotoMAGAZIN (Germany) (fotoMagazin), Nr. 5/2008, p30ff.
blocked URL
Results
Gear:
* 40D: Canon 40D + EF 70-200 f/2.8 L IS USM (3230,- German list price)
* 1D: Canon 1D-mkIII + EF 70-200 f/2.8 L IS USM (6390,- German list price)
* 1Ds: Canon 1Ds-mkIII + EF 70-200 f/2.8 L IS USM (10290,- German list price)
* D80: Nikon D80 + Nikkor 70-200 f/2.8 VR G IF-ED (3260,- German list price)
* D300: Nikon D300 + Nikkor 70-200 f/2.8 VR G IF-ED (4120,- German list price)
* D3: Nikon D3 + Nikkor 70-200 f/2.8 VR G IF-ED (7140,- German list price)
* E3: Olympus E3 + Zuiko 50-200 f/2.8-3.5 ED SWD (3100,- German list price)
* K20: Pentax K20D + DA* 50-135mm f/2.8 ED IF SDM (2180,- German list price)
* A700: Sony Alpha 700 + 70-200 f/2.8 G SSM (3800,- German list price)
(Order and prices as chosen by original print, K20 has best price in the field due to affordable DA* lens.)
Result (Test #4 -- Relative ratio of out-of-focus shots, central AF):
1. A700: 2.1 %
2. 40D: 6.4 %
3. D3: 9.2 %
4. D300: 10.0 %
5. 1D: 11.7 %
6. 1Ds: 14.3 %
7. E3: 17.8 %
8. D80: 23.2 % (but 10.7 % with multi-field AF)
9. K20: 25.6 % (and 43.6 % with multi AF)
Result (Test #3 -- Absolute number of in-focus shots, central AF):
1. 1D: 30.2 images
2. D3: 21.6 images
3. 40D: 20.6 images
4. A700: 15.3 images
5. 1Ds: 15.0 images
6. D300: 13.5 images
7. D80: 8.6 images (10.7 with multi AF)
8. E3: 7.4 images
9. K20: 5.8 images
3.7 images (with multi-field AF, other multi-field AF results ommitted here)
Method:
* Continous shots (burst, continous AF aka AF.C) of an approaching red Chrysler PT cruiser, tripod mounted and panned centering onto radiator grill. Photograph besides road. Road straight. Test done on a closed military airport.
* Controlled speed: 100 km/h (62 mph).
Starting photo burst at 100m distance (109 yards, 328 ft). Stopping when passing by.
Time to drive through 100m: 3.6s.
* Focal length: 35mm equivalent of 200mm (135mm for K20).
* Aperture: f/2.8 or widest (Olympus).
* Shutter: 1/320 or shorter.
* Light: Daylight overcast with grayness. Constant conditions.
* Storage: JPEGs onto Sandisk Extreme III.
* Statistics: 6 series, averaged.
Short verdict (mine):
While the K20 comes out last, it did do its job surprisingly well after reading all the complaints around. We all don't position the K20 as the sport photographer's machine gun and rightly so with only 20% of the available peak performance on the market (1D).
However, it is more than sufficient for everyday photography with ~75% of images in focus.
If I may make my personal scale:
Outstanding: 95-100% (relative ratio of in-focus shots)
Excellent: 85-95%
Good: 75-85%
Average: 25-75%
Poor: 0-25%
Then I get:
* Outstanding: A700
* Excellent: 40D, D3, D300, 1D, 1Ds
* Good: E3, D80, K20
====================================================================
Sad but true. When you hang it around your neck it senses your shirt and starts focusing. I remember in the SLR days, the camera needed two sensors to be detected before eye-start will activate. Grip sensor and eye sensor. If I'm not wrong, only A700 and A900 have both grip and eye sensor.
A850 no eye-start AF and the rest have eye sensors only to activate eye-start AF
If you're referring to me, I did click and I found something that confused me, which I've highlighted in a previous post and no one has yet been able to shed light on why the results seem a little off statistically. (ie I'm not saying brand X should be better)
If they can make a hash of the results statistically then it doesn't bode well for the rest of the figures or the test either.
lawrencewkl
Member
Oberon is right. My old film minolta 700si have a grip sensor and a eye sensor. Both need to be activated for the eye start AF to work. It is also very easy to disable this function via a switch. Dun understand why Sony wants to implement a feature half way. It has been a love / hate feeling since Sony took over Minolta.
It isn't a half-hearted implementation... there are reasons for it... the A700 was released with the same eye-start sensor plus grip sensors... but in Europe, some people being allergic to Nickel complained about the grip sensor containing nickel resulting in an allergic outbreak... so SONY removed the grip sensor and has since not put it back for perhaps a lack of suitable material not containing nickel...
The A900 does not have the eye-start feature nor the grip sensor...
The A700 was the fastest focusing and more accurately focused camera back then based only on the central double-cross AF sensor... the conclusion made by quite a few websites was that the double-cross AF sensor by SONY was quite clearly superior...
I've no idea if the A900 is that fast or even faster when used with a matched lens...
However, in my subjective experience when in real life context, the wide AF sensor of the Canon 1D series is clearly much superior to the Sony's AF system... the speed at which the wide AF on a Canon 1D can lock on (even on a blank but textured wall) is quite a bit amazing... especially at low light and action...
As for Dominium's test using the 24-70 for each brand, it simply says two things - 1) The Sony AF module on the A900 is on par with the rest of the world where the central AF sensor is used. 2) The CZ 24-70 is a very close match for the A900 and the other brands may not have matched the lenses to the body so well... finally, I still feel impatient with the A900 when using my cheap consumer lenses... in part they do not have SSM and when I use the 70-300G SSM, the AF performs miles better... however it's still not of the same standard as for the 1D series. However, when testing the 1D series, I was using the top L lenses (85mm, 24-70, etc)... so my experience is rather subjective since the lenses used were of different classes...
I also have no idea if the A900's wide AF function is accurate or fast... I never use it... I prefer to use the central AF sensor... it's a habit developed from the film days... so my question perhaps to the contributors to this thread is - do you usually use the central AF or let the camera decide where to focus by using the wide AF option?
Jed, the discrepancy could be due to the erratic AF behaviour of the 1DMkIII which lots of Canon users complained about and which Canon has not been able to fix emphatically... apparently the new 1DMkIV will not have this problem...
I've no idea if the A900 is that fast or even faster when used with a matched lens...
However, in my subjective experience when in real life context, the wide AF sensor of the Canon 1D series is clearly much superior to the Sony's AF system... the speed at which the wide AF on a Canon 1D can lock on (even on a blank but textured wall) is quite a bit amazing... especially at low light and action...
As for Dominium's test using the 24-70 for each brand, it simply says two things - 1) The Sony AF module on the A900 is on par with the rest of the world where the central AF sensor is used. 2) The CZ 24-70 is a very close match for the A900 and the other brands may not have matched the lenses to the body so well... finally, I still feel impatient with the A900 when using my cheap consumer lenses... in part they do not have SSM and when I use the 70-300G SSM, the AF performs miles better... however it's still not of the same standard as for the 1D series. However, when testing the 1D series, I was using the top L lenses (85mm, 24-70, etc)... so my experience is rather subjective since the lenses used were of different classes...
I also have no idea if the A900's wide AF function is accurate or fast... I never use it... I prefer to use the central AF sensor... it's a habit developed from the film days... so my question perhaps to the contributors to this thread is - do you usually use the central AF or let the camera decide where to focus by using the wide AF option?
Jed, the discrepancy could be due to the erratic AF behaviour of the 1DMkIII which lots of Canon users complained about and which Canon has not been able to fix emphatically... apparently the new 1DMkIV will not have this problem...
I understand what you're saying about the 1DIII's flakey focusing, but in this situation, *as presented in the extract of the article anyway* the statistics seem to be contradicting each other
Either the accuracy percentage is wrong, or the number of absolute in focus frames is wrong, or the figures are from separate tests...
I understand what you're saying about the 1DIII's flakey focusing, but in this situation, *as presented in the extract of the article anyway* the statistics seem to be contradicting each other
Either the accuracy percentage is wrong, or the number of absolute in focus frames is wrong, or the figures are from separate tests...
Not sure if the accuracy percentage is visual or just from the camera's focus beep? Maybe the numbers are indeed wrong...
XkaOnslaught
New Member
i laugh at all the emphasis at AF SPEED when AF ACCURACY is more important imo.
- Status
Similar threads
