I now have 2 1.4x TC - a Canon and a Kenko Pro 300 DG and will be testing them on a 300 f/4L IS. Unfortunately I don't have a 2x TC to compare to. However, I'll still be able to assess quality of the stacked converters. Given that Canon slows down AF speed by 35% with a 1.4x TC and 70% with a 2x TC, then stacking will result in faster AF as the camera only sees one TC, the one attached to the lens mount. This has been verifed by a number of people I've talked to.
later in the same thread I, Phil Bonner, commented on my experience with the 1.4xTC as follows:
When I 1st bought them I tested 70-200IS (+and -) f/1.4TC AFspeed and it is slower with the 1.4 TC, but, only by a fraction of a second. I used A1 Servo and had my son run at me from across the yard about a dozen times. Originally I thought I may be having a problem and inquired here at FM. Ben Hone, San Diego, straightened me out. It's in the literature somewhere too.
In my testing described above, with and without the TC, the TC slowed down my auto focus acquisition by about 1/3. In practical terms w/o the TC the 70-200IS focus locked onto my son in about 2 or 3 running steps coming towards me. With the TC it acquired focus in about 3 or 4 running steps coming towards me, i.e.: Not 4-6 running steps as would be the case if the TC slowed down auto focus by the 50-67% you are claiming.
The loss of focus speed is not bad at all on the 70-200IS which, BTW, has specially optimized auto focus for speed. This is the only Canon 70-200 lens that is speed-optimized.
FYI. The TC literature only states that the TC will slow down auto focus but did not specify the degree.
kecohen please leave your snide attitude at the door - thanks.
Jul 26, 2005 at 10:40 PM
The Big Bad Offline • • • • [X]
p.4 #2 · F2.8 vs. F4? Is it really THAT signifigant?
Its been well documents by Chuck Westfall and countless others than the reduction in AF speed is about 50-60% of the 1.4x and up to 75% for the 2x TC.
Its not really a debatable issue as its actually programed into the firmware of the camera that when the TC is added, it slows the AF.
Now with a cheap TC without all the pins, it doesnt do this automatically, but due to the stop or two loss, the end result is the AF still can and usually will slow down a little.
Im sorry Phil but a test with how many steps your son can take isnt as convincing as Chuck Westfall's word on the matter.
I personally have not heard of a special AF optimization on the 70-200 either. A few of the super teles like the 300 and 400 2.8 though did get rewritten drive algorythms however when they got IS.
the 70-200 2.8 is a very snappy lens in terms of AF though and its servo tracking is buttery smooth. Its about as fast as it comes though technically the 300 2.8 has the edge but were talking thousandths of a second most likley
p.4 #3 · F2.8 vs. F4? Is it really THAT signifigant?
I shot a dress rehearsal of "Arsenic and old lace" directed by Gregory Cohen at the Long Beach Playhouse. Six out of the eight shots in this gallery at f2.8. Not only did the fast lens make the shots possible, but the short DOF is great.
p.4 #4 · F2.8 vs. F4? Is it really THAT signifigant?
Big difference, would not be able to do much of the work I do with the f4, Examples on my website at www.secondfocus.com, scroll over "webBooks" and click on "Arnold 2005" for one gallery all with the 2.8. Also the main page photo is witht he 2.8.
Why should I give more credence to your hearsay assertion quoting Chuck than to pixel perfect's hearsay statement quoting other sources? Have you tested this on a 70-200IS? I have.
Therefore, I prefer to rely on that which my own 1st hand testing/eyes have shown/seen. And that is that the 1.4x TC degrades the AF speed on my 70-200 IS about 35%, and not 50-67% as you and kecohen claim.
Maybe this 50-67% AF performance hit applies more to non-optimized lenses but the impact of the TC on an AF optimized 70-200IS could be less? But there's now way it is half as fast with the TC.
BTW kecohen here is how I came up with the 400mm:
200 mm lens + 1.4x TC + 1.6x crop factor (1.4x+1.6x = 2x). Not exactly rocket science!!!!!
Jul 27, 2005 at 01:48 AM
The Big Bad Offline • • • • [X]
p.4 #6 · F2.8 vs. F4? Is it really THAT signifigant?
Phil Bonner wrote:
Why should I give more credence to your hearsay assertion quoting Chuck than to pixel perfect's hearsay statement quoting other sources?
Well, Chuck is the manager of the Camera Divison Technical Information Dept of Canon U.S so he probably has a pretty good idea about tecnical figures for Canon gear wouldnt you agree ?
Then theres you, some random poster on an internet forum who conducted the ever so scientific test of having your son running at you and seeing how many additional steps it took for the AF to lock on ?
seems a pretty good reason to give more credence to the offical word of Canon IMHO
p.4 #7 · F2.8 vs. F4? Is it really THAT signifigant?
Big Bad,
You can drop Chucks name and impressive titles until the cows come home. But I am still more impressed by the results from my own testing seen with my own eyes: There is not a 50% los of focus speed.
For example, with all due respect, your ignorance of the optimization of the 70-200IS auto focus causes me to even further question the credibility of your opinion of how the 1.4xTC impacts the auto focus performance of the optimized 70-200IS.
The question remains: Does a 1.4x TC cause less degradation of AF perfomance on such an AF optimized lens as the 70-200IS as it does on non-optimized lenses? What makes you so smug and sure that this supposed 50-67% percent decline in AF performance affects all the lenses equally accross the board? If so then why the 50-67% variation?
Untill I see a link with old Chuck baby specifically stating that the 70-200IS is equally affected by the 50-67% degradation of performance, I am inclined to believe my own first hand testing and eye witness results rather then second hand hearsay; Especially hearsay from someone who has just demonstrated a lack of familiarity with an upgrade of the AF system which gives it an advantage over most other lenses.
Jul 27, 2005 at 03:02 AM
The Big Bad Offline • • • • [X]
p.4 #8 · F2.8 vs. F4? Is it really THAT signifigant?
Phil, this whole thing is just absurd, what are you 10 ?
Your being stubborn and childish in the face of proven facts direct from Canon.
I dont understand the AF system on EF mount lens ? Well excuse me for saying so but Ive owned every single DSRL Canon has produced with the exception of the 1Ds mkII, and have pretty much owned every EF mount lens Canon has produced as well. I am not the worlds foremost expert, but I think anyone will will confirm that I do in fact know what I am talking about.
How do I have a lack of famaliarity when I specifically mentioned that the AF drive algorythms have been updated, "speed optimized" as you like to call it ?
Seriously your making a complete fool of yourself. How can you possibly think your kid running equates to a scientific test ? Its simply not a quantifiable measurement to come up with a percentage from.
At what distance do you even test from ? You do of course realize that the AF performance slows down in relation to the subject to camera distance as it decreases right ?
The very fact that you think you know more than Chuck Westfall is downright funny. Perhaps you should take over his postion at Canon since you apparently are more knowledgeable on the topic than him.
perhaps you can rewrite all of Canons tech notes as well since you obviously have a more scientific testing method than Canon has access to
Jul 27, 2005 at 03:24 AM
The Big Bad Offline • • • • [X]
p.4 #9 · F2.8 vs. F4? Is it really THAT signifigant?
and here is "old chuck baby" as you like to call him, email address:
p.4 #12 · F2.8 vs. F4? Is it really THAT signifigant?
Big Bad:
All I said was that I am more inclined to believe my own first hand, eye witness results than your 2nd hand hearsay quotes; Especially when you had just demonstrated your ignorance of the optimized AF characteristics of the lens in question.
It seems logical to me that said AF optimization could give the 70-2000IS an edge over non-optimized lenses that suffer the 50-67% degradation of AF perfromance. In other words, the optimized 70-200IS may only suffer about a 35% degradation. At least that's about as much as I see with mine.
It seems you have a very shaky memory. First you misquote me by claiming I said I knew more than Chuck. I never said that. Secondly, you have stated a blatant untruth when you denied that you said that you did not know of the AF optimization of the 70-200IS.
Maybe this will help jog your memory:
Big Bad said:
I personally have not heard of a special AF optimization on the 70-200 either.
Let's just stick with the main question rather than spewing all of your unpleasant personal infectum.
The question is: Does the 1.4xTC affect the AF perfomance of the optimized 70-200 IS in an equal manner as it affects other non-optimized lenses?
BTW, the more you attempt to "baffle me with BS" the more you just lose credibility in my book. For goodness sake, if you don't know the answer, just say so!
Edited by Phil Bonner on Jul 27, 2005 at 01:20 AM GMT
p.4 #13 · F2.8 vs. F4? Is it really THAT signifigant?
You can say that again! Ha! Hey I've been visiting the RG site reading Chuck;'s stuff for years - at least until about 6 months ago when he stopped posting there. Where have you been?
Big bad said twice:
and here is "old chuck baby" as you like to call him, email address:
p.4 #14 · F2.8 vs. F4? Is it really THAT signifigant?
Come on guys, this topic was on the difference between a 2.8 and 4.0 f stop, not on the effect of a TC on the AF of the 70-200. You both made your points more than enough. Let's get back to the original topic.
And BTW yes, to me the difference was worth it. If it wasn't just for the difference in speed and DOF, it was for IS and the wonderful bokeh as a result of the circular aperture mechanism and the position of diaphragm and lens elements that makes this lens so unique. But if you are happy with your current 70-200/4.0 I would think twice to upgrade. I mean it is a lot of money and you can buy other glass to complement your setup for it. Rent the lens for some days if possible and see if you notice the difference and feel if you like to spend the extra money on the lens. If so, you won't regret it, it is a wonderful piece of glass, really.
Peter
Edited by Peter de Weerdt on Jul 27, 2005 at 11:55 AM GMT (Reason: typo)
p.4 #15 · F2.8 vs. F4? Is it really THAT signifigant?
In my experience, I've never been able to shoot at an aperture higher than f/2.8 indoors. Arena's are well lit, but often get you to ISO 1600 at 2.8 and 1/500th, which is the limit for my 1D.
Sure with the 20D you can shoot 3200, but in those situations, you're also sttretching the limit of the light. I've had to shoot at ISO 3200 and f2.0, so f/4 just wont cut it.
That being siad, I like a lot of the f4 lenses by canon. The 70-200 is a sweet little piece of glass, as is the 300 f/4 non IS.
I want to try out a Sigma 100-300 f/4 if I ever need to have a backup rig for daytime stuff.
But as far as indoors, 2.8 isn't a luxury, it's a necessity.
In daytime, the extra stop helps freeze action even more, and really drops that background with some great blur.
p.4 #16 · F2.8 vs. F4? Is it really THAT signifigant?
I'm very grateful for this thread as I'm going thru the same agonising process myself. Thanks to all who have posted with their thoughts.
I think the difficulty is that all three have their advantages - normally the most expensive one is best, and from the brief test of the IS version I did the results were fabulous (yummy bokeh and amazing resolving power) but I guess I worry about its weight as an all-day travel lens. But on the other hand it just seemed so usable - and that may be the clincher. I tried the other two lenses and never seemed to get quite the magical results that I did with the IS - maybe my hands are just too wobbly ;-)
I do remember reading somewhere about how the 70-200 IS lenses were prone to be unreliable - has this improved or was it just a few rogue lenses?
Jul 27, 2005 at 10:32 AM
Lars Johnsson Offline • • • • • • • Upload & Sell: Off
p.4 #17 · F2.8 vs. F4? Is it really THAT signifigant?
About the speed with extenders. There is nothing in the documentation that came with the extenders saying how many % the sped is dropping. At least not in both of mine extenders. It only say it will slow down.
But if you have the EF lens workIII book. Page 226 is about extenders and it say.
1,4X II the speed drops by one half. 2X II the speed drops by 1/4
p.4 #18 · F2.8 vs. F4? Is it really THAT signifigant?
Hopefully this doesn't stray too far from the initial topic, but something I'm curious about is the difference in autofocus accuracy (on the 20D) between a F2.8 and a F4 lens. My understanding (perhaps incorrect) is that the center cross autofocus sensor is accurate to +/-1 DOF for a F4 lens but +/-1/3 DOF for lenses F2.8 or faster.
After a month or two of agonizing over what lens to buy (and reading every lens review I could find), I've decided to get a 17-40 F4 as a first lens for a new 20D. My question is, what practical difference will I see due to the lower accuracy AF? Will I find myself manually correcting the focus more often than I would with a F2.8? Because the DOF is greater with a F4, is precision less important?
Thanks for any advice and comments.
Jul 27, 2005 at 12:38 PM
The Big Bad Offline • • • • [X]
p.4 #19 · F2.8 vs. F4? Is it really THAT signifigant?
Lars Johnsson wrote:
But if you have the EF lens workIII book. Page 226 is about extenders and it say.
1,4X II the speed drops by one half. 2X II the speed drops by 1/4
Im sure phil will say thats just mored hearsay because it doesnt match his running kid steps test.
The only way anyone can have more "proof" than going off of Canon's own provided figures is to go into the actual firmware of the camera and figure out how much its programed to slow it down. When the camera "sees" a TC is attached, it changes its AF behavior in a certain manner to slow the AF and allow the additional precision. We all know that it seems, but why one of us simply cant accept Canon's figure is beyond me.
Phil, if you realize and admit that Canon was correct when they said the TC slows down AF, why can you not also accept that Canon's figures are correct ?
Chuck Westfall and now also Lenswork III arre both not good enough for you ? How many sources does it take ?
p.4 #20 · F2.8 vs. F4? Is it really THAT signifigant?
I do not doubt that the documentation says that the AF speed drops by 1/2. But, since this has not been my experience with my 70-200IS, in regualr shooting and in my informal testing, then I still question if this applies to those lenses with the AF optimization.
For all we know this may be a theoretical number (just like their MTF Charts) because as my AF testing has demonstrated, I do not lose half of the AF performance of my 70-200IS when I use the 1.4xTC.
