100-400 IS vs 70-200 IS and 2x-II TC (full size shots)

DavidP #28649 · Sep 6, 2001

OK, I'm posting the full shots now from my 3 test shots (see previous thread for details).

The differeces here are:

1) I "normalized" the shots by setting the background to white and the lettering to black by using the black-point and white-point eye-droppers in Levels in PS.

2) I applied some USM to the shots: Amount=100%, Radius = 0.5, and Threshhold = 0. Since we almost always apply sharpening to the D-30 pics, I felt this to be appropriate.

This shot is with the 70-200 (non-IS) and the 2x-II TC (at 200mm x 2)

http://www.cox-internet.com/dpennybaker/70-200_X2.jpg

This shot is with the 70-200 IS (non turned on) and the 2x-II TC (at 200mmx2)

http://www.cox-internet.com/dpennybaker/70-200_IS_X2.jpg

This shot is with the 100-400 IS (not turned on) at approximately 400mm

http://www.cox-internet.com/dpennybaker/100-400.jpg

WARNING: These files are about 1.5 MB each. They were saved as Size-12 JPGs in PS.

CAVEAT: The overwhelming concensus thus far is that the 100-400 IS shot isn't valid. There was probably some camera shake (from the tripod collar not being completely tightened). I plan on reshooting this shot, hopefully tomorrow night.

Until then, those with fast connections and who like to analyze these types of test shots are encouraged to take a peek, and offer up your analysis.

Of primary interest to me:

1) Comparison of the 70-200 IS vs. the non-IS model

2) Incorporating the data from the edges -- does this change your conclusions (esp. pertaining to the 100-400 comparison).

Until tomorrow, enjoy!

And thanks for all the input.--The Unofficial Photographer of The Wilkinsons
http://thewilkinsons.xroad.com

Fred Miranda · Sep 7, 2001

DavidP,

To my eyes, the 70-200IS version is slightly less sharp than the "non-IS" version.

If you bring the files into Photoshop and zoom both at 200%, there is a noticeable difference in sharpness.

There is some motion blur on the 100-400IS shot. The lens is moving from right to left.

Even if there was no motion, I believe the "old" and "new" 70-200L would be pretty comparable if not better than the 100-400IS at 400mm f/5.6
Could you re-shoot the 100-400IS shot again?

Best,
FRED

DavidP said:
OK, I'm posting the full shots now from my 3 test shots (see
previous thread for details).

The differeces here are:

1) I "normalized" the shots by setting the background to white and
the lettering to black by using the black-point and white-point
eye-droppers in Levels in PS.

2) I applied some USM to the shots: Amount=100%, Radius = 0.5, and
Threshhold = 0. Since we almost always apply sharpening to the
D-30 pics, I felt this to be appropriate.

This shot is with the 70-200 (non-IS) and the 2x-II TC (at 200mm x 2)

http://www.cox-internet.com/dpennybaker/70-200_X2.jpg

This shot is with the 70-200 IS (non turned on) and the 2x-II TC
(at 200mmx2)

http://www.cox-internet.com/dpennybaker/70-200_IS_X2.jpg

This shot is with the 100-400 IS (not turned on) at approximately
400mm

http://www.cox-internet.com/dpennybaker/100-400.jpg

WARNING: These files are about 1.5 MB each. They were saved as
Size-12 JPGs in PS.

CAVEAT: The overwhelming concensus thus far is that the 100-400 IS
shot isn't valid. There was probably some camera shake (from the
tripod collar not being completely tightened). I plan on
reshooting this shot, hopefully tomorrow night.

Until then, those with fast connections and who like to analyze
these types of test shots are encouraged to take a peek, and offer
up your analysis.

Of primary interest to me:

1) Comparison of the 70-200 IS vs. the non-IS model

2) Incorporating the data from the edges -- does this change your
conclusions (esp. pertaining to the 100-400 comparison).

Until tomorrow, enjoy!

And thanks for all the input.
--
The Unofficial Photographer of The Wilkinsons
http://thewilkinsons.xroad.com

Pondria · Sep 7, 2001

FRED said:
DavidP,
To my eyes, the 70-200IS version is slightly less sharp than the
"non-IS" version.
If you bring the files into Photoshop and zoom both at 200%, there
is a noticeable difference in sharpness.

I agree. And I have been expecting this. In order to incorporate IS feature, basically a lens or a group of lens has to be floating to be controlled by piezo actuators ( or similar stuff ). No free lunch !

Member said:
There is some motion blur on the 100-400IS shot. The lens is moving
from right to left.
Even if there was no motion, I believe the "old" and "new" 70-200L
would be pretty comparable if not better than the 100-400IS at
400mm f/5.6
Could you re-shoot the 100-400IS shot again?

Me too, please ;-)

Member said:
Best,
FRED

DavidP said:

OK, I'm posting the full shots now from my 3 test shots (see
previous thread for details).

The differeces here are:

1) I "normalized" the shots by setting the background to white and
the lettering to black by using the black-point and white-point
eye-droppers in Levels in PS.

2) I applied some USM to the shots: Amount=100%, Radius = 0.5, and
Threshhold = 0. Since we almost always apply sharpening to the
D-30 pics, I felt this to be appropriate.

This shot is with the 70-200 (non-IS) and the 2x-II TC (at 200mm x 2)

http://www.cox-internet.com/dpennybaker/70-200_X2.jpg

This shot is with the 70-200 IS (non turned on) and the 2x-II TC
(at 200mmx2)

http://www.cox-internet.com/dpennybaker/70-200_IS_X2.jpg

This shot is with the 100-400 IS (not turned on) at approximately
400mm

http://www.cox-internet.com/dpennybaker/100-400.jpg

WARNING: These files are about 1.5 MB each. They were saved as
Size-12 JPGs in PS.

CAVEAT: The overwhelming concensus thus far is that the 100-400 IS
shot isn't valid. There was probably some camera shake (from the
tripod collar not being completely tightened). I plan on
reshooting this shot, hopefully tomorrow night.

Until then, those with fast connections and who like to analyze
these types of test shots are encouraged to take a peek, and offer
up your analysis.

Of primary interest to me:

1) Comparison of the 70-200 IS vs. the non-IS model

2) Incorporating the data from the edges -- does this change your
conclusions (esp. pertaining to the 100-400 comparison).

Until tomorrow, enjoy!

And thanks for all the input.
--
The Unofficial Photographer of The Wilkinsons
http://thewilkinsons.xroad.com

Click to expand...

Jim C. · Sep 7, 2001

Fred,

I didn't zoom in w/photoshop, but I did notice that the 70x200L (non IS) also looked a tad sharper. (upper left corner test pattern)

When I looked at the upper right corner, I had a hard time calling which was sharper.
I wondered if you could confirm what these tired old eyes are seeing?
(19" at 1280x960)
Thanks,

Jim C.

FRED said:
DavidP said:

OK, I'm posting the full shots now from my 3 test shots (see
previous thread for details).

The differeces here are:

1) I "normalized" the shots by setting the background to white and
the lettering to black by using the black-point and white-point
eye-droppers in Levels in PS.

2) I applied some USM to the shots: Amount=100%, Radius = 0.5, and
Threshhold = 0. Since we almost always apply sharpening to the
D-30 pics, I felt this to be appropriate.

This shot is with the 70-200 (non-IS) and the 2x-II TC (at 200mm x 2)

http://www.cox-internet.com/dpennybaker/70-200_X2.jpg

This shot is with the 70-200 IS (non turned on) and the 2x-II TC
(at 200mmx2)

http://www.cox-internet.com/dpennybaker/70-200_IS_X2.jpg

This shot is with the 100-400 IS (not turned on) at approximately
400mm

http://www.cox-internet.com/dpennybaker/100-400.jpg

WARNING: These files are about 1.5 MB each. They were saved as
Size-12 JPGs in PS.

CAVEAT: The overwhelming concensus thus far is that the 100-400 IS
shot isn't valid. There was probably some camera shake (from the
tripod collar not being completely tightened). I plan on
reshooting this shot, hopefully tomorrow night.

Until then, those with fast connections and who like to analyze
these types of test shots are encouraged to take a peek, and offer
up your analysis.

Of primary interest to me:

1) Comparison of the 70-200 IS vs. the non-IS model

2) Incorporating the data from the edges -- does this change your
conclusions (esp. pertaining to the 100-400 comparison).

Until tomorrow, enjoy!

And thanks for all the input.
--
The Unofficial Photographer of The Wilkinsons
http://thewilkinsons.xroad.com

Click to expand...

Shabok · Sep 7, 2001

DavidP said:
1) Comparison of the 70-200 IS vs. the non-IS model

2) Incorporating the data from the edges -- does this change your
conclusions (esp. pertaining to the 100-400 comparison).

IMHO:

The 70-200 non-IS is NOTICEABLY sharper than the IS version, particularly in the lower left corner. Actually, I find this corner to be quite blurred...

The 70-700 non-IS seems also slightly sharper than the 100-400, but that will maybe change if you shoot it again...

My current sharpness result (from best to worst) :
1) 70-200 and 2x-II TC
2) 100-400 IS
3) 70-200 IS and 2x-II TC

Shabok

Asif · Sep 7, 2001

Do you have access to the old 2x extender? I'd be interested to see the same picture from a 70-200/2.8 non-IS + old 2x @ F5.6

Shabok said:
DavidP said:

1) Comparison of the 70-200 IS vs. the non-IS model

2) Incorporating the data from the edges -- does this change your
conclusions (esp. pertaining to the 100-400 comparison).

Click to expand...

IMHO:

The 70-200 non-IS is NOTICEABLY sharper than the IS version,
particularly in the lower left corner. Actually, I find this corner
to be quite blurred...

The 70-700 non-IS seems also slightly sharper than the 100-400, but
that will maybe change if you shoot it again...

My current sharpness result (from best to worst) :
1) 70-200 and 2x-II TC
2) 100-400 IS
3) 70-200 IS and 2x-II TC

Shabok

DavidP #28649 · Sep 7, 2001

Sorry, but I don't.

Asif said:
Do you have access to the old 2x extender? I'd be interested to see
the same picture from a 70-200/2.8 non-IS + old 2x @ F5.6

DavidP #28649 · Sep 7, 2001

Member said:
To my eyes, the 70-200IS version is slightly less sharp than the
"non-IS" version.

That was my initial impression. Not enough to dissuade me from keeping the IS version, though.

Member said:
There is some motion blur on the 100-400IS shot. The lens is moving
from right to left. Could you re-shoot the 100-400IS shot again?

That's planned for tonight. I may re-shoot the 70-200 IS again, too - I don't like the fact that it's rotated.

JCDoss · Sep 7, 2001

David,

Just to start things off on the right foot, I'd like you (and everyone else) to know that I just ate a hearty bowl of Chef Boy-Ar-Dee's ravioli (which for peculiar reasons, is served only once per two weeks in my house). Now that I have a stomach full of southern Italian cuisine, I'm ready to make some comments on your newest tests.

First, it took an unusually long time to download these files. Yes, I was on 56K, but it seems it was slower than it should have been.

I think in all three photos, the left side seemed less sharp than the right side. I wonder about whether the cameras were exactly perpendicular... a shallow DOF may be to blame if the test charts were shot at an oblique angle.

Overall, I rank the three in the following order:
1) 70-200 non-IS + 2xTC
2) 70-200IS + 2xTC
3) 100-400IS

IMO, the 70-200IS + 2xTC ranks second by the hair on its chinny-chin-chin.

Strange, I think the 70-200IS+2xTC outperforms the 100-400IS in the upper half of the frame (letters and lines are sharper), but the opposite is true in the lower half.

What is crystal clear is that the non-IS version outperforms both lenses ON A TRIPOD! Also, I presume that both IS-equipped lenses had IS off?

The next test would be to take some handheld photos with all three lenses with IS on. (This is the whole point of IS anyways, isn't it?)

JCDoss

JCDoss · Sep 7, 2001

David,

The bottom line that I failed to mention in the last post is that for quality long telephoto work, you'll probably need to get a good fast prime. With the D30's 1.6x multiplier, a 300/2.8 becomes a 480, and the 400/4DO (if it ever re-appears) would become a 640/4. Put a 1.4xTC on either lens, and you've got 674/4 and 896/5.6 (!) with minimal quality loss at your disposal.

The only problem is the exorbitant cost of these large aperture primes. Last I checked, the 300/2.8 was over $4K, and the 400/4 DO didn't exist yet (at least commercially).

JCDoss

Karl Guttag · Sep 7, 2001

It could be a lot of things, but it is most likely that David was slightly off in alignment. The DOF at 400mmF5.6 taken from about 15 feet away (since David was only using an 8x10'ish size target) is only about one-third of an inch. I doubt David had that kind of accuracy in his set up. Even the sightest error in centering, leveling or the like will build up to a lot more than 1/3 of an inch over 15 feet.

Karl

Shabok said:
DavidP said:

1) Comparison of the 70-200 IS vs. the non-IS model

2) Incorporating the data from the edges -- does this change your
conclusions (esp. pertaining to the 100-400 comparison).

Click to expand...

IMHO:

The 70-200 non-IS is NOTICEABLY sharper than the IS version,
particularly in the lower left corner. Actually, I find this corner
to be quite blurred...

The 70-700 non-IS seems also slightly sharper than the 100-400, but
that will maybe change if you shoot it again...

My current sharpness result (from best to worst) :
1) 70-200 and 2x-II TC
2) 100-400 IS
3) 70-200 IS and 2x-II TC

Shabok

DavidP #28649 · Sep 7, 2001

JCDoss said:
I think in all three photos, the left side seemed less sharp than
the right side. I wonder about whether the cameras were exactly
perpendicular... a shallow DOF may be to blame if the test charts
were shot at an oblique angle.

This may very well be the case. I really didn't make a concerted effort to get the lens perfectly perpendicular (just by eyeball). Why? I'm sort of at a loss on how to really do that.

Does anybody have any suggestions in this regard?

Whatever the method, I don't think it'll be easy. It's hard enough as it is to get the whole thing adjusted so that the target is centered and not rotated. (In fact, the target IS rotated in two of the shots). I'm beginning to see the limitations of cheap ballheads.

Member said:
IMO, the 70-200IS + 2xTC ranks second by the hair on its
chinny-chin-chin.

Is it neck-and-neck with the 70-200 non-IS, or the 100-400 IS ?

Member said:
Also, I presume that both IS-equipped lenses
had IS off?

Yes. I had posted shots with the new lens with IS both on and off yesterday -- nobody seemed to think there were any differences -- at least nobody commented on it.

Interestinly, the manual STILL says not to use IS on the tripod. Even though I thought this new lens was supposed to work with IS on a tripod.

Member said:
The next test would be to take some handheld photos with all three
lenses with IS on. (This is the whole point of IS anyways, isn't
it?)

I tried hand-holding the 70-200/2.8 IS + 2x-II yesterday. Like the 100-400 IS, it's a very heavy/bulky combination. I have a hard time holding it completely still, even with the IS, without SOME form of support. (Better start pumping some iron instead of eating lasagna, I guess).

I think I'll do some testing with my Sima VideoProp, both with and without the IS.

DavidP #28649 · Sep 7, 2001

Karlg said:
It could be a lot of things, but it is most likely that David was
slightly off in alignment. The DOF at 400mmF5.6 taken from about
15 feet away (since David was only using an 8x10'ish size target)
is only about one-third of an inch. I doubt David had that kind of
accuracy in his set up. Even the sightest error in centering,
leveling or the like will build up to a lot more than 1/3 of an
inch over 15 feet.

I agree. Any suggestions on how to get it perpendicular?

PS -- I'm not sure my wife is gonna let me build a jig in the hallway for this purpose.

Edmund Tan · Sep 7, 2001

DavidP said:
This may very well be the case. I really didn't make a concerted
effort to get the lens perfectly perpendicular (just by eyeball).
Why? I'm sort of at a loss on how to really do that.

Does anybody have any suggestions in this regard?

Ermm.. sorry to intrude like that but ever thought of using a bubble meter?

DavidP #28649 · Sep 7, 2001

I wish I had a 400/5.6 to test for a comparison. That lens is reportedly quite a bit sharper than the 100-400 IS at 400mm. Assuming for the moment that I sell the 100-400 IS, I could certainly use those funds to get the 400/5.6

If the 400/4 DO IS comes out at a price under $3,000 it'll be very tempting to me. That would still allow me to use a 1.4x adapter and have AF on the D-30.

Trading the 100-400 IS in for the 400/5.6 is certainly an option, but I'd miss the IS. So, at this point, I'm not sure if I'll sell the 100-400 IS or not. It's still a great lens (with the 1.4x adapter) for shots at the zoo.

I wonder how the 70-200/2.8 IS would work with BOTH adapters? The manual says not to do this, however. It mentions optical quality degradation (though I wonder if it'd be any worse than the 100-400 IS with a 1.4x), but also that metering would be a problem. I'm not sure why this would be. At any rate, I could always use Exposure Compensation to counteract that effect.

If I can run a few tests of the 100-400 IS and 1.4x vs. the 70-200/2.8 IS with the 2x-II and 1.4x, and conclude that the difference in quality is still fairly negligible, then I may go ahead and sell the 100-400 IS at this time.

Member said:
The bottom line that I failed to mention in the last post is that
for quality long telephoto work, you'll probably need to get a good
fast prime. With the D30's 1.6x multiplier, a 300/2.8 becomes a
480, and the 400/4DO (if it ever re-appears) would become a 640/4.
Put a 1.4xTC on either lens, and you've got 674/4 and 896/5.6 (!)
with minimal quality loss at your disposal.

The only problem is the exorbitant cost of these large aperture
primes. Last I checked, the 300/2.8 was over $4K, and the 400/4 DO
didn't exist yet (at least commercially).

DavidP #28649 · Sep 7, 2001

I'm not sure I understand how a bubble-meter is going to help me align myself in a panning sense. (left side of target vs. right side of target)

Edmund Tan said:
Ermm.. sorry to intrude like that but ever thought of using a
bubble meter?

Edmund Tan · Sep 7, 2001

DavidP said:
I'm not sure I understand how a bubble-meter is going to help me
align myself in a panning sense. (left side of target vs. right
side of target)

Opps.. pardon me.. did not see that part.. guess its time for me to go rest.. way past bedtime now..

Karl Guttag · Sep 7, 2001

The easiest thing I can think of to do is to get a bigger target.

DoF increases by the square of the distance, so if you double the target size and double the distance to fill the veiwfinder, then the DoF goes up by 4X.

While I have not done this (I did do it in just 2 dimensions to make sure I was centered), in theory if you had 4 strings or measuring tapes each attached to the corners of the target and ran them to the center of the lens so that the length of the strings/tapes are equal (say clamp them together at the point they converge), then you are lined up (at the point of a pyramid). This may be a bit hard to do in practice unless you go to the trouble to mount the target on something solid with holes for the 4 strings/tapes.

Karl

DavidP said:
Karlg said:

It could be a lot of things, but it is most likely that David was
slightly off in alignment. The DOF at 400mmF5.6 taken from about
15 feet away (since David was only using an 8x10'ish size target)
is only about one-third of an inch. I doubt David had that kind of
accuracy in his set up. Even the sightest error in centering,
leveling or the like will build up to a lot more than 1/3 of an
inch over 15 feet.

Click to expand...

I agree. Any suggestions on how to get it perpendicular?

PS -- I'm not sure my wife is gonna let me build a jig in the
hallway for this purpose.

Karl Guttag · Sep 7, 2001

FRED wrote:

FRED said:
DavidP,
There is some motion blur on the 100-400IS shot. The lens is moving
from right to left.
Even if there was no motion, I believe the "old" and "new" 70-200L
would be pretty comparable if not better than the 100-400IS at
400mm f/5.6
Could you re-shoot the 100-400IS shot again?
Best,
FRED

Fred, in the related topic David started yesturday, I too commented that it looked like the camera had panned slightly. But David went out for lasagna and had ravioli last night instead of re-shooting the shot

.

Shooting a good lens test is tougher than one might at first think with these long focal lengths. And it takes a while to get it all the things done to get them sorted and posted.

Karl

Shabok · Sep 7, 2001

You're right, it's the most logical explanation of this phenomenon.

Now, looking only at the center part of the graph, my overall placing remains unchanged

Shabok

Karlg said:
Karl

Shabok said:

DavidP said:

1) Comparison of the 70-200 IS vs. the non-IS model

2) Incorporating the data from the edges -- does this change your
conclusions (esp. pertaining to the 100-400 comparison).

Click to expand...

IMHO:

The 70-200 non-IS is NOTICEABLY sharper than the IS version,
particularly in the lower left corner. Actually, I find this corner
to be quite blurred...

The 70-700 non-IS seems also slightly sharper than the 100-400, but
that will maybe change if you shoot it again...

My current sharpness result (from best to worst) :
1) 70-200 and 2x-II TC
2) 100-400 IS
3) 70-200 IS and 2x-II TC

Shabok

Click to expand...

100-400 IS vs 70-200 IS and 2x-II TC (full size shots)

Forum Pro

Fred Miranda

Guest

Senior Member

Senior Member

Senior Member

Well-known member

Forum Pro

Forum Pro

Member

Member

Senior Member

Forum Pro

Forum Pro

Active member

Forum Pro

Forum Pro

Active member

Senior Member

Senior Member

Senior Member

Keyboard shortcuts