Do EFS Lenses Really Give Better Image Quality on 1.6 Bodies?

This is a serious question as I've never owned an efs body since they created

that designation...As some of you have that have read in the past know I have a

1D Mk II (and a MK 1 before that) but I did carry a digital rebel (the first

one) to europe to cut wieght when i went in 2004. I had much of the glass I

have now so those were the lenses I took and the pics were pretty good. Is

there a reason that they can't just make good lenses that fit both

cameras...I'm sure no pro would have been upset using a 17 - 55 2.8 (or even

better a 16 - 55 2.8) or something like that....you get my drift

 

The real question is this - (i understand the light going straight to a smaller

sensor theory) Do the efs lenses really produce a better image than a "normal"

eos lens with the 1.6 cameras...If not shouldn't canon just abandon that concept

for the most part (I guess the only time it really makes sense is with the 10 -

22 type lenses)

 

thanks

JC

I think if you look at the fact that the EF-S lenses are made for a reduced imaging circle, you will see that they can make a cheaper lens by using smaller optics. Thats where money is saved. I guess a full frame 17-55 2.8 which performs as good as the EF would be much more expensive to build

The same small size advantage applies by pushing the back element about 4mm nearer to the sensor, but I thing it may be a much smaller contributor than the reduced imaging circle

Except for the 10-22, Based on the optical performance of the few efs lens available, I dont believe that the image quality has benifited from the back protuding efs design.

 

Keeps me thinking if it's really neccessary to make the back of the efs 18-55, 17-85 , 17-55 and 60mm macro , protude.

so it seems that it comes to cash...which is what i thought. Both forcing people that start in EFS to buy new lenses if they decide to go full frame and by playing a marketing game of the "made for digital" lens...

 

could it possibly be more expensive to build a 16 - 55/70 ish lens than the price of buying both the $1400 16 - 35 and the 1150ish 24 - 70?

I wonder...something that we'll never know but I would love to...

Actually, if you want to see what the EF-S 17-55 lens is like in full frame, just look through an EF 28-90mm lens. Its characteristics seem virtually the same--build, crispness, etc. etc. Clearly this was the model in terms of price, quality for the EF-S kit lens.

"so it seems that it comes to cash...which is what i thought. Both forcing people that start in EFS to buy new lenses if they decide to go full frame and by playing a marketing game of the "made for digital" lens..."

 

No one is "forced" to buy anything. EF-S owners who see it in their interest to buy EF-S lenses can do so, just as EF-S owners who choose to buy full-frame lenses can do the same.

 

As far as "marketing games," does Canon explicitly say that the EF-S lenses give better image quality with digital sensors, or do they use the "optimized for digital cameras" phrase primarily as shorthand for "focal lengths that are optimized for crop bodies"? I'm not saying you're wrong; I just would like someone to point me to Canon's exact language if Canon is indeed promising better image quality. (I suppose I should poke around Canon's website.)

 

Finally, wrt to "shouldn't canon just abandon the EF-S concept for the most part (I guess the only time it really makes sense is with the 10-22 type lenses)"...

 

7 of the 8 EF-S lenses start out at 18mm or shorter, and all of these <18mm lenses would be substantially larger if they had to cover full-frame. The only exception to the <18mm rule is the 60mm macro, which I believe is far smaller and lighter than any other 1:1-capable combination made by Canon.

 

Bottom line: the size/weight/price advantages for EF-S lenses don't seem to be merely marketing spin - but again, photographers who expect to go full-frame can opt for full-frame capable lenses if they see no advantages to EF-S (most users of the EF-S 10-22 and 17-55 IS seem to have no regrets!).

The EF-S is simply smaller, lighter and cheaper to build. That is a plus for many folks who are drawn to the smalle and light DSLRs that use these lens. The extra weight does not bother me so I don't buy the EF-S lens for my 30D. I plan to keep my lens around for a long time and I hope to have FF oneday. They are no better in quality and pretty much all are cheaper in build (due to the market they are aimed at). Infact many say that it is better to use normal lens on croped sensor because you use the best part of the lens, and as we all know the edges are alway the worst part of any lens.

 

If you don't like EF-S don't buy EF-S. I don't!

 

Jason

BTW, at B&H you can only find 9 Canon EF-S Lens, as compared to 110 of the "normal lens". Of course out of that 110, they offer two of each lens (USA or Import) so there is only about 50. But still 50 normal lens compared to 5 of the EF-S lens. Thats 10 times as many.

 

When you look at it that way, what is the problem?

 

Jason

Your assumption is that everybody will eventually upgrade to FF, which might or might not be true. There is an interesting article by Bob Atkins speculating about the future of the APS-C format (http://www.photo.net/oped/bobatkins/full_frame.html), and basically wondering if it's just a short term phenomenon at the forming stage of the new digital photography era, or if that format will survive in the medium and long term. The article is copyrighted 2004, so it is getting kinda dated. Bob's last sentence is that: "In 5 years time I guess we'll find out", and that's probably were he went furthest from the truth, because of his 5 years only 2 are left, and I don't think in two years time the future of APS-C will have been decided quite yet.

 

Bob's main hypothesis is that APS-C is basically just a cost saving exercise compared to FF, and as soon as somebody can offer FF at a reasonable price, APS-C will inevitably be abandoned. IMHO the reason that APS-C could be more persistent than we initially assumed might be that it has something to offer over FF other than cost saving, which a lot of photographers want, and that's more reach in tele lenses. Whenever you have a tele lens that you can mount on a FF body and a APS-C body, you will have more reach with the APS-C body. That is enormously attractive to all the soccer moms and backyard bird shooters of this world, and these folk buy a lot of cameras!

 

Personally, I bought a Rebel XT almost 2 years ago and I have no regrets. Until that time I was shooting a Minolta XD-7 and had never seen the need to upgrade to AF, but the instant feedback, possibility to shoot an almost unlimited number of frames, and opportunities of post processing in the digital darkroom without a scanning step that a DSLR had to offer were very attractive to me. I could not have made the switch to a FF body due to cost consideration, but I could swing the Rebel XT with the kit lens and a 420ex. I have since sold my entire Minolta collection with lenses from 16mm to 400mm and acquired several EOS lenses, incuding EFS 10-22, EFS 17-85 IS, EF 70-210 USM, and EF 100mm macro. My rationale for buying the two EFS lenses is that I want wide angle capabilities now, and I want to have a mid range zoom that extends in the wide angle as well as the tele range, to avoid an ridiculous number of lens swaps during any shoot. I bought the lenses used or refurbished at good prices. If I sold them now, I am sure I could realize a profit, and if I sell them in a few years time I hope I will not loose too much. At any rate, the XT ($700), 10-22 ($600) and 17-85 ($400) come in at about $1700. That's well below the 5D price, and with a 5D I would want something like a 16-35 and a 24-105, and that's an extra $2k. APS-C has made DSLRs affordable and accessible for a lot of people. I think that's great!

Good EF lenses fit both types of cameras. EFS lenses do not perform any better than other

lenses on crop sensor cameras - it is just that they _only_ work on crop bodies.

EF-S (reduced lens to focal plane distance) probably is capable of producing some marginal improvement in image quality for highly retrofocal lenses, i.e. lenses with a very short focal length.

 

However there have been enough examples now from Sigma, Tokina and Tamron of non EF-S short focal length lenses to say that the improvent is small and that lenses with a "normal" back focus distance can be made which work perfectly well.

 

The reduced image circle (as opposed to the short back focus) is an advantage in that you can make lenses which might be more difficult to make and more expensive if they were required to cover a full frame image. However the cost saving seems small, given the current pricing of EF-S lenses vs. full frame lenses. I suspect that a 17-55/2.8L full frame lens could be produced, but it's easier (and probably cheaper) to do it with the reduced image circle of EF-S.

 

I suspect the APS-C (1.5x/1.6x) format is here to stay for the forseeable future. I expect to see more full frame cameras and I expect to see models at lower prices, but I don't think they will soon (if ever) displace the consumer APS-C DSLRs. They may be analagous to 35mm and medium format cameras, which co-existed for decades, with most amateurs shooting 35mm (now shooting APS-C) and many professionals shooting medium format (now shooting full frame digital).

It was my understanding that it was considerably more difficult and expensive to produce larger sensors, price up a hasselblad digital back if you need proof!. As this technology is refined fullframe cameras will reduce in price I have no doubt.

I'm not sure if this is what you are asking, but my 16-35/2.8 (on 5D and 20D)looked better than my 10-22EFs on my 20D at equivalent and actual FL and apertures.

 

http://www.nagelhome.com/phototest/10-22%20vs%2016-35%2022mm.jpg

 

m

That was 22mm

 

here is 16mm

 

 

http://www.nagelhome.com/phototest/10-22%20vs%2016-35%2016mm.jpg

 

m

The other factor to consider is that APS-C DSLR are smaller and lighter.

 

Even at the same price point (APS-C and FF), a lot of non-Pros may very well opt for APS-C for the combined weight and telephoto reach advantage.

that sorta was part of my point...but what i'm saying is this...

it's 1000 for the 17 - 55 2.8 and it's 1400 for the 16 - 35 and 1100 for the 24 - 70...if they could produce a full frame that was less than 2g's the covered 16 - 70 it would be a great lens...that just about everyone that shoots both (which is becoming increasingly common) would buy...

Imagine that lens for wedding shooters that have 5D's and 40D's as back ups...

apparently it's impossible but the fact is that none of us work for canon and we don't know how much they are really making per lens. One thing I learned in school is that Henry Ford made it possible for everyone to own a car by introducing mass production...Once the initial production r/d costs are absorbed it costs them very little more to produce the 1400 16 - 35 than the 680 17 - 40 - these lenses are not hand crafted...they are computer generated assembly line produced...

"If they could produce a full frame [lens] that was less than 2g's the covered 16-70 it would be a great lens...that just about everyone that shoots both (which is becoming increasingly common) would buy... Imagine that lens for wedding shooters that have 5D's and 40D's as back ups... apparently it's impossible but the fact is that none of us work for canon and we don't know how much they are really making per lens."

 

Generally speaking, if none of the dozen major companies that produce lenses for SLRs make a particular focal-length lens, it's not because of some nefarious "marketing game" or conspiracy against the buyer: there simply isn't enough demand for it at the price point it would take to justify it. If the 16-70, FF, sub-$2k lens you want were easy to build, Tamron, Sigma, Tokina etc. would be all over it. The fact that none of them are should tell you something.

 

Frankly, it appears that it isn't very easy to make a high-IQ 2.8 zoom that covers FF and goes from 16-35, let alone twice that far. I doubt that Canon, Nikon, and the others have the capability for producing a 16-70 lens cheaply and are just withholding it so that they can sell more lenses in less-impressive focal lengths! (Yes, hyperzooms are out there - albeit not from ultrawide FF to short telephoto - but at least with where technology is now, photographers who are concerned with high IQ tend to shun them.)

 

We'd all like to have a pocketable 10-1000 f/1.4 image-stabilized full-frame zoom lens with no distortion that sells for $200. The fact that no one makes such a lens shouldn't be interpreted to mean that the camera- and lens companies are conspiring to fleece us gullible consumers.

fairly certain i didn't ask for a 10 - 1000 1.4...what i said is that I don't believe that it is that much harder to produce any lens over an other...

once the line is set up the costs go down as more are sold...

Joseph Carey wrote "...Once the initial production r/d costs are absorbed it costs them

very little more to produce the 1400 16 - 35 than the 680 17 - 40 - these lenses are not

hand crafted...they are computer generated assembly line produced..."

 

But what an assembly line! see http://www.canon.com/camera-museum/tech/l_plant/

f_index.html You can see why the 500 f/4 L is a) as good as it is and b) why it costs what

it does, there is a large amount of skilled manual assembly and testing along the way, and

that's after the extensive molding grinding and polishing of how many elements per

"lens"?

I don't think you completely understand how complex these are. The optical formula and

manufacturing tolerances for those two lenses are not likely to be as close as you seem to

be assuming (f/2.8 vs f/4 size, complexity , grinding, polishing, tolerances etc...) The

production volume for these is not like the microprocessor and RAM chip economics for

example, and then there are all the people of the forums who complain about the slightest

issue with one of these lenses. Sure, Canon makes their money but it's supply and

demand and their quality is generally pretty high. Plus what Ralph said :)

ok...fair enough. But do you really believe it costs canon 550 to install IS in the 70 - 200 or that it really costs twice as much to essentially add one stop to the 17 - 40 to get down to the 16 - 35

 

In any event...I was just saying that I wish they would stop with the efs. the aforementioned 17 - 40 is a great example of a lens that they could market to the 1.6 crowd very easily if they would add 10 mm to it.

a 17 - 50 f4 L lens would sell lots...truth is I sell lots of it now to rebel and 30/40D folks... and then the 70 - 300 which was one of the lenses i complain about. Not due to sharpness but more the quality of the build - it is always kinda loose and just doesn't feel as solid as some other companies lenses

Tests of my EF-S 10-22mm vs. my 16-35mm on a 40D at f/8 show them to be almost

identical. I think the 16-35mm is a little sharper overall, but you really gotta pixel peep to

see it.

 

But the 10-22mm is a little more than half the weight of the 16-35mm. After several

years of lugging a 1D2N/16-35mm combo around the 40D/10-22mm combo is a

welcome relief to my back!

 

-Mike

Are there any third party wide-angle primes optimized for smaller sensors?

No. The only thing that maybe comes close to that is the Sigma 30/1.4, but that's a "normal" lens, not a wideangle.

 

As far as I know there are no primes under 30mm designed only to cover an APS-C sensor.

 

This is probably because full frame lenses already exist and the market for such lenses would be limited. You can buy a EF14/2.8 if you've got enough money and if you want one badly enough. Though an EF-S version might be a bit smaller and lighter, I doubt it would be much cheaper and I doubt they'd sell very many, so economy of scale wouldn't come into play and lots of development cost would have to be supported by each lens sold.

 

There may be some prime wideangle 4/3 format lenses. I don't really follow that market.

Pentax seem to be the brand to get if you want (wide-angle) primes for small sensors:

 

http://www.photo.net/equipment/pentax/#wide_angle

In general, looking at MTF charts and test results, the EF-S lenses do often have better centre sharpness. On the other hand the corner sharpness falls off in the same way that it does with an EF lens on full frame.

 

Generally I don't think the IQ advantages and disadvantages are all that compelling in the real world.

 

The main advantage with the EF-S lenses seems to be being slightly smaller and lighter in weight and having more convenient focal length ranges for the format.