Unadulterated Junk

[soupdragon]

If it's got EF-S written on it or, it,s got a plastic mount, don't touch it with a barge pole.
What were Canon thinking when they designed these lenses?
Is there any point in having an 8mp camera if the bloody lens only resolves to about 4mp.
I have tried all of the EF-S lens range because I can't afford L glass and taken them all back to the shop.

Dear Canon get it together or I am buying a bloody nikon.

[Naytwan]

My 17-85 IS has served me well.

[Skip Souza]

My 17-85 IS has served me well.
Glad to hear that. It is on my wish list.

[tim]

What were Canon thinking when they designed these lenses?

I would think "lets give people a cheap starter lens that's not too bad, since anyone who knows anything will buy their own lenses anyway".

[Andy_T]

Soupdragon,

I am sorry to hear about your experience, it certainly is not similar to mine.
While not stellar, my EF-S 18-55 is plenty sharp at f/8.0 and certainly does give me the full 8 MP I paid for ... which is more than I really expected for a 100$ lens.

Also - as you mentioned plastic lens mounts in general - the other lens with that feature, the EF 50/1.8 certainly does not have the characteristics you describe.

I have not tested the EF-S 17-85 and the EF-S 10-22 lenses, but from all I heard especially the 10-22 should be 'an L lens without the label'. What were your results with these EF-S lenses?

While it certainly is possible that you could have received a dud lens (or even more of them), let me respectfully suggest that you verify if your tests have been conducted in a controlled and proper way.

Best regards,
Andy

[GyRob]

yes some of Canon's lenses are utter rubbish my 75 -300 for instance.
Rob.

[kraterz]

The EF-S 60mm macro is supposed to be a cracker. However I agree on the 18-55, it's quite a pathetic lens.

[Jesper]

yes some of Canon's lenses are utter rubbish my 75 -300 for instance.
Rob.But that's not an EF-S lens with a plastic lens mount that soupdragon was talking about. All lens manufacturers have good and bad lenses.

[Huckaback Photo]

Soupdragon
You do sound rather pi..... off with EFS lenses, obviously you have either had bad copies or possibly some other reason the results are not great.
Presumably you have some examples done with EFS lenses, so please put some shots on here to let us all judge for ourselves.
Now I don't actually own a EFS lens, however I have seen many amazing shots done with said lenses & at times it is hard to tell from norm or indeed L versions as of course there are even bad copies of these.

Now check this out to see what I did with a borrowed EFS lens,
it just blew me away, I normally use all L series on my 1D mk 2 & D 60 bodies.

http://www.photography-on-the.net/forum/showthread.php?t=70062&highlight=huckaback+photo


Martin (Huckaback Photo)

[Huckaback Photo]

Tim
I think what you said above is quite true & indeed a lot of people realise this.

The research done on here by various members of this forum is great.
we can often point others toward top quallity kit and lenses if thats needed, the big crunch of course is the cost as not all have £/$1000 + per each lens, available to spend.
So the EFS is there to maybe fill a slot in the market, however the canon quallity should also be there, otherwise like our friend above he will not buy the Canon brand again.

One final thought. Does anyone on here know if Lenses are graded in any way, are some sold off real cheap maybe cause they were refubished or sent back to canon to be sorted.
only a thought but sure I've seen refurbished stock on ebay for sale ?

Martin

[lancea]

I'm confused soupdragon that you say you've tried all the EF-S lenses, but can't afford L. There's certainly 2 L lenses that cost less. And like Andythaler I'm pretty pleased with my 18-55. I imagine there's plenty of amateurs could happily stick with that as their only lens - and produce good shots. The shots taken with any of the (non kit) EF-S series all look very good to me, and the various tests support that. I'm sure everyone else is curious if there's more to your story. Why so damning in your judgement?

[Tom W]

Lens frustration is the symptom; 17-40L is the cure.

[kenyc]

If it's got EF-S written on it or, it,s got a plastic mount, don't touch it with a barge pole.
What were Canon thinking when they designed these lenses?
Is there any point in having an 8mp camera if the bloody lens only resolves to about 4mp.
I have tried all of the EF-S lens range because I can't afford L glass and taken them all back to the shop.

Dear Canon get it together or I am buying a bloody nikon.

I'd be particularly interested in how you arrived at the resolution of the entire family of efs lenses being 4MP. Do you have measurements or images that detail this?

KAC

[Jack W.]

My 17-85IS and 10-22 both work great.
My 18-55, though certainly a cheapie, can still produce surprisingly good results.
No complaints here with any of them.

[photog]

Go ahead and by Nikon if you want.

I have the EF-S 17-85 and recently bought the EF-S 60 Macro. The 17-85 is an excellent lens...Suprisingly it beat the pants of of the two 24-70Ls I tried and had to return for serious focus issues.

My EF-S 60 is absolutely stunning. It's even slightly better contrast and color than my EF 100mm Macro and sharper than my EF 50/1.4. I use the 60 as my walkaround and carry my 70-200 2.8 L IS for zoom.

[soupdragon]

OK!

Firstly I had no intention whatsoever of upsetting EF-S lens owners.
Secondly "all EF-S lenses" was an exaggeration, the ones I have used and taken back are, 17-85, 10-22 and still own but don't use the kit lens.
Yes the 50 1.8 does have a plastic mount and is not an EF-S but is still crap.
The basis for my judgement is this (if you can be bothered to read any more of this drivel)
The 50 1.8 never produced a sharp image at any aperture all the time I used it, and I really did try to get on with it.
The EF-S models I owned/used did not perform reliably in as much as at some apertures/focal lengths they were good but at others bad (I really don't like trusting to luck)
My points for comparison are the 50 1.4 and my personal fave the 100 2.8 macro.
Yes I know they are not zooms but they perform in a manner I expect and produce reliable well colour balanced sharp images.
For those unable to understand sarcasm the 8mp performing like a 4mp is how I perceive the degradation in quality.

"Any more Inane questions?"

[sdommin]

I have the EF-S 17-85 "kit lens", and I've been pleasantly surprised by it. For example, this was handheld with my 20D (external flash)...
http://home.att.net/~sdommin/z_portrait.jpg

This is a 200% unsharpened, unprocessed crop...
http://home.att.net/~sdommin/z_locket.jpg

That is certainly not "unadultered junk", and I would put its sharpness against any L-lens, any day (and I do own a few L-lenses).

[jfrancho]

Could be a nut loose behind the viewfinder?

j/k

[soupdragon]

If that's un-sharpened and un-processed, how come (on my monitor) the models top is a different colour?

[sdommin]

If that's un-sharpened and un-processed, how come (on my monitor) the models top is a different colour?

The crop is indeed unprocessed. The full picture was given a quick auto-levels just to make it more presentable so that you could get an overall view of the area that is shown in the crop (I would also straighten the photo, too). Sorry for the confusion, but I didn't think that this has anything to do with the central point of this thread.

[skyphix]

Funny, my 50 1.8 has a metal mount... and I definitely wouldnt call it crap.

Im poor, all I can afford for a wide angle is the 18-55 (kit lens) and it does the job.

Also, the Ef-S line hasnt been out for very long, and although Im not happy about its inability to be used on anything other than the 300/350/20d's, its cheap enough for me to not worry about breaking and it takes fine enough photos for me to deal with it.

The 50 1.8 (and 650 it was attached to)
http://non.skyphix.com/2005/Things/Small/650Kit4.jpg

Test photo taken with my 50 1.8
http://non.skyphix.com/2005/Things/Small/50mmTestShot28.jpg

[Andy_T]

Skyphix ... he's referring to the 50/1.8 MKII (the version you can buy new today).

Yours (and mine) is the older MKI version. This one sells for more money used now than the new one if you buy it new ... not because the image quality was better, but because it is supposedly more durable.

Still, the assertion that you 'can not get a single sharp image at any aperture' with the new 50/1.8 II is contrary to the experience of 99% of forum members who own this lens and know what they are supposed to do with it. Bad lens again ???

Best regards,
Andy

[mdr]

Even jumping ship and buying Nikon is not going to help you. Nikon lenses in the same price class will give you the same quality (or the lack of it) and inconsistency. Whether you go for top of the range Canon L or Nikon AF-S for the consistent quality you're after, you'll have to pay more whichever make you choose.

[soupdragon]

SDOMMIN:

Whilst your picture in it's own right is commendable, just lightly pressing the auto levels button in photoshop does a fair amount of processing.
I could be wrong on this but, I think it adjusts colour balance, contrast & to some degree, sharpness.

[Southswede]

I would think "lets give people a cheap starter lens that's not too bad, since anyone who knows anything will buy their own lenses anyway".


I was thinking the same thing.
The kit lens, I got with the DRebel served me well.

[sdommin]

Whilst your picture in it's own right is commendable, just lightly pressing the auto levels button in photoshop does a fair amount of processing.
I could be wrong on this but, I think it adjusts colour balance, contrast & to some degree, sharpness.

Soup, I don't think we're connecting here. Forget about the overall portrait picture. That one WAS "processed" - put it out of your mind. Look at the close-up view of the locket. That one is unprocessed - nothing was done to it except that I blew it up to 200%, saved it as a JPG, and put it on the forum post. When you wrote that EF-S lenses were "unadultered junk", I wanted to show you that my 17-85 was not junk, and that it was a pretty darn good lens. Don't worry about color balance, look at the sharpness and detail.

Since you've tried a bunch of EF-S lenses and found them to be lacking, why don't you post some examples of the poor results that you obtained (surely you must have saved something from your experiments).

[JZaun]

Here is a different take on this issue. I had a 18-55 kit lens. It was sharper than my 17-85mm lens :? I did a side by side test with the 18-55 and the 17-85 and sent the 17-85 IS back. Then I ordered the 17-40 (L) and did a side by side with the 18-55 and the (L) just barely beat it out :? I kept the 17-40 (L) but in NO way were any of these lens JUNK. All took good pictures, it was just how picky I wanted to be that made the final decision. I could have been happy with any of them if I had just printed the pics and had I not done a side by side test and pixle peeped each. My 17-40 cost 5 times as much as the 18-55 but I really do not feel its 5 times better :) Maybe just a little better!!

Just my 2 cts

JZ

[rdenney]

If it's got EF-S written on it or, it,s got a plastic mount, don't touch it with a barge pole.
What were Canon thinking when they designed these lenses?
Is there any point in having an 8mp camera if the bloody lens only resolves to about 4mp.
I have tried all of the EF-S lens range because I can't afford L glass and taken them all back to the shop.

Dear Canon get it together or I am buying a bloody nikon.

Do you think Nikon's bottom-of-the-line lenses are any better?

Canon makes three grades of lenses: Consumer, Prosumer, and L-series. The consumer lenses are very light, very plastic, and often have a plastic mount. Some of them are good optical performers, but many only perform acceptably when used very carefully. But they meet the requirements of those who will never make a print bigger than 4x6. The price for the consumer lenses ranges up to about $300 or so. I have a 35-70 zoom that falls into this category (but not into my camera bag), and the 50/1.8 does also. I keep mine around, though, for when I'm taking pictures in conditions I don't want to subject my better stuff to.

Then, there are the mid-grade Prosumer lenses. These lenses are priced from as low as about $225 to maybe half a kilobuck. They often (but not always) have USM focusing. Some have image stabilization. They tend not to be as fast as their L-series counterparts, but some of them are optically quite as good. The 50/1.4 is an example, the 20-35 USM is another, and the EF-S 10-22 is still another (though it breaks out of the price bracket). Many of these are first-class lenses that don't get the L designation because they don't use special glass and some aren't build for commerical duty cycles.

The L-series lenses are also made in two grades. There are the entry-level L-series lenses such as the 17-40/4L and the 70-200/4L, but most are intended for commercial use and are built for high duty cycles. They are optically excellent and in many cases superior within their category, of course. They are also pricey.

So, it's just like my grandmother used to say, you get what you pay for.

(If you want high quality and low price, you have to give up convenience and buy the high-end products of obsolete systems and adapt them to your camera.)

Rick "who wonders why people complain that cheap stuff is cheap" Denney

[skyphix]

Skyphix ... he's referring to the 50/1.8 MKII (the version you can buy new today).

Yours (and mine) is the older MKI version. This one sells for more money used now than the new one if you buy it new ... not because the image quality was better, but because it is supposedly more durable.

Still, the assertion that you 'can not get a single sharp image at any aperture' with the new 50/1.8 II is contrary to the experience of 99% of forum members who own this lens and know what they are supposed to do with it. Bad lens again ???

Best regards,
Andy


:D Andy, I realize that, it just wasn't specified and being a vicim of several broad generalizations (young males cant drive, young people just drive slow imports, all imports are slow, young people dont have any experience in the field I work in, etc) I tend to throw caution to the wind and common sense out the door. Just giving him a hard time :D

I can see where he's coming from - if y ou're used to L lenses and then you go to a lens with a plastic mount and plastic body it feels a bit weak, however, the quality of images (at least from the kit lens, per its price, and what I've seen of the 10-22) aren't so terrible that they are unusable...

[sp00g3]

Personally, not to sound like an a-hole, but if soupdragon is having issues with a 50 1.8 mk1 or 2, then the issue is with the photographer and not the equipment. I have never taken a bad (unsharp) photo with this lense.

[ed2day]

Give Soupdragon the benefit of the doubt. I don't get 6MP worth out of my 18-55 kit lens. The resolution is clearly limited by lens(it has a focus issue). My plastic 50mm is great. I think however that it's wrong to generalize, esp. EF-S. Plastic mounts mean cheaper lenses which probably means more lax specs/QC. But too many people have had great experiences with EF-S lenses. I suspect whatever problems Soup is having won't be fully cured with full frame metal mounts or even "L" lenses.

[sodakred]

I too am surprised by Soup's adamance. I have two EF-S lenses, the 17-85 and the 10-22. I have the 20D, so the 10-22 is an absolute kick to use...the composition possibilites seem endless even when surrounded by mundane subjects. I constantly am amazed at how sharp it is w/out much distortion when used properly.

The 17-85 is my walk-around lens. I'm constantly craving the 24-70L as the lens I would prefer to have in this range, but the more I use the 17-85, the more I enjoy it and find that it really takes excellent photos. The USM is superb, the AF is fast and the IS allows you to hand hold at very low speeds. Simply put, I still can find enough (or any) "drawbacks" to justify spending the extra cash. I also compared it to the 17-40L, which was very slightly better, but the range of the 17-85 trumped it in my book.

I also have a 70-200 2.8L and I derive as much enjoyment from the results of the 17-85 EF-S as I do from the 70-200, accounting for different applications of course.

If I ever upgrade to full frame and lose mounting compatability, I'll keep or sell the 17-85 with my 20D. I would not like to part with the 10-22 under any circumstances (maybe I just need to stick with the 20D).

Finally, I have the 50 1.4 as my low light "bar lens." It's tack sharp, but I struggle to get good, pleasing photos with it probably more than the 17-85.

As said in either The-Digital-Picture site or Luminous Landscapes, the EF-S lenses fit "hand and glove" with the 20D...a great platform.

I have no experience with the 18-55 kit job or the 60mm macro.

I'm trying not to be overly sensitive as the owner of 2 EF-S lenses. I can afford better (i.e. the 24-70L)...just haven't found the "why" yet.

[DocFrankenstein]

My sig :rolleyes:

[shiato storm]

50mm mkII f/1.8...great lens. probably the best lens canon do for the money!!
ok so its got a plastic mount but if you look after it certainly you will have a good load of pictures out of it. as a prime its quality is (pretty much) assured, and at 1.8 you can't really complain. cost me all of £54... (thats about $100 to US, $150 CA I think)...bit of a bargain since its more than that from honk kong on ebay, then add shipping expenses...
yes its no USM but it does the job well.

! got a 24-70 L new for about 2/3 retail price. if you know where to look top quality can be cheap too! in photography you can either get lucky or spend tons of money. I manage both :)
whether thats a good thing or not I'll let you know as I sell my work to pay the rent! :o

[Naytwan]

I just took this with my 50 1.8 MKII about 10 minutes ago.

Shutter Speed 80
F 1.8
ISO 100
Handheld
20D
Fill Flash

http://img.photobucket.com/albums/v425/naytwan/625747c6.jpg
Straight from the camera, no processing.

http://img.photobucket.com/albums/v425/naytwan/e5893963.jpg
100% crop.

I think I paid about $60 plus shipping US for the lens from BH.

[Skip Souza]

Naytwan, that is the cutest little nose I have seen.

I don't know why all the uproar about the supposed lack of quality of plastic mounts. My Glock is mostly plastic and my Colt is virtually all metal. I'll bet my life on the Glock any day over the Colt.

[kenyc]

OK!

Firstly I had no intention whatsoever of upsetting EF-S lens owners.
Secondly "all EF-S lenses" was an exaggeration, the ones I have used and taken back are, 17-85, 10-22 and still own but don't use the kit lens.
Yes the 50 1.8 does have a plastic mount and is not an EF-S but is still crap.
The basis for my judgement is this (if you can be bothered to read any more of this drivel)
The 50 1.8 never produced a sharp image at any aperture all the time I used it, and I really did try to get on with it.
The EF-S models I owned/used did not perform reliably in as much as at some apertures/focal lengths they were good but at others bad (I really don't like trusting to luck)
My points for comparison are the 50 1.4 and my personal fave the 100 2.8 macro.
Yes I know they are not zooms but they perform in a manner I expect and produce reliable well colour balanced sharp images.
For those unable to understand sarcasm the 8mp performing like a 4mp is how I perceive the degradation in quality.

"Any more Inane questions?"

Oh, I see it's JUST YOUR OPINION, you don't really have any measurements and it's just exaggeration, I see. I thought that's what smilies were for. :)

KAC

[kenyc]

The crop is indeed unprocessed. The full picture was given a quick auto-levels just to make it more presentable so that you could get an overall view of the area that is shown in the crop (I would also straighten the photo, too). Sorry for the confusion, but I didn't think that this has anything to do with the central point of this thread.

I think the central point of the thread is that he was just trolling for trouble.

KAC

[soupdragon]

No, the central point of this thread is, if you settle for what Canon define as acceptable you'll can it for the rest of us.
I don't need to resort to extensive lens testing to be able to gauge the difference in image quality between various optics, I can see the difference on my monitor.
More to the point, I notice the 350D default camera capture settings enhance the image by upping in camera sharpness, saturation & contrast.
Why should this be so do you think? I guess it is to give that WOW factor to the out of box first use.
Whilst I have no method of quantifying the performance of lenses it seems to me to be rather pointless having a sensor with more pixels than a lens can resolve to.

I had hoped this thread would have led to healthy debate as opposed to a slanging match.
The saddest part for me though is, some of the members of this forum are producing truly outstanding images but in some cases appear out of focus or soft. I am prepared to wager in many instances this will be down to budget optics.

[kenyc]

No, the central point of this thread is, if you settle for what Canon define as acceptable you'll can it for the rest of us.
I don't need to resort to extensive lens testing to be able to gauge the difference in image quality between various optics, I can see the difference on my monitor.
More to the point, I notice the 350D default camera capture settings enhance the image by upping in camera sharpness, saturation & contrast.
Why should this be so do you think? I guess it is to give that WOW factor to the out of box first use.
Whilst I have no method of quantifying the performance of lenses it seems to me to be rather pointless having a sensor with more pixels than a lens can resolve to.

I had hoped this thread would have led to healthy debate as opposed to a slanging match.
The saddest part for me though is, some of the members of this forum are producing truly outstanding images but in some cases appear out of focus or soft. I am prepared to wager in many instances this will be down to budget optics.

Well, you clam that lens can't resolve what the sensor can, but yet you have no way to prove that. Again it's just you opinion and other (many others) don't seem to agree.

Maybe it's your monitor. :)

KAC

[soupdragon]

Well it's not just my opinion is it!

For instance:
The canon 135 f2.0 L is reputed to be the sharpest lens money can buy.
Now look at one of the budget priced bottom end cheapy zooms.
Do you think on the same camera, the image clarity, light fall off and colour balance will be the same?

No!??

Why is that do you suppose?

Well I'll tell you.

If the 135 produces a sharper image than a cheap zoom then the sensor must, by logic, be able to resolve at a higher level. Thereby annulling the need for a high resolution cmos sensor if the lens cannot resolve detail to a level exceding that of the sensor.

[Croasdail]

Not to put fuel on the fire here, but it would be nice if there was some standard naming methodology here Canon could use. They have the venerable red L line which we all associate with superiour optics and build. But below that there is little distinction between lenses. Price is always a good indication - but not always the most accurate. It would be nice if Canon would officially have a high-end consumer line as a specific group. EF-S is not a product line, rather a form factor and within EF-S ther will be a wide range of lenses and levels of Quality.

As to the issue of resolving power - there should be some objective tests that can truely rate a lenses abilty. Sharpness is being very subjectively rated here. What may be sharp to me may not be sharp at all to you. I know I have seen plenty of shots showing off a particular lenses sharpness that I have thought to be soft. And sharpness without deep saturated color or contrast is less then appealing. And lastly, sharpness is not always desirable depending on the supbject mater - Portraits for one. As to who is right or wrong here... no one is. I belong to a sports car club and am happy to get any car on a track... conversly my boss has been racing for some years and what is acceptable to me is quit boring to him. So IF-S so far may not be cutting it for the thread owner - doesn't mean it isn't the perfect lens for someone else. I ramble way to long... trying to avoid work... cheers!

[Andy_T]

Objective measures of sharpness would be resolving power in lines per square mm or MTF charts, I guess.

Still, the question is how helpful these really are in judging the quality of a lens.

Best regards,
Andy

[Tom W]

Ultimate resolution isn't based solely on the weakest link - it is based on the resolving power of the entire system, including both the lens and the sensor. You can have two lenses that both resolve better than the sensor, one by a small margin and one by a larger margin. The difference will still be noticeable even though the sensor is outresolved by both lenses.

[soupdragon]

Tom:

You are talking beyond what I understand about sensors.

How, if you excede the resolving power of a sensor with two lens types, one marginally and one to a greater degree, does this affect line pairs per millimeter?
Given you can calculate the lppm for any known sensor, how then can you fit another line in between the rows of pixels?

[Croasdail]

couldn't there be a simple test the identifies the smallist line that a camera can resolve to at a specifed image size and distance. At some point though it jumps into the world of highly specialized lenses... and there must be some threshold where it is equal to what the average persons eye can recongize - and beyond that any difference is lost? Getting way off track now.... and still avoiding monday morning work - though have checked email.

[Tom W]

Can't give an exact formula, but the general trend follows the "sum of the reciprocals" rule where:

1/T=1/L+1/S where T=total resolution, L=lens resolution, and S=sensor resolution. There's much more to it, and I have read on this a few times. I'll try to find a link to the actual articles that explain this a lot better than I can.

[Andy_T]

Well, until Tom comes up with more detailed links, this would be some kind of example: RDenney's test of some of his MF lenses (http://www.rickdenney.com/lens_testing.htm).

He uses a modification of the USAF1951 test (http://www.efg2.com/Lab/ImageProcessing/TestTargets/)chart. In this test, you photograph different sized printouts of this chart with a given lens and see which is the smallest one where the lines are still clearly discernible ... that would be the 'extinct' resolution of a lens/camera combination.

Best regards,
Andy

[Tom W]

couldn't there be a simple test the identifies the smallist line that a camera can resolve to at a specifed image size and distance. At some point though it jumps into the world of highly specialized lenses... and there must be some threshold where it is equal to what the average persons eye can recongize - and beyond that any difference is lost? Getting way off track now.... and still avoiding monday morning work - though have checked email.

Depends on how large you want your final image. On a 4X6 inch print, nearly any lens will give good resolution. 8X10 starts showing more differences, and larger prints will show even more. 100% crops on the computer screen (the equivalent of something like a 30X45 inch image, more or less depending on screen resolution and pixel count) will show defects that will never reveal themselves in typical sized prints.

The theory of the circle-of-confusion is based on what the average person's eye can see on a somewhat standardized print size at a standard viewing distance. There's variances in what is a standard print size, what is a standard viewing distance, and what is considered to be visible to the average person's eye (Leica's circle of confusion might well be smaller and more stringent than that used by Canon). The greater the degree of enlargement needed from sensor (or film) to final image, the smaller the circle-of-confusion must be.

[soupdragon]

I have put some more thought into this and the best I can come up with is this:

An 8mp sensor (350D) is 22.2mm on its longest side.
It has 3456 pixels along this side.
So if you divide the number of pixels by the overall length you get 155.67 pixels per mm.
So divide this by 2 to get pixel pairs and you get 77.83.
This assumes that the pixels are effectively in a square pattern (I really don't know, so if any of the above is not true that blows my theory out of the water)

In summary, it appears, given my limited understanding of CMOS sensors, that an EOS 350 D can produce images with a maximum resolution of 78 lppm.

Thus any lens incapable of resolving to this or above is not fully utilising your sensor.

But I could be wrong!

Feel free to keep the hate mail coming.

[Tom W]

Can't say it isn't true, but it is only true to the extent that the contrasty boundaries used to measure resolution line up with the boundaries of the individual photosites. If a boundary falls dead-center on a photosite, that individual photosite will see the boundary as the sum of the light energy. If it is half-black and half-white, then it will be a mid-gray response from that individual pixel. The pixel immediately to the dark side will see pure black, and the white side, pure white (I'm leaving color out as that will complicate things considerably). The fine line will be 2 pixels wide rather than 1. Thus, the resolution in that circumstance will be much poorer. So, your real-world resolution will not be as great as the theoretical limits.

Incidentally, film has this same issue, but instead of photo-detecting grains being arranged in a grid-like pattern, they are randomly placed on the film. And more sensitive film (higher ISO) has larger granules and thusly will not resolve as finely as slower film.

Getting to the lens' input into this situation - most lenses will outresolve most sensors on DSLR cameras, though the corners may go south on some lenses at some apertures.

Norman Koren touches on the system resolution a bit in his series on MTF:

http://www.normankoren.com/Tutorials/MTF.html

Some of it is way to geeky for me, but this part caught my eye:

Resolution of an imaging system (old definition)— Using the assumption that resolution is a frequency where MTF is 10% or less, the resolution r of a system consisting of n components, each of which has an MTF curve similar to those shown below, can be approximated by the equation, 1/r = 1/r1 + 1/r2 + ... + 1/rn (equivalently, r = 1/(1/r1 + 1/r2 + ... + 1/rn )). This equation is adequate as a first order estimate, but not as accurate as multiplying MTF's. [I verified it with a bit of mathematics, assuming a second order MTF rolloff typical of the curves below. It's not sensitive to the MTF percentage that defines r. The approximation, 1/r^2 = 1/r1^2 + 1/r2^2 + ..., is not accurate.]

Note that the he says the the sum of the reciprocals is adequate for estimation, but that the sum of the squared reciprocals (second order) is not. This is drifting into calculus which I haven't seen in 10 years. I cringe each time I see "dx", even if it's just a camera's name.

[soupdragon]

Well judging by the lack of response that's to deep for us to understand!

Many thanks for the details though.

[csnudelman]

The crop is indeed unprocessed. The full picture was given a quick auto-levels just to make it more presentable so that you could get an overall view of the area that is shown in the crop (I would also straighten the photo, too). Sorry for the confusion, but I didn't think that this has anything to do with the central point of this thread.

True, it has nothing to do with the point of this thread. As long as we are discussing things that have nothing to do with this thread...the second picture makes me long for my, now gone, Mamiya RB67.
;)

[Tom W]

Well judging by the lack of response that's to deep for us to understand!

Many thanks for the details though.

I register about 8 on the geek scale of 1-10. This stuff is 10. I skim it, but I'm not about to do calculus again - I'd rather take some pictures once in a while.

Perhaps everyone is off reading the Norman Koren web site now. :)

[Andy_T]

I very selectively read the Norman Koren website ...
only the pages with pictures and not too many equations :lol:

My take on it is ... some lenses are better than others, and with the right tools measurebaters can find out, which are which and tell us normal users about it ... ok with me :wink:

Best regards,
Andy

[kenyc]

Can't say it isn't true, but it is only true to the extent that the contrasty boundaries used to measure resolution line up with the boundaries of the individual photosites. If a boundary falls dead-center on a photosite, that individual photosite will see the boundary as the sum of the light energy. If it is half-black and half-white, then it will be a mid-gray response from that individual pixel. The pixel immediately to the dark side will see pure black, and the white side, pure white (I'm leaving color out as that will complicate things considerably). The fine line will be 2 pixels wide rather than 1. Thus, the resolution in that circumstance will be much poorer. So, your real-world resolution will not be as great as the theoretical limits.

Incidentally, film has this same issue, but instead of photo-detecting grains being arranged in a grid-like pattern, they are randomly placed on the film. And more sensitive film (higher ISO) has larger granules and thusly will not resolve as finely as slower film.

Getting to the lens' input into this situation - most lenses will outresolve most sensors on DSLR cameras, though the corners may go south on some lenses at some apertures.

Norman Koren touches on the system resolution a bit in his series on MTF:

http://www.normankoren.com/Tutorials/MTF.html

Some of it is way to geeky for me, but this part caught my eye:



Note that the he says the the sum of the reciprocals is adequate for estimation, but that the sum of the squared reciprocals (second order) is not. This is drifting into calculus which I haven't seen in 10 years. I cringe each time I see "dx", even if it's just a camera's name.

Tom Thanks for this info. I'll check out the link. I'm pretty geeky myself, having been in the computer industry (in various aspects) all my adult life. :)

KAC

[Todd Jacobsen]

If it's got EF-S written on it or, it,s got a plastic mount, don't touch it with a barge pole.
What were Canon thinking when they designed these lenses?
Is there any point in having an 8mp camera if the bloody lens only resolves to about 4mp.
I have tried all of the EF-S lens range because I can't afford L glass and taken them all back to the shop.

Dear Canon get it together or I am buying a bloody nikon.

Based on your "testing" I'd recommend you go to nikon. Canon just doesn't fit your individual needs.

[Tom W]

Based on your "testing" I'd recommend you go to nikon. Canon just doesn't fit your individual needs.

Well, they do have some 4 megapixel cameras over there on the "dark" side, which should match their lenses. ;)

[rdenney]

I had hoped this thread would have led to healthy debate as opposed to a slanging match.

That's all that's left once you take quantitative measurement out of the equation. What did you expect?

Yes, pixel spacing narrower than about 8 microns is going to expose the optical limitations of photographic lenses. The same was true for film. Lenses vary considerably in their limiting resolution, based on whether they are design to be cheap, light, fast, or to have a wide zoom range. Everything is a compromise, and it's as true for Leica and Nikon as it is for Canon.

We have been trained in the digital photography age to think of the digital sensor as being the limiting element. We calculate how big a print we'll be able to make based on the pixel count, just as an example. But that is because the memory needed to hold large digital images is dear, and we don't want to waste any of it.

A truly faithful (to reality) digital sensor would have pixels so small that the optical faults are the lens are always the limiting factor in potential print size. We should be able to profile the sharpness versus aperture of even the best lenses, if we don't want the digital sensor to limit us. Then, we wouldn't need an anti-aliasing filter, because the moire patterns would be smaller than the lens blur. Until that time, digital can only be described as "good enough", but it will also represent a compromise with reality (including how much storage space we can carry around with us). Thus, rather than complaining about the lenses not being up to the sensor, as you are doing, you should be complaining that the sensor isn't clearly more detailed that all the lenses, all the time.

The solution is to measure performance the way we used to: This lens will let me make a good 8x10 print; that lens is sharp up to 11x14; this other lens gets fuzzy beyond 5x7; no sense using fine-grain film with that lens, or no sense wasting the potential of this lens on Tri-X.

Judging lenses based on uncorrected images on your monitor is like judging your lenses by looking at a color negative through a loupe. You can tell a few things about resolution, but very little about tonality. A digital image is not a transparency, which is designed to be directly viewed at the outset. A digital image is just information to be manipulated into something worth viewing by adjusting color, contrast, tonal placement, and so on. We've always done that, too. The Zone System, to name one 75-year-old methodology, is based on that concept. If a Digital Rebel hot-rods the image to make it look better out of the camera, then that's what is appropriate for all the SLR newbies buying an entry-level DSLR, right? Shoot raw and you won't have to deal with that limitation.

The better Canon lenses are outstanding, as tools. If you are unable to be persuaded of that, then perhaps you should go try Nikon.

Rick "who doesn't blame his equipment without knowing why" Denney

[Croasdail]

I am with you all the way until....

A digital image is just information to be manipulated into something worth viewing by adjusting color, contrast, tonal placement, and so on....

Photoshop and it's like products are great tools - but they should not be a replacement for good fundimental techniques and quality equipment. A less the desiable lens - even if you do shoot RAW - gives you much less to work with then a quality lens will. Every person has their own workflow strategy. The really good people spend very little time photoshoping thier images. I could go on forever on the subject of how people spending hours learning how to make marginal photos look good rather then how to take great pictures to start... but thats another subject.

and to your comment....

The better Canon lenses are outstanding, as tools. If you are unable to be persuaded of that, then perhaps you should go try Nikon.

the man never said that, it was a very specific complaint about ef-s lenses and their build quality. In general there is a huge issue with lens quality from all manufactures - having to have brand new lenses sent in for calibration is rediculas - and you see that a lot - hence all the post of "did I get a good one". So two issues with your statement.. we all know that Nikon's consumer line has the same issues so sending the chap in that direction does nothing- and we should not be so offended when someone takes issue with Canon - they are not god. When someone spends $500 plus for something - they should not have to be asking "if they got a good one" or not.

[Bob_A]

Soup, just wondering, but how much USM do you typically apply to your images in PS?

Bob

[soupdragon]

Croasdail:

Finally someone who actually understood what I was saying.

[soupdragon]

Bob A:

The amount of sharpening varies tremendously.
When photographing people I tend not to use sharpening as much as in some cases it is not flattering for the subject (alternatively I use an EF-S lens to acheive soft focus)
When doing macro I use sharpening more often, mainly because I like to see the hairs on a bees legs for instance.

[Tom W]

I am with you all the way until....



Photoshop and it's like products are great tools - but they should not be a replacement for good fundimental techniques and quality equipment. A less the desiable lens - even if you do shoot RAW - gives you much less to work with then a quality lens will. Every person has their own workflow strategy. The really good people spend very little time photoshoping thier images. I could go on forever on the subject of how people spending hours learning how to make marginal photos look good rather then how to take great pictures to start... but thats another subject.

True, a good image to start with requires less work than a mediocre image. But mediocrity is much more a function of what's behind the viewfinder than what's inside the camera nowdays. Even the inexpensive kit lens has been shown to produce great images in the right hands. Granted, it lacks a lot of features and is rough at the extremes, but it is the best $100 zoom around.



and to your comment....

the man never said that, it was a very specific complaint about ef-s lenses and their build quality. In general there is a huge issue with lens quality from all manufactures - having to have brand new lenses sent in for calibration is rediculas - and you see that a lot - hence all the post of "did I get a good one". So two issues with your statement.. we all know that Nikon's consumer line has the same issues so sending the chap in that direction does nothing- and we should not be so offended when someone takes issue with Canon - they are not god. When someone spends $500 plus for something - they should not have to be asking "if they got a good one" or not.

The issue of "bad" copies is grossly overblown IMHO. I have 10 Canon lenses. I've had a couple of Sigma lenses, and have owned and sold a few other Canon lenses. All work as advertized except for 3. Two of those 3 were bought used, and the third was a brand new Tokina which I sent back. It was obviously bad as it wouldn't focus anywhere near the subject. Of the two used ones, one was a Sigma prime and the other a Canon that had seen better days. The point? No Canon lens that I have bought new has had any calibration problems.

There are people on DPR that claim to have tried 7 copies of a particular lens and never found a good one. I question their method of testing a lens more than anything. The internet is a great place to share ideas. It is also a great place to proliferate bad information. And finding 7 "bad" copies of a particular product - a product which is used and enjoyed by many others - is more than bad luck. It's a statistical anomoly.

So most of the time, when people ask "did I get a good one?", it is not because of any real problem but because of the proliferation of bad information on the 'net that makes one question whether any lens is good, even when it is obviously working fine.

My $.02 for what it's worth....

[soupdragon]

Tom W:

You Said:

"True, a good image to start with requires less work than a mediocre image. But mediocrity is much more a function of what's behind the viewfinder than what's inside the camera nowdays. Even the inexpensive kit lens has been shown to produce great images in the right hands. Granted, it lacks a lot of features and is rough at the extremes, but it is the best $100 zoom around."

Well done, you've just bolstered my original post.

Can you not imagine yourself in that "once in a lifetime, got it framed perfectly & lighting bang on" situation only to have to explain to people, "it's rough because because I had to use my lens at it's extreme wide angle!"
Or, "sorry, it lacks the corner to corner illumination feature"
Or, "you would be able to read the logo on his cuff links if this lens had the make this picture sharp button"
Or, "No, I couldn't afford the right hands to go with this lens"

[ssim]

Soup

You seem to want L quality for the price of a lower budget lens. You get what you pay for. Plain and simple. I have 13 Canon lenses and don't have significant issues with any of them. You have to find their "sweet spot" in the f-stop range, although most of mine are pretty good throughout most of the ranges.

I have seen alot of really great stuff on here from all of the EF-S lenses. If you don't like the performance that you are attaining, then dump it and move on.

[soupdragon]

SSIM:

You are absolutely right on all your points of course, I just have had a frustrating time with new glass lately.
You know how it is when you get a new toy, race home, un-box it, discard the instructions, photograph the old ladies bunch of flowers and get the results into ps as quick as possible.
Part of my problem is having dragged a Hasselblad around with me for many years and I guess I just expect all glass to be as good.
Still, I live in the hope that peasants like myself will one day own L glass.

Anyway, I offer my unreserved apologies if I have upset or offended anyone who uses this forum.

[rdenney]

Photoshop and it's like products are great tools - but they should not be a replacement for good fundimental techniques and quality equipment. A less the desiable lens - even if you do shoot RAW - gives you much less to work with then a quality lens will.

With respect, you are missing my point. Let me try it a different way:

A slide is meant to be put into a projector and projected on the wall. Therefore, it is formulated so that the shadows are dark and the highlights are very bright, with a steep tonal gradation in between for realistic contrast. That's why slide films have an S-shaped characteristic curve. They are designed to spread a narrow slice of scenery values over a wide display range.

Negatives, on the other hand, are not meant to be viewed directly. They are meant to be put in an enlarger, and printed onto paper designed to reverse the negative accurately. Thus, the dark shadows, bright highlights, and steep tonal gradation in between are built into the paper. If the negative did it, it would provide no room for interpretation when it is printed--the S-shaped tone curve would be built in. It's a lot easier to add contrast than it is to remove it. Negatives take a wide range of scenery values and compress them into a narrow range on the film, and you have to spread it back out when you make a print. That's why color negatives have so little apparent contrast, and why black-and-white negatives with lots of contrast are actually rather hard to print.

Digital images are more like negatives. Instead of an S-shaped characteristic curve, they are far more linear, with just a slight toe on the highlight end. The sensor can separate tonal values that extend beyond the range of most display devices, so you have a wide tonal range compressed into a narrow display range. The processor in the camera can do one of two things: It can apply that S-shaped contrast enhancement, or it can display a compressed image. The first strategy is designed to make the image look realistic at the first go, but the second provides more room for the photographer to make his own decisions about tonal placement, as with negatives. Thus, as with negatives, digital images from high-end DSLR's are intended to provide the room to make manipulations.

None of that means that Photoshop is a replacement for good technique. But it is a philosophical difference. With negatives, the print is the target and the image-making process is not complete until the print is made. So, the visual appearance of the negative is not important as long as it supports the making of the desired print. With slides, the slide itself is the endpoint (printing slides used to be something that was done only in desperation--then Cibachrome and its successors came along--but that was relatively recently in the scheme of things; also, I always exposed slides intended for printing quite differently than slides intended for viewing to help control contrast).

If digital images are more like negatives, then image editing would be an integral and expected part of the process to produce the final display image. We may do that automatically in the print-making machine at the local minilab, or we may do it ourselves in Photoshop. But it's still an expected step.

We can set up some cameras to produce direcly viewable images out of the camera, but only by programming the camera to apply the same sort of processing mentioned above. So, we either do that processing in the camera, the computer, or at the lab, but we always do it.

Again, that is unrelated to technique. We may adopt techniques to minimize processing depending on our objectives, but since the output of the sensor is non-viewable raw instructions, we still have to process it at some level and somewhere.

My point in bringing it up is that the expectation that lenses should be matched to sensor performance is ultimately both unrealistic and undesirable. The perfect sensor would be the one that records such fine detail that the blur of even the best lenses obscures aliasing. We just don't have the technology to do that yet. We may find that we don't really need it.

The point of the cheaper EF-S lenses (and I take your point that Soupdragon was talking only about those lenses) is to fill some gaps that would be very difficult to fill with lenses that have full-frame coverage. One of those gaps is a cheap, general-purpose zoom for consumer applications where the 4x6 print is the target display. Another is a good-quality ultra-wide zoom that is reasonably carryable. And so on. Canon's 18-55 fills the first gap, and the 10-22 fills the second.

That a cheap lens performs cheaply should come as no surprise. The size of the final print is still limited by lens quality, just as it has always been, no matter what price point you pursue.

I highly doubt that the cheapie lenses produced by Nikon are any better than the cheapie lenses produced by Canon.

These lenses still do much more than state-of-the-art lenses of just a few decades ago. Back in the 70's, a lens that zoomed from 65% of normal to twice normal would have been rare and expensive, with questionable performance. A lens that zoomed from 36% of normal to 75% of normal would have been unthinkable at any price. Yet the 18-55 and the 10-22 accompllish both these tasks, for less money (in real dollars) to mid-grade primes back in the 70's.

It should be noted that the 8-MP digital Rebel with the 18-55 costs less than the 3-MP fixed-zoom digicam of perhaps five years ago that provided the same basic range of coverage.

Rick "who thinks one should not expect a cheap lens to provide 75 lines/mm resolution, even if the sensor does, but who notes that some cheap lenses actually do when carefully used" Denney

[Croasdail]

Very well put my most eliquint co-contributor. It really all comes down to expectations, and what you feel you should get for your money. Fortunately and unfortunately we have some amazing photo posts here that we would all love to have the capability to reproduce, but we either lacked the proper equipment, or more often then not the experience to achieve those vaulted levels ourselves. And frustration sets in....

And not to date my self too much, but I remember the early days of cibachrome very well rolling that dang tube back and forth across my dorm room only to scratch the print trying to get it out of the tube. Trust me, I am enjoying the electronic experience ever so much more.

It is all about the synchronisity of being in the right location, with the right equipment, and the right skills the make the majic of a wonderful image possible. And the more you practice - the more often those three eliments come together.

I learned a lot from you all on this - thanks.

[soupdragon]

Wow, someone has actually done home Cibachrome, awsome.

By the way, just to reinforce my wingeing (if there is such a word) Amateur Photographer Magazine, this week (well I got mine yesterday) produced an add on about lenses and guess what?... that's right, they really caned the Canon 18-55.
In fact, of all the budget price zooms tested Canon came a strong last.
It really guts me to think I bought a turn-key digital soulution only to find part of it is a door stop.

[rdenney]

And not to date my self too much, but I remember the early days of cibachrome very well rolling that dang tube back and forth across my dorm room only to scratch the print trying to get it out of the tube.

That was a piece of cake compared to making contrast masks. As i said, it's easier to add contrast than to remove it.

I used my 16x20 Cibachrome tube right up to about a dozen years ago, when I put my Ansel Adams phase in the portfolio boxes and stopped doing black and white. It was great for making 16x20 black and white prints in my darkroom that was only big enough for 11x14 trays. It would just fit across the rails of the sink and I could roll it back and forth.

I had given up on doing my own Cibachrome ten years before that. It was just too hard to live with losing all that shadow detail, and preserving it with contrast masks was only really practical with 4x5 transparency film, which I had to process, too. I'm sure I'm not the only one here who has developed E-4 (which dates me horribly) in deep tanks with metal sheet-film hangers. I'd be quite happy to never have to do that again.

To Soupdragon: What were the other lenses in the test in the magazine you mentioned? I once owned a Samyang 18-28 zoom (similar coverage in 24x36 to the 10-22). I used it without shame on my Canon F-1, because it was the only ultrawide zoom I could afford. I thought the images looked fine, and that lens cost $150 (20 years ago). I use that lens now with a Canon P adaptor on a Leica screw-mount Russian rangefinder, as a dual-focal lens, using either the 28mm position on the Russian turret finder or in the wide end with a 16mm Sea-and-Sea viewfinder that leaks to much to still use underwater. As you can tell, I don't mind playing with cheap stuff, as long as I keep my expectations in check. And not all cheap stuff is inexpensive. After all, I own a Kiev 88CM medium format camera with five backs--now that's cheap junque. But it cost more than your Digital Rebel.

Remember that the 18-55's Prime Objective was to hit the sub-$100 price point. That Samyang of mine is pricey by comparison (especially in real dollars). It was not targeting those who have a high image quality standard, but rather those who wanted a wide-to-tele zoom on the Digital Rebel who never make prints bigger than 4x6.

Have you actually made prints with it yet? The images look terrible shown at 100% on a 100-dpi monitor, even when made with state-of-the-art glass. But at 100 dpi, the image on a DRebel will print at 23x35 inches, which is a bit beyond the scope of an APS camera with however many pixels.

Rick "who thinks throwing the 18-55 in the trash at least won't dent your wallet" Denney

[Andy_T]

Amateur Photographer Magazine, this week (well I got mine yesterday) produced an add on about lenses and guess what?... that's right, they really caned the Canon 18-55.


Well, finally a test from a reputable publication :wink: that really teaches all of us who mistakenly thought their photos with the 18-55 looked great.

Thanks for sharing :rolleyes:

Best regards,
Andy

[PaulB]

Just look at the price of the 18-55mm when bought as part of the kit - not as a stand-alone purchase.
The lens does what it is intended to; be a cheap, get you going, OK for small prints, optic. That's before you get seduced into buying more, bigger, better, more expensive lenses.Many people will NEVER need, or want, or lust after, anything else on their 300/350D (or even their 20D) however heretical that may be to the members of this forum!
As for the AP review of 'standard' zooms - the Nikon costs a lot more and is considered a fine optic in its' own right (and has a longer range of 18-70mm), the cheaper Sigma is still more expensive (remember the 18-55mm kit price) and the other Sigma is a lens aimed far higher, both in specification at f2.8 and price. Rather like comparing chalk and cheese.
Have just checked the prices here in the UK and the 18-55mm adds around 8.5% to the price of the 350D body whilst the 18-70mm Nikkor adds 26% to the price of a D70..........
Lets keep things in perspective please.

[kenyc]

Just look at the price of the 18-55mm when bought as part of the kit - not as a stand-alone purchase.
The lens does what it is intended to; be a cheap, get you going, OK for small prints, optic. That's before you get seduced into buying more, bigger, better, more expensive lenses.Many people will NEVER need, or want, or lust after, anything else on their 300/350D (or even their 20D) however heretical that may be to the members of this forum!
As for the AP review of 'standard' zooms - the Nikon costs a lot more and is considered a fine optic in its' own right (and has a longer range of 18-70mm), the cheaper Sigma is still more expensive (remember the 18-55mm kit price) and the other Sigma is a lens aimed far higher, both in specification at f2.8 and price. Rather like comparing chalk and cheese.
Have just checked the prices here in the UK and the 18-55mm adds around 8.5% to the price of the 350D body whilst the 18-70mm Nikkor adds 26% to the price of a D70..........
Lets keep things in perspective please.

ding.....ding.....ding.....

KAC

[soupdragon]

Ok:

Here's the perspective:
Canon are sure proud enough to boast about the 350D features and quality but, I see nothing on the box that says "the lens included probably isn't worth a flying fart".
Yeah! Yeah!, sure you can buy the body alone but, for mugs like me, cheap was really attractive.
I based my purchase on my first experience with Canon whereby my A1 came with a budget price 50mm 1.8 which was s@*t hot for the money.
Like I said before, I am a total ham at photography but, when I fluke a top shot I don't want it marred by a technical issue.

[PaulB]

soup,
I have sympathy with you on this, believe me. BUT.
There's always a but isn't there?
But, "you get what you pay for", "there's no such thing as a free lunch" and all the rest of the homilies...........I know it doesn't help when you've paid your money and made what seems a wrong choice, just grit your teeth and get on with taking pictures. If the lens doesn't WORK then return it, if it doesn't work as well as you like then learn to get the best out of it in the meantime and save up for something better for the future.
I have seen images on the web from a 300D and the 18-55mm by (I think) a Polish photographer which I wish I had taken - regardless of the lens and camera used. So it isn't junk, just made to a price, so what? That 18-55mm cannot be any worse than lenses I used in the 1960s and 70s, and made money out of - simply because I couldn't afford any better, back then the 18-55mm would have been a marvel at any price.

[drisley]

I have read/seen many, many stories/comparisons that show the 18-55mm lens is very close in image quality when compared to 17-40L, especially stopped down at the wide end.
As far as plastic mounts go, my 50/1.8II is an amazing lens that probably out resolves the 20D.

[soupdragon]

You are all correct in what you say (including the but but but bit)

I shelled out for the 100 2.8 macro and sing it's praises on high, however (notice I avoided using the but word there) it was an unexpected cost, and I am sure there are people on this forum worse off than me that are drooling at the thought of having quality optics.
The bit about EF-S optics out performing L glass just says to me I got a Friday afternoon example.
That said (still no but word) if I tried enough of of them and could be bothered to tabulate the lens characteristics I am sure I would get some good results.

[Tom W]

Tom W:

You Said:

"True, a good image to start with requires less work than a mediocre image. But mediocrity is much more a function of what's behind the viewfinder than what's inside the camera nowdays. Even the inexpensive kit lens has been shown to produce great images in the right hands. Granted, it lacks a lot of features and is rough at the extremes, but it is the best $100 zoom around."

Well done, you've just bolstered my original post.

Can you not imagine yourself in that "once in a lifetime, got it framed perfectly & lighting bang on" situation only to have to explain to people, "it's rough because because I had to use my lens at it's extreme wide angle!"
Or, "sorry, it lacks the corner to corner illumination feature"
Or, "you would be able to read the logo on his cuff links if this lens had the make this picture sharp button"
Or, "No, I couldn't afford the right hands to go with this lens"

Sorry if I came across a little rough in the last post - I was up early, and preparing for a 575 mile drive yesterday.

Anyway, I think that others have answered your question somewhat - you're not going to get hassy performance out of the $100 kit lens. Just isn't going to happen. For one thing, you're comparing a small format DSLR with medium format. For another, you're comparing a very inexpensive tool with top-notch equipment. That's kind-of like saying that your since your Saturn GL doesn't perform as well as your Ferrari, it is junk.

Know your lenses, know their limitations - all lenses have them. All lenses vignette - some more than others. All wide angle lenses are a series of compromises. Zooms are comprised of more compromises than primes. Use that knowledge so that you don't have to explain problem images to your clients (and if you're gaining clients, you might want to step up a notch on the equipment if you find this doesn't meet your needs).

Also realize that you're not selling 100% screen shots of your images - you're selling 4X6, 8X10, maybe 11X14's. The kit lens can do all of those sizes under most circumstances, as can most lenses.

As for the right hands - you'll have to supply your own. You can't buy them unless you hire a photographer.

[soupdragon]

Tom:

I have to say I did not read any abrasion into your post.
In fact your replies seem so composed I assumed you did Yoga or Stress Management.
I'm the one that should calm down (bloody canon corporate bean counters)

[Tom W]

Tom:

I have to say I did not read any abrasion into your post.
In fact your replies seem so composed I assumed you did Yoga or Stress Management.
I'm the one that should calm down (bloody canon corporate bean counters)

No, I generally take my agressions out on the road. :)

Not really - I was in a bit of a hurry to get under way, but I still managed to erase and retype my response about 4 times before I hit the send button.

The bean counters are bad enough, but the corporate lawyers really mess things up.

[soupdragon]

True

[rdenney]

I shelled out for the 100 2.8 macro and sing it's praises on high, however (notice I avoided using the but word there) it was an unexpected cost, and I am sure there are people on this forum worse off than me that are drooling at the thought of having quality optics.

Note that for the kit lens to perform as well as an L lens it has to be used under very specific circumstances. One is at the wide end (the 55 end is dodgy, according to all reports). Another is well stopped down. I could match my 70-200/4L with the 75-300 cheapie that I took back after a day, but only at small apertures and then only after enduring horrible ergonomics. The L-series lenses are designed with versatility as a design objective, and perform well at wide apertures and over a broader range of conditions.

But I hear you regarding the unexpected cost of better lenses. But it's important to understand what is happening here, and I can illustrate that best using my own case as an example. Before buying a 10D, I did all my serious work in medium format and used 35mm only for family snaps. I had an Elan IIe and a cheapie Sigma 18-35, a nicer Sigma 28-70/2.8, and a cheapie Pheonix 100-400. At one point I added a 14mm rectilinear wide, also a Sigma. None of these lenses is at the top of any category except low price, and a couple are at the bottom of their category in quality. But I judged my results based on 4x6 prints in the photo albums we would show to family to tell the story of vacations and visits. My serious photos on those vacations were made using medium format stuff (Exakta 66 and various Ukrainian junque of the same lens mount, two Pentax 645's, or, depending on the look I want, a Rolleiflex; not Hasselblad but still solid, commercial-grade stuff and more usable for me than Hasselblad).

When I bought the 10D, I thought I'd use the lenses I already had. First thing I had to do was have all the Sigma lenses rechipped, of course. That was a disappointment. But then I made a discovery: Those lenses are crap! Yup, they are unadulterated junk, to use your description. I was horrified! After all, I made this image on Velvia with the Pheonix:

http://www.rickdenney.com/images/pinnacle_in_last_light_lores.jpg

It looked okay to me on Velvia, but it was trash on the 10D.

What was different?

I was looking at the image at 100% on my computer monitor, that's what. Before, I was looking at a slide through a 10x loupe (which is equivalent to a 10x15-inch print), or I was looking at 4x6 prints. In some cases, I was projecting scanned images using a computer projector at fairly low resolution (about the same as above). None of those viewing methods test the limits of lens performance the way looking at a 6 or 8 megapixel image at full resolution does.

So, the lens hadn't changed, but my standards just took a big jump because I was looking at everything in the equivalent of a 20x30-inch print. I've never owned anything in small format that would be a good performer at those print sizes. None of my old high-end Canon stuff in FD mount could, that's for sure.

That's why I asked if you had actually made any prints. You might be surprised that an 8x12-inch print is a lot more forgiving than that computer monitor.

And I have paid much more for lenses than I paid for the 10D. Two things happened: 1.) the 10D was good enough for me to use it for serious work, and 2.) my quality standards therefore jumped requiring an upgrade in equipment. That doesn't mean I've had to buy only L-series lenses by the way. I've gotten very good performance from a range of optics, including some cheapies in M42 screw mount used with an adaptor.

But there is no way a 15x23 sensor, even with a billion pixels, will allow images of the same quality as the 56x56 frame of a Hasselblad, even without the high-end expense of Zeiss Oberkochen lenses. It just doesn't start with as much information.

Rick "who thinks image quality has to be judged at the final intended display" Denney

[Roy Gray]

Hi for a start the 50mm 1.8 is one of the best std lens made its sharp and most people use it as a std for quality of sharpness also I have the efs lens and as others have said its a good allround lens sharp pictures at F8 are good, would suggest you look at your camera and check that it is ok.

happy shooting:D

[Tom W]

...

...Rick "who thinks image quality has to be judged at the final intended display" Denney

Well said. Hopefully, the final intended display is not a 100% crop.

[soupdragon]

RDENNEY:

Nice pic, you're just showing off now.
You, having used medium format will know the the sensation of projecting a large format chrome (that's my baseline standard for subjective image quality)
I still find it hard getting used to not having an acre of slide/negative to play with but then that's the price we pay for progress.

ROY GRAY:

I don't feel it's the camera as the images I get using the 100 macro are as good as I expect from an 8mp camera.
I owned the FD 50mm 1.8 and could guaruntee good results at almost any aperture but the new EF 50mm 1.8 I owned (for about three days) was not good). Given what others are saying I attribute this now to poor quality control.

As for printing, therein lies another tale of woe.
I may well start another thread on the apparent lack of unified standards between monitors and printers.

And completely off subject, I saw a couple of Swallows/Swifts yesterday evening which, legend would have it, marks the start of the summer. They are too bloody fast for me to photograph though.

[rdenney]

You, having used medium format will know the the sensation of projecting a large format chrome (that's my baseline standard for subjective image quality)
I still find it hard getting used to not having an acre of slide/negative to play with but then that's the price we pay for progress.


Yes, projected medium-format slides are stunning. But I never paid for a medium-format projector and rarely used medium format for chromes. I've done quite a bit of black and white and also color negative work with medium-format, however. My scanner will scan a 6x6 negative to about 6 megapixels. Even downsampled after that, I can tell the difference in tonality between many of those shots and the 10D images. The larger format just integrates far more information into the image, when displayed at the same size. Format is king.

That's why I've started a savings program for the upcoming Pentax 645 Digital. It will be cheaper than the current 1DsII, I already have a range of lenses for it, and I love the interface on those cameras. The sensor is estimated to be a 35x43mm size with 18.6 megapixels. That will easily support optimal 13x16 prints and still excellent 16x20 prints. People only think their 20D's make good 13x19's, heh, heh.

Get a monitor calibration kit. It will be worth the money. Then, you can have prints made at a lab and they will be close. The printer will be, um, closer. But you have to spend a bit of money to get a printer with good profiles or to have a profile made. Correct your base image to look the way you want it on the monitor, and then save it. That's your corrected image. Then, make a copy and adjust it to look the way you want on the printer (including cropping and resizing if necessary, plus color corrections and sharpening). That process is targeting. Correction and targeting are separate process, and if you keep them separate, life will be much easier.

Remember, digital images are like negatives--they are a step in the process to a final display form.

Rick "who didn't find it that much easier to dial in Cibachrome or C prints using color heads" Denney

[Croasdail]

Rick "who didn't find it that much easier to dial in Cibachrome or C prints using color heads" Denney

at least you had a color head... I had to use those dang filter packs... and the one filter I always needed was always missing. but that is another story for another day... reminds me of a monty python skit on who had it worst.. cheers!

[soupdragon]

RDENNEY:

Is it possible to calibrate a laptop LCD?????????????????
I notice in photoshop there are embedded profiles for canon printers, (I have the pixma 5000 if that means anything) I got bored reading the manual do you know if there is a profile for it anywhere.

CROASDAIL:

I still snuggle up to my enlarger sometimes (kind of a comfort blanket thing) and it does have dial in filters.
I was going to scan the bin all my chromes and negs but the effort involved outways the net benefit.
And I still do B+W printing as I don't seem to be able to get the tonal range (yet) with PS/digital.

Thanks for your time guys:

[Jon]

The better Canon lenses are outstanding, as tools. If you are unable to be persuaded of that, then perhaps you should go try Nikon.
the man never said that, it was a very specific complaint about ef-s lenses and their build quality. In general there is a huge issue with lens quality from all manufactures - having to have brand new lenses sent in for calibration is rediculas - and you see that a lot - hence all the post of "did I get a good one". So two issues with your statement.. we all know that Nikon's consumer line has the same issues so sending the chap in that direction does nothing- and we should not be so offended when someone takes issue with Canon - they are not god. When someone spends $500 plus for something - they should not have to be asking "if they got a good one" or not.

Croasdail:

Finally someone who actually understood what I was saying.

Dear Canon get it together or I am buying a bloody nikon.

Which was, as near as I can tell, if Canon doesn't stop making lenses that you don't approve of, you'll change over to Nikon, which you seem to be holding up as an across-the-board gold standard for lens quality.

[rdenney]

Is it possible to calibrate a laptop LCD?????????????????
I notice in photoshop there are embedded profiles for canon printers, (I have the pixma 5000 if that means anything) I got bored reading the manual do you know if there is a profile for it anywhere.


Printer profiles will change with the kind of paper you use, too.

Whether you can calibrate your laptop display depends on the display. I have used two laptops where I attempted to calibrate the display. Laptops usually don't give you any hardware controls over color balance, and thus it is quite difficult to truly calibrate the monitor. (Calibration is adjusting the monitor to your preferred viewing light, and profiling is measuring the monitor so that software will know how to display color so that it comes out right. Hardware calibration does not usually decrase gamut, but software calibration and profiling does. The better the hardware calibration, the less the profile will make big changes, and the less gamut you will lose.)

I calibrated both using software calibration, but I had to choose a very blue viewing light to keep from reducing gamut too much.

The problem with the newer display is that the color balance for highlights changes with angle of view. Thus, the top half of my screen is greenish, while the bottom half is pinkish. I hate it, and I just don't use it for critical color work.

If you are going to use your laptop for this, see if it supports a second monitor. Then, put a trusty CRT display as your second monitor, and calibrate and profile the system on the secondary display. Then, view photos and do your photoshop work on that display. I've done that with this laptop and it works okay. If you need to do photo work on the road, then you'll have to be really choosy about the laptop you buy. The screens on laptops are not as good as they were just three years ago. My laptop from three years ago is 15", 4:3 aspect (so it will fit in my brief case and backpack), and 1600x1200 resolution. The best I could do for the new one was 1400-ish by 1050, or some such, and still have a machine that would fit in my briefcase. And the new display has more contrast but that awful color shift with angle of view; the old did not suffer from that problem.

I don't know if the Mac laptops are optimized for the display, but I'll bet they are more likely to be than most Windows laptops that are optimized for gamers and office workers.

Rick "who does critical color work at home" Denney

[rdenney]

And I still do B+W printing as I don't seem to be able to get the tonal range (yet) with PS/digital.

In my opinion, you will never get the look of a silver print with ink. It's just too different. If it's the look of silver you like, stick with your darkroom.

But that doesn't mean you can't get excellent black-and-white prints from printers. For one thing, the choices for matte-finish papers is much better for ink printers than for wet-process prints. And those matte finishes can produce some extraordinarily deep and rich blacks just not possible with the kind of silver-based print paper we can buy now. The new Epsons set to come out presently supposedly have a range of black and gray inks making it possible to make full-range prints using only neutral-tone inks, and that is going to give the after-market quad-tone arrangements a run for their money.

A printer that only has one neutral ink (black) will not make good black-and-white prints using that ink. It has to dither too much to make the lightest tones, and your highlights will block up. I get much better results printing the monochrome image using all the inks (set the image to RGB after using the channel mixer to make the image monochrome--even if it's a scanned black-and-white negative). Then, the blacks will use the black ink plus a mix of the colored inks and it will be much richer. But you'll get subtle crossovers. These crossovers are sometimes interesting and therefore not always bad. Toning the print slightly minimizes them considerably.

You have to judge ink prints on their own merits. They will always be different from chemical prints, but that also means that in some ways they are better. Because of the editing power in the computer, I find I usually get much closer to my previsualization.

Rick "who thinks digital display requires a fresh set of eyes sometimes" Denney

[Jon]

RDENNEY:

Is it possible to calibrate a laptop LCD?????????????????
I notice in photoshop there are embedded profiles for canon printers, (I have the pixma 5000 if that means anything) I got bored reading the manual do you know if there is a profile for it anywhere.


Yes - Check out, for instance, the ColorVision Spyder, Monaco EZ Color or Optix, or the Gretag-Macbeth EyeOne (http://www.bhphotovideo.com/bnh/controller/home?O=NavBar&A=search&Q=&ci=6443).

Check out Canon's web site if you're looking for the Pixma colour profiles.

[soupdragon]

RDENNEY:

Sorry for the delayed response I use my works computer for internet so can only respond on weekdays.

My laptop lcd has the same problem with viewing angle.
I can improve this a bit by moving away from the screen, sadly my arms won't then reach the keyboard.


JON:

I'll check it out, though I can find no method of adjusting my lcd monitor other than the brightness.
Unless it does some kind of software thing (which is all a bit beyond me)

[soupdragon]

JON:


I see from your equipment list you own an F-1, did you have the FD 50mm 1.8?
If you did, I can't beleive you think the new plastic mount 50mm 1.8 is anywhere near as good.
BTW:
No I would not hold Nikon as the across-the-board gold standard for lens quality, the reference standard would be the CZ 80mm Planar (or any of the Hasselblad range for that matter).

[Andy_T]

No I would not hold Nikon as the across-the-board gold standard for lens quality, the reference standard would be the CZ 80mm Planar (or any of the Hasselblad range for that matter).

No problem. Look HERE!!! (http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=30059&item=7520179196&rd=1&ssPageName=WDVW) :D

Best regards,
Andy

[soupdragon]

Nice one:

How on earth did you stuble across that nugget?

[Andy_T]

I get very pleasing results with a CZJ Sonnar 135/3.5 and a Russian Jupiter-9 85/2.0 (also Sonnar design) in M42 mount, so I recently gave some consideration to getting a CZ Planar 85/1.4 (which is not the one you mentioned, as I found out) or Sonnar 85/2.8 in Contax mount and a corresponding adapter ...

With knowledge like that in the back of your head, the eBay search function is normally all you need...

However, I currently feel that I will most likely rather go the Canon 85/1.8 route, because AF is really helpful on a 20D (at least for me).

Best regards,
Andy

[AXENA]

Naytwan, that is the cutest little nose I have seen.

I don't know why all the uproar about the supposed lack of quality of plastic mounts. My Glock is mostly plastic and my Colt is virtually all metal. I'll bet my life on the Glock any day over the Colt.

I know we are way off subject here, but the GLOCK, while a very nice weapon, DOES have some serious flaws. I NJ, the state police I believe switched from the GLOCKs bue to rounds jamming up. A plastic vs. metal issue? Who knows. But I think if I was to trust my life with something, it would be a metal Colt.

On the other hand, I have had some great results with the kit lens which I origianlly bashed when I first got it.... just needed to learn it's "quirks." Once you get used to them, it does take some decent shots. I don't think I would rate it anywhere near my primes, or my first L-glass which will be here in a day or two, but I guess it is all in the user. Would I trust the 18-55 with portrait work or sports action... no. Will I use it for walking around. Absolutely.
-GSL

[Steve Parr]

the state police I believe switched from the GLOCKs bue to rounds jamming up. A plastic vs. metal issue? Who knows. But I think if I was to trust my life with something, it would be a metal Colt.

I thought the Glock was some kind of poly-carbon fiber deal as opposed to plastic...

Steve

[soupdragon]

I thought the Glock was a type of wooden shoe, warn by tulip growers, and had a slightly higher resolution than an 18-55 efs.

[AXENA]

LOL
-GSL

[kenyc]

I know we are way off subject here, but the GLOCK, while a very nice weapon, DOES have some serious flaws. I NJ, the state police I believe switched from the GLOCKs bue to rounds jamming up. A plastic vs. metal issue? Who knows. But I think if I was to trust my life with something, it would be a metal Colt.

...
-GSL

yes off topic, but Glocks are far more reliable than the competition, particularly 1911 style pistols. They were designed from the ground up (originally for the Austrian army) to keep shooting when others failed. :) I don't trust my life to anything else.

KAC

[C.S.I.]

Hmmmmmm........I actually get exceptional images from both my 18-55 and 50 1.8.......just lucky I guess :lol:

[jfrancho]

I thought the Glock was a type of wooden shoe, warn by tulip growers, and had a slightly higher resolution than an 18-55 efs.I thought "glock" was short for glockenspiel.

[rdenney]

I thought the Glock was some kind of poly-carbon fiber deal as opposed to plastic...

What do you think those carbon fibers are imbedded in?

Rick "who thinks 'plastic' is a broader term than most people realize" Denney

[rdenney]

I know we are way off subject here, but the GLOCK, while a very nice weapon, DOES have some serious flaws.

The only beef I had with the 9mm Glock that I tried was that the slide came too close to rubbing my hand. But I could shoot a tighter pattern with it than with my own 9 (a Tanfoglio/EAA Witness). Even so, mine was good enough to get a perfect score on my Texas concealed-carry permit test, when I lived in Dallas in a neighborhood with a rash of armed and violent car-jackings.

I also have a Beretta 25 that if I ever have to use without thinking will probably cut a slot in my hand.

Where I live now, I can carry the weapon on my belt in full view without a permit, though I choose not to. I'm out in the country where accuracy with a BB gun is more useful than with a handgun.

Rick "who has big hands which is good for accuracy and bad for slide interference" Denney

[drisley]

Back to the original topic about the EF-S lenses being junk, check out the picture in this thread (http://photography-on-the.net/forum/showthread.php?t=76706) taken with the 15-55mm kit lens. It's as good as anything I've seen from any wide angle L lens.

[soupdragon]

Back to the original topic about the EF-S lenses being junk, check out the picture in this thread (http://photography-on-the.net/forum/showthread.php?t=76706) taken with the 15-55mm kit lens. It's as good as anything I've seen from any wide angle L lens.


It's a warming image but I'm not convinced.
I find it very hard to tell from web images so small if the resolution is good or not.
If the 18-55 is as good as L glass? how do you explain the price difference?
This thread has gone on too long!

To summarise:

Possible soupdragon induced errors:

Can't hold a camera steady enough @ 1/2000th.
Not figured out how to use AF Points.
Not got AF switched on.
Can pick a dud EF-S from forty paces.
Not an accomplished photographer.
Should try looking through the viewfinder instead of the manual.
Expects 8mp to be as good as kodachrome.
Have not found the lens sweet spot.
Had camera strap dangling in front of lens.
Wouldn't know a good picture if I saw one.
Makes too many generalisations.
Not figured out how to set captured image quality.
Haven't posted any pictures therfore am not worthy to have an opinion or make observations.
Should not say bad things about Canon gear on this forum (naughty naughty)
Should have bought a Nikon.
Should do something obout receding hair line.
Should not drive a crappy Ford Mondeo.
Ought to get a life.

Did I miss anything?

[kenyc]

Back to the original topic about the EF-S lenses being junk, check out the picture in this thread (http://photography-on-the.net/forum/showthread.php?t=76706) taken with the 15-55mm kit lens. It's as good as anything I've seen from any wide angle L lens.


It's a warming image but I'm not convinced.
I find it very hard to tell from web images so small if the resolution is good or not.
If the 18-55 is as good as L glass? how do you explain the price difference?
This thread has gone on too long!

To summarise:

Possible soupdragon induced errors:

Can't hold a camera steady enough @ 1/2000th.
Not figured out how to use AF Points.
Not got AF switched on.
Can pick a dud EF-S from forty paces.
Not an accomplished photographer.
Should try looking through the viewfinder instead of the manual.
Expects 8mp to be as good as kodachrome.
Have not found the lens sweet spot.
Had camera strap dangling in front of lens.
Wouldn't know a good picture if I saw one.
Makes too many generalisations.
Not figured out how to set captured image quality.
Haven't posted any pictures therfore am not worthy to have an opinion or make observations.
Should not say bad things about Canon gear on this forum (naughty naughty)
Should have bought a Nikon.
Should do something obout receding hair line.
Should not drive a crappy Ford Mondeo.
Ought to get a life.

Did I miss anything?

Yeah, the part about the Glocks.
:)

KAC

[Andy_T]

Did I miss anything?

No, seems to be a pretty concise list so far :lol:

As you know, the first step in improving yourself is self-awareness :p

Best regards,
Andy

[soupdragon]

KENYC:

Ammendment to post.

Should buy a pair of Glocks and so I can make informative judgement on Dutch footware.

Andythaler:

My yings and yangs are coming into focus now.

[Tom W]

KENYC:

Ammendment to post.

Should buy a pair of Glocks and so I can make informative judgement on Dutch footware.

Andythaler:

My yings and yangs are coming into focus now.

I think that you've got it all together now. Buy a Glock and your troubles will be over. They're very easy to shoot accurately.

And keep the Ford Mondeo, especially if its paid for already. Having no car payment presents a bit of freedom (and is considered un-American by some). And it gives you a bit more money to buy expensive lenses.

[soupdragon]

Does 3 red stripes make you a sergeant???

[Tom W]

Does 3 red stripes make you a sergeant???

Depends on where you place them.

[AXENA]

The only beef I had with the 9mm Glock that I tried was that the slide came too close to rubbing my hand. But I could shoot a tighter pattern with it than with my own 9 (a Tanfoglio/EAA Witness).....

LOL.... nothing, and I mean NOTHING, hurts as bad as getting one's hand smacked by the slide! When learning to shoot, I made the major amateur mistake of having my thumb behind the slide! Needless to say, the weapon ended up on the floor next to me writhing in pain. Never made that mistake again, and learned early on that the weapons instructor is your friend, no matter how many times they yell at you..... they do that for a reason when they see you doing something really STUPID!

I do like the Glock.... now I like the H&K a little better.... although rifles are still my fav for sport shooting. Can't carry here in NJ.... lucky you! Fav handgun is still the Eagle .50 - accurate and feels great in the hand. Wow.... did we ever digress from shooting photos on this one...... oh well... guess shooting is shooting.... be it photos or targets... either way, if your eye is off, so is the shot!

So I suppose in some weird way, this topic is photography related. Actually, it may be good practice for photogs to go out and target shoot..... develops a sense of aiming directly at your subject (target) and getting that right "shot." :-)

Ok... back to photography.....
-GSL

[RbrtPtikLeoSeny]

I'm still trying to convince my dad to let me have a bb gun..... faught for 6 months for the paintball gun..... o_o.....

[drisley]

Back to the original topic about the EF-S lenses being junk, check out the picture in this thread (http://photography-on-the.net/forum/showthread.php?t=76706) taken with the 15-55mm kit lens. It's as good as anything I've seen from any wide angle L lens.
It's a warming image but I'm not convinced.
I find it very hard to tell from web images so small if the resolution is good or not.
If the 18-55 is as good as L glass? how do you explain the price difference?

Hey, sounds like you've made up your mind then already, and no amount of pursuading will convince you. No biggie, the money you spend on the L lenses is yours. Heck, I myself just purchased a L wide angle lens.
One of the big reasons there is a price difference is construction of the lenses.
The Tamron 28-75mm lens has been shown to rival the Canon 24-70L optically, and the price difference is HUGE! One of the reasons is that the Canon is built like a tank, while the Tamron is not.
Also, in my studies of marketing, we were taught about different pricing strategies, and one was elite pricing, where a company purposely marks up a product by a huge amount to give the "illusion" of quality. Thus the product will appeal to a richer segment of the population (or professionals in the industry).
This may or may not be going on in the case of L lenses, but it's hard to say.

[soupdragon]

drisley:

I watched the Canon promo DVD that came with my 350 and I seem to remember it mentioning a different type of glass in some of the elements.
I can only assume this results in a reduction of chromatic abberation.
Did you miss the bit on this thread where I mention the Amateur Photographer lens test report?
I think that confirms either all Canon lens' are iffy or just the kit (depending on your stand point)
In case you hadn't noticed, this thread has degenerated to a discussion about semi-automatic wooden shoes.
And anyway, yes, I am bitterly jealous of anyone with L glass.
Grumble Mumble, I want, I want, I want, me me me, gimme gimme,

[Jon]

Did you miss the bit on this thread where I mention the Amateur Photographer lens test report?
I think that confirms either all Canon lens' are iffy or just the kit (depending on your stand point)

Or just the sample they tested . . .

Go back and read Persian Rice's troubles in getting a good Tamron 28-75 if you really believe that every single instance of a given lens design is absolutely identical in quality across the board.

[soupdragon]

Jon:

What's a DOMKE?

[Jon]

Jon:

What's a DOMKE?

Simply the finest and most usable camera bags in existence (http://www.tiffen.com/domkecat01.htm), designed by a photographer for the Philadelphia Inquirer when the paper offered to pay for making whatever camera bags the photographers wanted to use. You'll probably see almost as many Domke bags around photo ops and press conferences as you will white lenses.

[soupdragon]

Why is it that most people ripped into me for this thread and yet time and time again I see posts about "what are the purple fringes on my pictures" or "the front element of my lens just fell off" or an entire thread devoted to "how to make the kit lens sharper"

Why?

[guitarman]

Well I can only speak for the 17-85 EFS. Yes it was junk. It gave me problems terribly with purple fringing and lens flare. Not too much the tack sharp I was promised but not to bad. After two copies I'd have to agree that it was junk. Went with the 17-40 F4L after that and never looked back.

[Mohawk]

You get what you pay for. And I would not blame Canon for making cheap lenses, because they do make better glass, but you have to pay to play. I found this out years ago when I became interested in photography. And it still holds true to this day. I know that my 20D and 1DMKII photos suffer with non Canon and some of the cheaper Canon consumer lenses, if the lenses work at all. Both of these cameras, in my opinion, require a high quality lens to get the desired results. I know allot of people complain about the costs of say an "L" lens, but it has always been this way with SLR gear. When was SLR photography ever inexpensive?!

I know that in the golf business, we had one heck of a surge during Tiger Woods hey day. It was easy to see that this was a fad, as the golf industry had never seen such times. Well, that did not last long after all of the dust settled. Most people were getting into golf for all of the wrong reasons and never had the tenacity to stick with it, or had any clue as to what they were getting into. The equipment was the first hurdle, not cheap for quality gear. Most went to Kmart or similar stores for clubs. Lessons, not cheap, but most people skipped that part, as it looks so easy on television. Green fees, whoa!! $60+ for a round of golf! Then there is practice time at the driving range, $6 for a bucket of balls!!! Etc., etc., etc...

What's my point? Photography is allot like golf. I see allot of people going from the P&S realm, putt putt golf, into the dSLR world, Tour Golf. Allot of things hold true for both activities, and the entry fee is not cheap. You may be able to afford a good driver (camera body), but that will not make you a good golfer. A good set of quality fitted irons will be needed to get the most out of the game, and you will need a few of those (lenses). Then you will need a putter, a bag and some balls (accessories). And then there are the practice sessions. And maybe some lessons would be a good idea.

I am not saying that you can't play with what you can afford, but don't expect to be a pro with Kmart equipment. Work with what you have, learn it's limitations and live with it till you can afford better gear. And practice practice practice!

Mike

[kenyc]

Why is it that most people ripped into me for this thread and yet time and time again I see posts about "what are the purple fringes on my pictures" or "the front element of my lens just fell off" or an entire thread devoted to "how to make the kit lens sharper"

Why?

You are clearly just here to cause trouble and stir the pot. This thread has been dead for over a month.

KAC

[MDJAK]

Wow, I just scanned this thread, haven't seen it before, and now am wondering if I made a mistake. I posted on another thread about taking a friend's girlfriend to B&H yesterday where, with my input, she purchased the 20D along with the kit lens. This lens cost 600 by itself. If it is as bad as people say, I feel quite bad about my advice. While I know she is not sophisticated enough to tell any difference, I still hope this is not true.

[mikesd]

The 17-85 IS kit lens is a very nice lens, I wouldnt worry to much.

[jfrancho]

Wow, I just scanned this thread, haven't seen it before, and now am wondering if I made a mistake. I posted on another thread about taking a friend's girlfriend to B&H yesterday where, with my input, she purchased the 20D along with the kit lens. This lens cost 600 by itself. If it is as bad as people say, I feel quite bad about my advice. While I know she is not sophisticated enough to tell any difference, I still hope this is not true.The kit lens being referred to here is probably the EF-S 18-55 f/3.5-5.6, which B&H has priced at $139.99.

[sodakred]

I agree with mikesd. I bought the 17-85 for a walk-around lens. It keeps getting bashed by people, probably due to its price (and EF-S fitting), which is slightly higher than it probably should be.

My copy makes wonderful photos and it has great reach on both ends. I keep trying to convince myself to get rid of it (somewhat inspired by all the bashing and wondering what I'm missing) but it just continues to perform nicely.

I also have a 70-200 2.8L, which is superb, but the 17-85 definitely holds its own. I also have a 50 1.4, which rarely gets used except in low-light non-flash situations, because the 17-85 covers that range and gives me equal or even better output.

Sodakred

[kenyc]

Wow, I just scanned this thread, haven't seen it before, and now am wondering if I made a mistake. I posted on another thread about taking a friend's girlfriend to B&H yesterday where, with my input, she purchased the 20D along with the kit lens. This lens cost 600 by itself. If it is as bad as people say, I feel quite bad about my advice. While I know she is not sophisticated enough to tell any difference, I still hope this is not true.

It's not. Re-read the thread the 18-55 is the "standard" kit lens and is about $100, the 17-85 is about $600 and it's definitely a better lens, but there is NOTHING wrong with the 18-55. Check out the pics on this forum taken with it or got to pbase and search for pics taken with it (under equipment or something like that).

KAC

[MadMesh]

I just bought the 10-22, I figure its got to be better than 3rd part stuff like sigma , tamron and tokina.

Im just hoping for a sharp copy, is there is such a thing.

[Gary_E]

this is an itty bitty crop from a brand new 50mm f1.4 yesterday.. i couldnt get get this resolution from my old 50mm 1.4 nikkors running kodachrome 25