10D and Good Glass Myth

The image quality difference with good glass is as drastically

different with the 10D as it is with film. It's not a slight

difference. You would have to be absolutely blind not to see it.

Unless you are printing 4x6 prints at Walmart. Even then so! This is

the most RIDICULOUS new myth on this forum. Why it is being

proliferated I have no idea. It is categorically untrue.

The myth that the difference between good and bad glass is not apparent on the 10D.

Jorge,

<p>

Many of the postings on this forum have made exactly the opposite case: i.e. the use of digital (and for most users this currently means the 10D) and the ability / inevitability of viewing / printing larger pictures (an actual pixels view on a 17" monitor is equivalent to viewing a 36x24" print) actually requires better lenses.

<p>

The one area where it can "improve" cheaper lenses is with lenses that have good central definition but are soft in the corners or show some vignetting - these deficiences will be less obvious with the crop factor of a 300D/10D/D60 etc.

<p>

An example of the latter is my 100-300 f4.5-5.6 USM zoom which performs better on my 10D than on film for the reasons given above and actually is a very useful and useable lens on the 10D.

Due to the crop/magnification factor of 1.6 I think lenses must be better (related to x1.6) at the center part when used with a D60/10D/300D than combined with 35mm body.

Derek, I am aware and agree with posts to the contrary. But every since Bob's review of cheap Canon zooms on the 10D there has been a rash on this forum of people believing that the quality glass on a 10D makes no difference. You need to read through post since that review and you will see what I mean. I hate to sound like am stating the obvious but it doesn't seem to be that way for some folks. It's just so ridiculous I had to post.

To believe that using a cheapie consumer lense on a 10D is just as good as a high-quality L lens is silly, of course.

 

But refusing to believe that the crop factor *can* give an effective performance enhancement to those same cruddy zooms compared to a full-frame sensor (be it film or digital) is equally silly.

 

There is a difference, but it's not worth concerning youself over if a) you're a pro shooting for guaranteed publishability, or b) the money is not your primary concern in the deal.

I too have noticed alot of opinions for lens quality being more noticable on digital,not less.

<br>One related point though....from what i've seen so far,the 10D has such clean and clear output that one can get some "completely acceptable" pictures with cheap lenses.A case of the camera making up for the lens to a point.But it's still nowhere near what a great lens can do.....

You can use cheap glass on the 10D and it can come close (note not equal) the better glass under the following circumstances: FIRST the cheap glass must be stopped down 3 stops, SECOND one end of the zoom range (on the 28~80 its the 80) will suck but the other end will be fine, and finally THIRD you will need to have better than normal technique to get the best from average glass.

 

Where does the "L" glass pay off, FIRST wide open, SECOND at both ends should be fine except for the 100~400 at the 400 end (note there maybe others), THIRD your technique may actually need to be better because of the heavier lens.

 

BOB wrote a really good article compare the 28~80 vs other zooms, it did pretty good when stopped down, go to the article section to see his review.

 

Note most cannon primes (even the cheap ones) will blow the cheap zooms out of the water.

 

GS

Jorge

 

You are wrong. I'll tell you why. A 10D cannot, under any circumstances, resolve more than 65 lp/mm in an image. That's an absolute upper limit based on pixel spacing and can easily be verified by experiment.

 

Just about any lens is better than that (at least in the center of the image), be it a 75-300 or a 300/2.8.

 

Now FILM can resolve 120 lp/mm or so. Lower quality lenses can't match that, but high quality lenses can.

 

SO....the result is that the DIFFERENCE between high and low quality lenses is LESS with the 10D. High quality lenses are still better, there's no doubt about that, but the DIFFERENCE is less.

 

Crap lenses still look like crap on a 10D. However the difference between a GOOD lens and a GREAT lens isn't as obvious as it would be if you were shooting Velvia and looking at the results under a 10x loupe.

 

 

The above is the 5 cent version of the situation. The $10 version goes into MTF of lenses, MTF of film, MTF of CMOS sensors and other esoteric stuff. There are also subjective factors like the fact that digital images often LOOK better (even when they have the same of even slightly lower resolution as film) because of the nature of the image and the lack of grain.

Bob,

It was nice of you to give Jorge a way out (subjective factors).

 

Jorge,

Bob is defining image quality as lp/mm in an image. Were there other qualities that you were referring to? Or, in other words, I'm sure there were experiences of yours that led you to your conclusion. What were they? Things other than lp/mm perhaps?

 

I'd like to offer a quote from an article at the following link: http://www.luminous-landscape.com/tutorials/understanding-series/understanding-mtf.shtml

 

"And, to make things even more indeterminate, it can clearly be shown that the actual image quality produced by a lens does not correlate at all well with its measured resolving power. For this reason resolution test results alone should not be considered a valid measure of a lens' goodness. This is why the optical industry as well as those that really want to understand how a lens is performing rely instead on its MTF." - Michael Reichmann

 

He says "goodness" instead of "optical quality" but I believe it still applies simply because, and as Bob stated, we're dealing with things that are subjective to each individual. According to the linked article, "resolution is not a quantifiable scientific absolute." Or put differently, it's also subjective.

I should have spent more attention to the $10 version.

The differences between "good" glass and "low" quality are detectable when using film,digital sensors, or on a lens test bench. If your personal standards are low; or the enlargement factor; they may both appear OK.

I know what I see and I know what I sell Bob. Lines per millimeter are a very small portion of the pie as far as optical quality is concerned.

Here's an example of what I've noticed with my 10D and lenses. I own a 200mm f/2.8L and also a 70�210mm f/3.5�4.5 zoom. In terms of resolution the L lens clearly outperforms the zoom. It's sharper wide open than the zoom is at f/8. But the zoom still gives me a snappy image at the 210mm setting. Not as much detail as the L lens but contrast is good.

 

Now let's print 11x14" shots with minimal cropping from both lenses. As it turns out the zoom is sharp enough wide open (f/4.5 at 210mm) that it can resolve detail you can't see in an 11x14" print at normal viewing distances. Since 11x14" is as large as I go (other than a few 13x19" prints done for kicks) the difference in resolution between the two lenses hardly matters. The zoom provides enough snap and enough detail that its print differs very little from the L lens print. The obvious difference is the zoom creates a warmer image. Only on close inspection do you notice small detail in the L lens print is better defined.

 

So, yes, you can certainly tell the difference between a decent lens and an excellent one with the 10D. But depending on what you do with your photos the difference may not matter much. I use my zoom most of the time due to its smaller size, lighter weight and greater versatility. Its resolving power and contrast are good enough for most uses. I use the 200 f/2.8L when I need the extra speed or the shallower depth-of-field.

 

-Dave-

"10D and Good Glass Myth." I've not heard this one. Did you make it up? The myth I've heard repeatedly is the 10D needs good glass more so than film. I assume this myth arose from bored techno geeks viewing image files at pixel level on computer monitors whereas in the past they merely viewed 4 x 6 prints.

Yeah, no doubt, those Tameron lenses are JUST AS GOOD as the highest end Canon lenses... Riiiiight...

Resolution is a measurable quantity. Sharpness is subjective. Image quality is subjective. Both Sharpness and Image Quality depend on resolution, contast and acutance.

 

The 10D responds well to changes in contrast, in fact as well as film does. The 10D also responds well to changes in acutance (edge sharpness). The 10D responds to changes in resolution (MTF if you prefer), but only up to a point. That point is the MTF curve BELOW 65lp/mm. It doesn't matter what the MTF curve looks like between 65 lp/mm and 500 lp/mm because, for one thing the anti-aliaing filter in front of the sensor filters out all that information, and for another thing even if it DID reach the sensor all it would do would be to create Moire patterns and other aliasing artifacts.

 

Now if the 10D was totally imune to lens quality, that would be great. I'd throw out my 300/4L and 500/3.5L and buy a 50-500 Sigma Zoom and pocket the spare $2000 left over. So yes, you can see the effect of better lenses. You see it most in terms of contrast when used wide open and at the edges of the image.

 

So there's no doubt that a 28-135/3.5-5.6 IS will give you better images than a 28-90/3.5-5.6. The real question is whether a 85/1.8 at f5.6 will give you a better image than a 28-135/3.5-5.6 at 85mm and 5.6, or whether a 300/2.8L (at f4) will give you better images than a 300/4L, or whether a 200/2.8L will give you better images than a 70-200/2.8L. On film I'd say the answer is yes if you use Velvia and examine the slides with a loupe. On a 10D I'd say the answer is no.

 

So while you can certainly tell the difference between bad and good, the difference between good and great does get masked to some extent by the limitations of the sensor in the 10D. It's a bit (though not exactly, so don't take this litteraly and start going on about film gaine etc.) like saying that using ISO 50 slide film you'll see differences betwween lenses that you wouldn't see on ISO 400 print film. I don't think you'll find a lot of people disagreeing with that statement. The recording medium DOES matter when it comes to revealing differences between lenses.

 

So I'm not selling off all my "L" glass. On the other hand I'm not selling off my consumer zooms either, since neither the "L" nor the "non-L" lenses are going to show detail at 65 lp/mm or above on a 10D.

 

BTW I'm not making this stuff up. I did some field tests using a 75-300 at 300mm and a 300/4L. The 300/4L images directly from the camera had a little more "snap" and a little better sharpness. However with suitable tweaking in PhotoShop to bring up the contrast a little and add a little sharpening, the 75-300 images were pretty good. Better than I would have expected and significantly closer to the 300/4L images than I would have predicted based on my previous experience with these two lenses shooting ISO 100 slide film. The 300/4L images were still better, but whereas with film I wouldn't shoot the 75-300 at f5.6 and expect to get usuable shots, with digital I would because of the effects I can achieve working directly on the digitally captured image in PhotoShop. Scanning the film images just doesn't work as well.

I in regard to fabrication on my part to posts in this regard on this forum this would be pretty lame don't you think. Read some posts butt head. I don't like being called a liar. Here is a quote from a post that was deleted (for some strange reason) a couple of days ago.

 

Jim Larson wrote: "But back to the topic: Suuure => the 10D is a cheap APS with a price worthy of a top end 1V price. . the atrocity is made worse by the fact that the "L" lenses we like are not effectively used as a result."

 

http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=005s2a

I see you just posted Bob. My previous comment was not directed at you obviously. So, previous responder to my post was questioning my having read posts in this regard or fabricating an observed recent line of thinking on this forum.

Bob I respect your opinion but I whole hartedly disagree. I think maybe we are looking for different qualities in our optics. For me good glass is that with will resolve well but also (and more importantly) give a higher degree of 3 dimensionality to a printed image. I want to make the surface disappear. This is essential for me compositionally. I want to establish relationships between foreground and background object via the illusion of 3 dimensions on a 2 dimensional plane. I don't see this being pulled off very effectively with most consumer glass. I also find a lens which is only useful at f/8 to be totally useless for my needs. I need my tools to give me the creative flexibly to work with spatial relationship at different aperture settings and focal lengths.

Hey Bob I have learned not to get into arguments if I am not knowlegable about the

subject so I won't get into this one. But I can say that I have read many post on many

forums and a very large majority (in fact all that I have read up until this one) say that

good glass is more critical with digital cameras than with film. That leads me to belive

that this has been the "real world experience" of many people. There are those, my

brother included, that can't except that things don't always come out "real world" the

way they should on paper, assuming that your info is correct or relevant. As I said I

don't know you may be 100% correct technically and maybe even in real world use,

but when the landslide of opinion that I have seen is on the other end of this it is hard

not to think that "real world" that is where things are at.

NOTE I did not say that you were wrong.

With all due respect, Pat

With film and sensors; all life doesn't revolve around lab tests; and shooting high contrast USAF resolution charts. <BR><BR>Better glass will have a higher contast & MTF at more practical resolutions; like 20 to 40 cycles/mm. Most better high end zooms; and most all fixed focal length lenses have a higher contrast than the plain Jane cheaper starter grade "consumer zooms"; and will give better results on sensors and film. The maximume resolution of each can be similar; but the better lenses will have a higher contast & MTF at lower more practical resoltions. This makes a better looking image. <BR><BR>This same argument happened 40 to 50 years ago; when film movie camera lens were used on early TV video camera tubes. Some worked well; some didnt. This is the reason that the IEEE and SPIE and other groups got serious with testing contrast versus resolution; which resulted in MTF testing being developed. Two lenses can test the same as far as maximum resolution; but one can be far better/higher contrast at lower resolutions; than another.<BR><BR> The special "lenses for video cameras" splitoff/argument many decades ago; is similar to the current special "lenses for digital cameras" that are hawked/discussed/argued today.

I also think it is important to state that when we speak of contrast in a lens it has noting to do with global image contrast. Global or overall contrast refers to how the overall tonal gradation is distributed in an image from lightness to darkness When we discuss contrast in a lens we are referring to "micro-contrast" something entirely different. Micro-contrast refers to the ability of the lens to differentiate between smaller and smaller details of more and more nearly similar tonal value. This is actually what creates dimentiality in an image because it allows the viewer to perceive greater amount of differentiation between boundaries at a higher level between areas of slightly different tonal values. Micro-contrast is for me the hallmark of high quality glass not resolution.

Patrick, I really wonder how much of what we see posted on the 'Net regarding lens performance is based on real-world use. When I look at photos taken with my 100mm f/2, 200mm f/2.8L and 70�210mm f/3.5�4.5 lenses at "actual pixels" resolution in Photoshop it's clear the zoom doesn't resolve as much detail at 100 & 200mm as the two so-called "primes." The zoom is also a bit lower in contrast. But when I make 11x14" prints, after adjusting contrast and applying some USM, the main difference between the zoom and the other two lenses pertains to color rendition. The zoom is warmer. I can only see a difference in resolution when I stick my nose up against the prints, and even then the difference is very subtle. At a normal viewing distance the difference disappears.

 

For me the "real world" means looking at prints because a print is the ultimate form my photos take. If resolution differences between lens A and lens B don't translate onto paper does it really matter that one lens outresolves the other?

 

-Dave-

David the difference for me is night and day from across the room.