Soft copy/Sharp copy.

[dkangel]

Originally Posted by Double Negative

See, that's the weird thing.

I've bought five new L lenses and one used - every one of them is spot on.

Not for me. You can ask my wife. I have a 98% failure rate on just about anything I buy. I can buy a new fan and it will not run. I can buy a new appliance and it will have missing pieces. Weird thing is I am not superstitious by nature being an engineer and all but after a while you start to wonder. In fact I just bought a scanner last week and all pictures I scanned had weird marks at the top of it. I cleaned the glass but it didnt help. Had to exchange it for another copy of the same model. After a while you get used to it.

[thekid24]

Originally Posted by dkangel

Not for me. You can ask my wife. I have a 98% failure rate on just about anything I buy. I can buy a new fan and it will not run. I can buy a new appliance and it will have missing pieces. Weird thing is I am not superstitious by nature being an engineer and all but after a while you start to wonder. In fact I just bought a scanner last week and all pictures I scanned had weird marks at the top of it. I cleaned the glass but it didnt help. Had to exchange it for another copy of the same model. After a while you get used to it.

Well then out of the kindness of my heart I feel it would be a smart choice to graciously let me have your camera gear. If you truely value them then you'll understand, and dont worry I keep all my camera gear in pristine condition, and I will gladly do the same to yours Remember the odds are too high, so its only a matter of time.

[tfitz]

I just purchased a Tamron 28-75/f2.8 for my XTi and I used it for a shoot with a model, or beginning model I should say. Anyway, I did some small testing with the lens the night prior with some object sitting on my desk. They came out razor sharp. Anyway, after the shoot (I had to drive a total of 3 hours in and out of NYC), got home, uploaded them to computer and wouldn't you know, almost every shot was out of focus a little. The close in shots seemed OK, but that's about it. I focused using the center spot right on the models face each time, and each time her face was a little out of focus! ARRRRRGGGGGG!

The model has since reviewed the photos and doesn't see anything wrong with them, but believe me, they are out of focus. I am going to post a couple here a little later so you can see what I mean. I was absolutely mortified when I saw these photos, especially since this lens has such a great sharpness reputation.

For the most part, I shot in shade (about 1-2PM) F2.8 and F4, ISO 200. Some shots were a shutter speed of 1/90 while others were 1/200 with the built in flash. That didn't see to make too much of a difference in IQ.

Is it possible and I mean reasonably possible, for me to have gotten a lens with one of the elements crooked or something. I would think it wouldn't take too much to totally throw a lens out of whack.

Like I said, it seems to be the close focusing that works, but normal to distant focusing seems to be horrible with a capital HORRIBLE.

Regardless I am returning it tomorrow for another one. If this gives the same performance, then I must be a total moron and not shooting correctly. I would think auto focus with 1 center point wouldn't be rocket science, but maybe it is.

And when I say these images are soft, they are soft. You will see.

Anyway, I am in total psycho mode about this. I rescheduled the shoot with the model for the middle of the week and we will see what happens.

Again, I will post the images a little later.... :S

[ed rader]

Originally Posted by Double Negative

See, that's the weird thing.

I've bought five new L lenses and one used - every one of them is spot on.

maybe your standards are different?

i know copy variation exists and some lenses are more problematic than others but i do agree that user error is the main culprit.

i've never received a bad copy but my first L lens quit AFing after about 50 shots and had to be returned, my first DSLR (300d) gave me an err 99 right out of the box and had to be returned and my second battery grip (20d) kept prematurely draining batteries and had to be returned and repaired.

ed rader

[Double Negative]

Originally Posted by ed rader

maybe your standards are different?

Well I will say that I don't sit in front of the computer staring at every corner at 100% pixel-level detail. But I know when shots are sharp. If I PP something and don't need to apply any USM at all, or less than 80/0.3/0 then I'm pretty happy. So far, all my L glass has given me that. Much less USM is necessary than my old 35mm f/2.0, 50mm f/1.8 - or certainly my former Tamon 28-300mm.

I invite people to take a look at the shots I happen to have online (http://litpixel.com/ee/) and show me some "soft copies."

Maybe I'm just lucky - I don't know. I certainly don't doubt that there ARE bum copies out there. Of course flukes get through. It's the way of things.

[ed rader]

Originally Posted by Double Negative

Well I will say that I don't sit in front of the computer staring at every corner at 100% pixel-level detail. But I know when shots are sharp. If I PP something and don't need to apply any USM at all, or less than 80/0.3/0 then I'm pretty happy. So far, all my L glass has given me that. Much less USM is necessary than my old 35mm f/2.0, 50mm f/1.8 - or certainly my former Tamon 28-300mm.

I invite people to take a look at the shots I happen to have online (http://litpixel.com/ee/) and show me some "soft copies."

Maybe I'm just lucky - I don't know. I certainly don't doubt that there ARE bum copies out there. Of course flukes get through. It's the way of things.

some people actually do test lenses and a lens that might be acceptable to you may not be acceptable to someone else.

i don't test lenses, check them for dust or pixel peep and my experience mirrors yours but i do know that canon has serious QC issues as i pointed out in my first post so it stands to reason that not all copies are equal.

ed rader

[Double Negative]

Originally Posted by ed rader

some people actually do test lenses and a lens that might be acceptable to you may not be acceptable to someone else.

i don't test lenses, check them for dust or pixel peep and my experience mirrors yours but i do know that canon has serious QC issues as i pointed out in my first post so it stands to reason that not all copies are equal.

I can agree with all of that.

One day when I'm not too lazy I'll print out one of those test charts and shoot all my glass at it... Or some batteries. I just thought that shooting in real life was the important test.

But seriously, I have been meaning to shoot some charts just for the hell of it to see for myself.

[cosworth]

Originally Posted by Uhland

I think on the internet and especially on this forum people like to fault the user more so than the lens.

It's hard not to when someone posts a soft copy thread and the shot was taken hand held, on a dreary winter day, ISO 100, slow shutter speed and wide open.

Let's look at this again: ISO 100, slow shutter speed and wide open.

Ok, lets get the shutter speed up to 1/400th and shoot at say f/8. OH NO! I can take the picture because I can't go over ISO 200! There might be noise at ISO 200! Oh no no no.

Ok, get out the tripod. Mirror lock up, nice contrasty image, non-moving subject, say 1/250th, F/8 and resulting meterable ISO.

Then and only then will we see what is wrong with a lens.

Sure there are bad lenses out there. I've never had one. I've never seen one in action either. But they exist. And as mentioned, only bad ones get talked about.

[tfitz]

Originally Posted by cosworth

It's hard not to when someone posts a soft copy thread and the shot was taken hand held, on a dreary winter day, ISO 100, slow shutter speed and wide open.

Let's look at this again: ISO 100, slow shutter speed and wide open.

Ok, lets get the shutter speed up to 1/400th and shoot at say f/8. OH NO! I can take the picture because I can't go over ISO 200! There might be noise at ISO 200! Oh no no no.

Ok, get out the tripod. Mirror lock up, nice contrasty image, non-moving subject, say 1/250th, F/8 and resulting meterable ISO.

Then and only then will we see what is wrong with a lens.

Sure there are bad lenses out there. I've never had one. I've never seen one in action either. But they exist. And as mentioned, only bad ones get talked about.

Agreed, many people are obsessed with pixel peeping and finding the fault in their lenses. However, in saying that, the same is true with the skeptics; often people are dismissed as being an 'amateur' or a noob and not really understanding the capabilities of the lens and expecting too much. All of that is usually true, in one for or another, but I think a balanced approach on these threads is warranted.

For my situation, it was a "drop-off" in performance of the new lens with respect to what I am used to seeing in the same shooting circumstances. I have done a few tests, but nothing very scientific; mostly to just get a feel for how the lens was behaving.

I have to say, the lens has disappointed me in the past 2 days consistently enough to bring it back for a new copy (good thing I bought it locally). HAHA, again I find myself at my laptop with no samples to post, but I will.

[tfitz]

Here is a sample of photos I took while testing the lens. Not this is taken outside around 11-12PM today, overcast sky

Left image: Tamron 28-75/f2.8: ISO 200, f4, 50mm, 1/250
Right image: Canon 50/f1.8: ISO 200, f4, 50mm, 1/250

As you can see the image on the right is much sharper. This should, IMO, be a no-brainer shot, point and shoot, right? Or am I missing something here? All of the blurred shots seem to exhibit the same sort of "bluriness" in that they have some sort of abberation going on? What do you all see here?

[Double Negative]

Originally Posted by tfitz

As you can see the image on the right is much sharper.

You may very well have something funny going on there. Good thing you're swapping lenses - it'd be interesting to see how the new copy performs in comparison.

Just to add one point to the mix; the nifty, as cheap as it is - is a very sharp lens, part of why being that it's a prime. That'll beat a zoom nearly every time. Of course I'm not saying that explains your images but something to keep in mind (that the left shall rarely equal the right in this case). Not to mention that the prime is stopped down a lot more than the zoom (despite the aperture "being the same").

[kitacanon]

Originally Posted by Uhland

Only look at the 100% crops for examples of lenses and even on them I have been seeing a lot of sharpening and moderate post processing done to them. Why someone would post process a image meant to be an example goes beyond me but some do.

Canon's own instructions for processing RAW files recommend using moderate USM (300% at .03). A lens is only as good as the photographer who knows the relationship between the lens, the camera, and the processing.

[tfitz]

Originally Posted by kitacanon

Canon's own instructions for processing RAW files recommend using moderate USM (300% at .03). A lens is only as good as the photographer who knows the relationship between the lens, the camera, and the processing.

I do understand the relationship. Both of these images are RAW unprocessed, to include having the ACR sharpen turned town all the way. So in other words, those images are exactly what were recorded. I would understand the deficiency if it were consistent across all of the lenses I own, but it is not. It reveals itself in this one lens only, so there is some measure of a scientific method being adhered to. As I see it, the USM addition to a RAW image is meaningless to this particular test as long as all photos have a control reference or multiple control references like the image on the right.

[jdos2]

I too am of the ilk that often get bad gear out of the box. The latest photographic example was the Sigma 24-70 f/2.8. I sent it back to Sigma, and it's much better now, though I must say that my faith in using it is... Diminshed, in case I have to send it back again.
My 50mm f/1.2 backfocuses. At any photography at short distances (2 meters or less) I take two shots (moving my head back---really, just hunching my shoulders to move back about a cm) and peep the sharper one. Canon is aware of the trouble with my lens as well as my trepidation to return something to them which they "don't see any complaints." Since I personally know of several folk that have complained, Canon telling me they've not heard of any makes me both want to start a conference call, AND to strangle someone... I don't trust them enough to fix the trouble because they out-and-out lie.

Luckily, I've had plenty of good luck, too. 35L? Perfect (perfectly perfect: it's a delight!). 85 f/1.2 & f/1.2? Perfect 16-35mm f/2.8? "Good enough (focuses fine, but isn't Contax G2 21mm or Biogon sharp, like I'd like: come ON Canon! Give us some good wide lenses!) 70-200 f/2.8? Beautiful. 50mm f/1.4? Nice.

Not so the 28mm f/1.8, but that's a different story.

It's not just the Canon stuff with which I've had trouble. Many things complicated come broken. I've sent a Mac back after only a week of use, then again at a year (that's a 20 percent failure rate), and another at the year point (day before warranty!). Contax G2s as mentioned above? Every one has gone back (a couple bodies: more than once!) and that cool Contax 6x6 rangefinder? Back to Jurgen 5 times, the last one to stay.

The Leica MP and all lenses came in perfect condition. Left that way too, when I went to the 5D (and digital).

Yaesu VX-7 (amateur radio hand held) was the first version. I went through three before giving up for a couple years. The latest I have, though, is fine.

I am coming to believe that many products are simply released without someone actually looking at them to say "yay" or "nay." People are expensive, and machines cheap. I don't think that Canon would have shipped my 50L, though, if someone would have simply done a close shot to see that the focus is quite obviously behind where the camera "locks." That's about the same with all of this stuff that comes broken. If someone would have <i>used</i> it once or twice, the problems would have come out in the wash.

Some folks have different luck. Mine is certainly mixed. When I get something that "just works well" out of the box, I smile, remembering the trouble I've had with those that don't.

As for my 50mm f/1.2, when Canon get's their heads out of... Um... When Canon figures out the fix, I'll send it off to 'em. Until then, it's close to the 80/20 rule, which is "good enough," though I'd not recommend it as "flawless."

[SunTsu]

Originally Posted by dkangel

Not for me. You can ask my wife. I have a 98% failure rate on just about anything I buy. I can buy a new fan and it will not run. I can buy a new appliance and it will have missing pieces. Weird thing is I am not superstitious by nature being an engineer and all but after a while you start to wonder. In fact I just bought a scanner last week and all pictures I scanned had weird marks at the top of it. I cleaned the glass but it didnt help. Had to exchange it for another copy of the same model. After a while you get used to it.

I had a crappy day yesterday, so I'm going to throw in my two cents about "luck." I've had my 5D for about two months now and am pretty happy with it. What I haven't been so happy with is the lenses that I've purchased so far. When I first started shooting with the 24-105, I was thinking, "Hm...not as great as I would have though after dropping 5K in gear". Luckily, I had the option of asking a bunch of people on this board and some people told me it was OK and some people told me it might not be. So finally, I got another copy, did some testing on a tripod and the differences were obvious. I even showed images (fit to my screen) to my wife and she thought the differences were obvous at F4. At F8, at 100% crop, she could definitely see a difference. I use her as my litmus test because I know she will let me know if I'm being too picky.

After keeping my second copy of the 24-105, I did some tests with the second copy against my 100mm macro. Everyone has told me that the 100mm macro should be sharper because it's a prime, but it's not. It's not blury like my first 24-105mm, but it's definitely not as sharp as my second copy of the 24-105mm. Like others have said, people only complain about the bad things, so I know I'm just a sample, but from my limited experience, I'll tell anyone there are differences from copy to copy. I think what kind of irks me is that "L" lenses are supposed to be the gold standard. They cost a fortune (I could have bought the next three generations of my SD900 for the same price) and are supposed to be the best-in-class. For that kind of money and the expectation that Canon sets, I don't think that bad copies should be extremely rare (less than .5%).

I come from the manufacturing world and in contracts, there are what's known as major, medium and minor faults. A cosmetic scratch is minor. Major is something that affects the intended purpose of the product. IMO, super sharpness is something that is the intended purpose of a $$$ L lens so QA should be weeding out the copies that are not super sharp. This is admittedly, a total guess on my behalf, but I think that the number of copies that are not super sharp is more than 1%.

PS: dkangel, I'll post how bad my luck is in the Accessories forum