can anyone help me figure out these test results?

[jann1033]

trying to decide if i have just gotten way pickier( or maybe just figured any lousy photos were my fault before) before i send this back to canon ( 28-135 IS)

i did a focus test
at f8 /70mm/tripod/ all focus points used. the focus is basically the same with IS on and off( normal according to canon, it's more to protect the is motor than iq), slightly sharper on but definitely sharper to one side...less sharp side is not as close so could be "tester stupidity" err "error" i mean.. that is the same for f5.6 except there both are noticeable softer.,,expected due to wide open for 70mm... sorry the scan stinks but you can get the idea

so it would seem like the focus is OK?

so next to the IS..canon guy said take it in low light with IS on and off which I did, and the difference between the two is clear so the IS is working....( can't seem to find that file but it is dramatically different)

this leads me to the inevitable although disheartening conclusion that my lens is just a hunk of junk, softer than i realized ( since at f8 the focus seems sharp-ish not tack sharp though imo)or maybe by some fluke some photos appear sharper to me( or uhumm ie maybe "beauty" in the eye of the beholder?) but the actual lens is working....anyone else come to any other conclusions? if I'm right no sense humiliating myself any farther by sending my lens back again to canon....( I figure that's what this site is for, self humiliation )
Thanks

 

[extreme8]

How far are you away from the focus target?
Assuming 3 feet, @ 70mm & f8, your limits for acceptable sharpness will be less than an inch on each side of your focus point.

 

[Groucho]

I confused, why a scan if you're taking digital photos?

I dunno. $400 is awfully cheap for an IS lens - the cheapest one they have, from a quick glance. After that it's $550, then $1,050, and it keeps going up from there. I would have to guess that a fairly large percentage of that $400 is going for the IS system rather than the optics. Is there an equivalent non-IS lens?

Still, the user reviews at BH are generally pretty positive, the main downsides seem to be slow focusing and iffy construction rather than image quality. There are multiple stories of lenses having to be returned to Canon, though, so I guess that's not that uncommon a problem.

You would probably disagree with the reviewer who wrote "Canon Tech Support is second to none!" though.

I'm sure a lot of it is expectations, though. Someone coming from a PnS would probably be overjoyed. Someone coming from using primes or $1k+ lenses would probably be unimpressed. As expected, some reviewers talk about how sharp it is, some talk about it being a bit soft, some say it's heavy, some say it's light...

At this point, I would guess that you're unlikely to ever be totally happy with the lens... I'd send it back to Canon one more time, since they're willing to look at it again, and if it doesn't make you overjoyed when you get it back - sell it and buy a lens that you will be happy using. Otherwise, any subpar photo you get with it, you'll always be wondering if it's you or the lens...

 

[jann1033]

I confused, why a scan if you're taking digital photos?

I dunno. $400 is awfully cheap for an IS lens - the cheapest one they have, from a quick glance. After that it's $550, then $1,050, and it keeps going up from there. I would have to guess that a fairly large percentage of that $400 is going for the IS system rather than the optics. Is there an equivalent non-IS lens?

Still, the user reviews at BH are generally pretty positive, the main downsides seem to be slow focusing and iffy construction rather than image quality. There are multiple stories of lenses having to be returned to Canon, though, so I guess that's not that uncommon a problem.

You would probably disagree with the reviewer who wrote "Canon Tech Support is second to none!" though.

I'm sure a lot of it is expectations, though. Someone coming from a PnS would probably be overjoyed. Someone coming from using primes or $1k+ lenses would probably be unimpressed. As expected, some reviewers talk about how sharp it is, some talk about it being a bit soft, some say it's heavy, some say it's light...

At this point, I would guess that you're unlikely to ever be totally happy with the lens... I'd send it back to Canon one more time, since they're willing to look at it again, and if it doesn't make you overjoyed when you get it back - sell it and buy a lens that you will be happy using. Otherwise, any subpar photo you get with it, you'll always be wondering if it's you or the lens...

i I erased them as i went along so i could be sure to keep track of the right ones and printed them straight from my printer hence had to scan the test i printed.yeah i know, not the smartest thing i ever did rotfl2: unfortunately probably not the dumbest either)


it is an old lens as well, introduced in the 1990s ( I just read that and was shocked) and has the old IS so that might account for it being cheaper. i don't think there is a comparable canon at that range...a sigma around that same range is about $230-250 ish i think so i guess that would be the closest( and that might be 18-125)

i'm just wondering what else could be wrong if the focus and IS are ok

the instructions for the test said to get as close as you could where it would still focus and fill the frame as much as you could with the chart....the center point of my camera was right on the "focus here" words but must not have been completely centered( should have counted the letters i guess to be totally accurate but didn't really think of it) since the paper is obviously not centered. the last focus point was on the "mm" of the right column so i would think should have been in focus since the chart is only approx. 4 " wide total. maybe i'll do it yet again

is it possible for a lens to be sharper on one side than the other? if not I have no ideas...

actually the 2 tech support and 2 customer support people were really helpful...it's more the technicians( or who ever from that dept calls people) that i think aren't...i can practically picture them when talking to me on the phone,, as in or or or maybe even .. it's a toss up who is more "helpful" canon or dell

 

[gruZ]

Jann,

It looks pretty good to me - but a few questions / suggestions:

Was the camera at a 45 degree angle to the piece of paper? That does make a difference. Also, I would try various apertures - open it wide up and you'll have the shortest DOF with that lens, and should be able to tell more readily if it is center, front, or back focusing.

One more suggestion - instead of having all focal points active, just use the center. Not only is the center point more accurate (it picks up contrast both horizontally and vertically to focus on), but there is also zero guess work as to which focal point the camera picked.

Lastly, were you using Mirror lock up and either a cable release or the self timer? Again, the object is to remove all potentially confounding variables, like camera shake, and only test the lens.

Have fun,

Scott

 

[jann1033]

Jann,

It looks pretty good to me - but a few questions / suggestions:

Was the camera at a 45 degree angle to the piece of paper? That does make a difference. Also, I would try various apertures - open it wide up and you'll have the shortest DOF with that lens, and should be able to tell more readily if it is center, front, or back focusing.

One more suggestion - instead of having all focal points active, just use the center. Not only is the center point more accurate (it picks up contrast both horizontally and vertically to focus on), but there is also zero guess work as to which focal point the camera picked.

Lastly, were you using Mirror lock up and either a cable release or the self timer? Again, the object is to remove all potentially confounding variables, like camera shake, and only test the lens.

Have fun,

Scott

i did the 45 angle, the wide open( 5.6) and then also where i think it is probably sharpest(8) i forgot the remote although i would think the whole thing would be blurry but maybe not...I'll try it tomorrow with the center point and remote and see if it changes anything

 

[boBQuincy]

As much as you are probably tired of doing tests, I would say one more may still be worthwhile. Many of Canon's focus sensors are about three times the size of what is shown on the screen, so it's easy to focus on some point other than what was intended.

One test that I liked was to use a single object (like a AA battery) for focusing, then add more at 45 degrees once the focus is locked. This way there is no question of what the camera was focused on.
The self timer works well for this. Use a piece of paper to mark where the objects should be placed and it should be easy enough to do in the 10 second countdown.

Some viewer programs (like BreezeBrowser) show which focus point was used, so you may be able to tell what was supposed to be in focus from that.

If the lens still doesn't focus well and give sharp images, it's probably time to give up on it. Canon just doesn't seem to get the concept of real customer service...

 

[jann1033]

well i redid the tests: tripod timer center point only, centered on "s" of focus here line and the results are(drum roll)
f5.6( wide open)

f8

these are both with the is on, the difference between is on and off is minuscule at best
showing the is is working however, you can figure out which is which

now to me and on the sheets i printed out it seems the 5.6 is very much not focused and could be focusing a little in front of where it should since the 15/64 and 3/8 seem the sharpest to me( below the text line) so i am just going to send these in with the lens instead of sending photos...looking at them online the difference is not as great , the text line is really hard to read on the copy for the 5.6 s, compared to the 8,s the f8 sharper from 10mm-10mm the f5.6 sharpest but still not really sharp from 6mm-10mm
i also tried bob's test and it appears to me about the same although i should have used the same identical batteries turned the same way for easier comparison...some day i might learn....don't hold your breath

 

[0bli0]

are you doing your test on a tripod? it should be done on a tripod with the chart properly at 45 degrees and the centre of the lens at the same level as the focal point. since you should be using a tripod, you should also turn IS off as th IS on the 28-135 is not compatible with tripods. it is important to have the lens level and the focus target at proper angle in relation to the sensor.

note f/8 and f/5.6 can mask front/back focus problems due to the DOF. i've said it before... baiscally i'm not a big fan of focus tests. if you're not doing them exactly right, you can't be assured of the results and then you tend to interpret the results as you've felt the lens is behaving. i highly recommend taking photos outside under natural light with a real world subject (kids work great for this). do so under a variety of conditions but it's best to focus with a single spot and ensure you have a shutter speed appropriate to the lens.