Sigma 10-20 - Undecided - 100% crop pic

swag72

swag72

Suspended / Banned
Messages
7,969
Name
Sara
Edit My Images
Yes
Undecided whether I think this is an acceptable copy of this lens - Kind of erring towards not.

Here's a pic with a 100% crop below. Taken with tripod and remote - Focus was on the bridge pretty much in the middle by the keystone.



100% crop, focussed as described above - Just seems very soft to me, expected better



The whole lot just seems soft, soft, soft

Help required apart from crit on the pic!!!!!
 
seems fairly soft to me... certainly much softer than my Tokina 12-24mm, not sure how it's supposed to compare to that.
 
Yeah i would have to say it looks soft to me, i would like to think it would be sharper than this.
 
You just had to pixel-peep didn't you ......:nono:



Get your money back, hire the canon from stewart & do a comparison shot otherwise you'll always wonder "is the Sigma soft / have I wasted money on the Canon"
 
I think you've focused somewhere ahead of the bridge.. I reckon you've focused on the blades on grass at the "front" of the frame?

The bridge is soft because its not your focus point.
 
I had a similar problem, I'm on my phone so can't link you but if you fancy filtering through my posts I've got my 100% crop which you could compare it with? :)
 
puddleduck said:
I think you've focused somewhere ahead of the bridge.. I reckon you've focused on the blades on grass at the "front" of the frame?

The bridge is soft because its not your focus point.
Click to expand...

Definately focused on the bridge key stone as I did it in live view on x10 magnification using MF - My eyesight aint that bad either!!!!

Just to add, it was at F8 at 20mm
 
swag72 said:
Definately focused on the bridge key stone as I did it in live view on x10 magnification using MF - My eyesight aint that bad either!!!!
Click to expand...


Hmmm.. well its out of focus in any event, so its no point looking at 100% crops.
 
One thing i will say is i am not a fan of the live view for it focus, i always feel it's off dont no why but i have noticed this in a few shots. I now only use it for family snap shots.
 
Looks about as soft as mine. Not sure that helps much, but not sure you are going to get a hell of a lot sharper.
 
Mine is a bit soft everywhere - I found that photoshop 'smart sharpen' 100%, 1 pixel helps a lot.
Is anywhere sharp in your pics (ie is the lens just focussing at the wrong place)? Are they soft everywhere or just on one side / corner etc?
I've put some full res unsharpened crops of a test shot on photobucket. They're at http://s252.photobucket.com/albums/hh40/parsleywok/
A series of crops and then the downsized original. (Landscape, not the moon.)
 
Is there somewhere you can link to a full size uncropped version?

It looks to me like there may be a tiny bit of motion blur. Also, did you shoot RAW and has this been processed at all? Ohh, and what's your exif details please? :)
 
no probs with mine, thus said I have just had a prism put on't glasses to cure double vision :D:D


showphoto.php
[/url][/IMG]
 
Exif data would be handy to see, also try autofocus and not MF to see what results you get this time around.
 
It's soft compared to what? TBH it's not a very revealing test unless you have got a directly comparable image, processed in exactly the same way, from a good lens of known sharpness.

If you want to see if you have a good copy or not, shoot at lowest f/number and check the corners. If they are equally sharp, or quite likely equally unsharp, then the lens is almost certainly a good copy. If they are unequal, the lens is not properly centred, the most common manufacturing defect or damage problem. Checking the centre of any decent lens, at f/8, will give very similar results.

I would not condemn the Sigma on the basis of what you have shown.
 
This was taken in Raw with no sharpening either in camera (neutral setting) or in CS3. There has been no processing at all. The exif is 1/250, F8, ISO 100, 20mm.

I have a 17-55 which is crackingly sharp, and I just feel that this would not cut it when compared to this. I guess I will need to do a direct comparison on something to see how it goes. Alternatively, I just return to Jessops and buy the Canon like I had planned to anyway!!

When you say 'compared to what' Rich, that is really compared to how sharp I would expect the pics to be in the first place to be honest.

Will do a test this evening in my garden compared to my 17-55. Thanks for your help so far.
 
Isn't it odd that a few different copies of the same lens can differ so much......

Mine is very sharp indeed, even wide open F5.6 @ 20mm.
(But then I bought it from Puddleduck, so wouldn't expect anything else :D )

Could you send it to Sigma for 'repair'?
 
Take the size of the bridge on first shot = 40 * 20 mm. Crop it and print out = 165 * 70mm. The first is 800 square mm the second is 165 * 70 mm = 11550 square mm. Therefore you have enlarged it 14.5 times. Did you expect it to be very sharp.
PS must say *** the first shot is poor.
 
In the old days the write ups on lens was just the same A reviewer would say "This is a very good lens", another would say about the same lens "like the bottom of a bottle". I thought I had the same trouble with a Tamron 10-24 lens.
 
I know that I shouldn't pixel peep, but I was so disapointed in the original picture, I had no choice!! And I was apprehensive about getting the Siggy in the first place, it's probably all just in my head now and I'll never think it's as sharp as a Canon would've been.
 
swag72 said:
This was taken in Raw with no sharpening either in camera (neutral setting) or in CS3. There has been no processing at all. The exif is 1/250, F8, ISO 100, 20mm.
Click to expand...

Well if there's been neither capture, creative nor output sharpening applied to the image then wouldn't it look rather soft anyway?

Once sharpened it looks acceptable, if not razor sharp, to me (although sharpening the 100% Jpeg crop rather than the Raw file).

Unsharpened:

100_-crop-bridge.jpg


Sharpened:

 
I agree with Deckard. I recently picked up a 10-20 and whilst playing around with it noticed the lack of sharpness and the incredible difference between images captured using "Mountain Mode" and AV mode in which I normally shoot. Obviously in the preset mode the image is a fully processed jpeg. On the other hand the raw image was, by comparison, terrible even when all settings where identical and the focus was identical.

I still can't fully decide whether to go raw or jpeg. In raw there is latitude for adjustment but the post-process overhead can be immense.

As a matter of interest why don't you try the comparison of a "Mountain Mode" and Av raw mode image of the same shot. After the first shot, switch your lens to MF so no change to focus can happen. You may be surprised at the result.

Russ
 
There have been some soft copy issues with this lens for a long time.
Luminous Landscape tested this lens against the Canon 10-22 and their test copy was very soft.
I would take a few test shots and if it still appears soft return it as not fit for purpose, or at least have it recalibrated.
Davol
 
OT but I'd go Raw any day mate. Using DPP Raw processing can be as simple as selecting all then clicking batch process [having applied a picture style or "recipe"] if you don't want to spend you life in PP, but it gives you the latitude for exposure/WB etc if you desire.
 
swag72 said:
This was taken in Raw with no sharpening either in camera (neutral setting) or in CS3. There has been no processing at all. The exif is 1/250, F8, ISO 100, 20mm.

I have a 17-55 which is crackingly sharp, and I just feel that this would not cut it when compared to this. I guess I will need to do a direct comparison on something to see how it goes. Alternatively, I just return to Jessops and buy the Canon like I had planned to anyway!!

When you say 'compared to what' Rich, that is really compared to how sharp I would expect the pics to be in the first place to be honest.

Will do a test this evening in my garden compared to my 17-55. Thanks for your help so far.
Click to expand...

Sure, you might have a duff copy but TBH I am skeptical of these reports. Folks buy a Sigma because it's cheaper, so reckon it can't be as good, so go pixel peeping for problems and, unsurprisingly, they find some. Mainly because they go looking for them, and don't set up a fair comparison. And please don't focus test it using that silly sheet of paper shot at 45 degrees :eek:

The difference between cheap lenses and expensive ones is rarely in central sharpness, and never at a mid range aperture like f/8. A guy on here recently compared a 16-35L against a kit lens and was pretty shocked when there was absolutely no difference at the centre when stopped down a bit. If you want to be picky, then look at the edges and corners, at lowest f/number, compare distortion and vignetting, see how they handle flare. Top lenses are also much better in terms of construction quality too, but it has to be said that the law of diminishing returns kicks in hard quite early.

I know you really rate your 17-55 2.8, as I do :) So shoot with that side by side at 18mm or so, f/number for f/number. I'm sure you can focus okay ;) and you've got loads of DoF there anyway, but use live view 10x again. Run them out using your normal PP routine.

But to cut a long game short, if you can find the cash, why not go for the Canon 10-22? It is of known high quality and (this is the unique bit) you can make it even better by treating it to the lens correction suite in Canon's DPP. Any distortion and vignetting is gone, and also most residual CA. I think it's a brilliant feature, and it will surely be incorporated into the next generation of cameras (Nikon already has something similar) but you'll not be able to do that with Sigma. (TBH you can do something similar with a programme called PTLens with a lot of third party brands, but it's not as good, and it will never become in-camera.)

The other wide zoom I like the look of is the Tokina 11-16 2.8. If you believe Ken Rockwell and most others, and I tend to, it's pretty damn good and nothing else does f/2.8. Of course what you lose is zoom range, too much for me, but that is a different kind of trade off than between the normal same-but-cheaper options.

Lokking forward to your results ;)
 
Rich, you talk too much sense!! Jessops contacted, I will return the Siggy on Friday and they've kept back a 10-22 Canon for me!! That will teach me for changing my mind!

If I don't do this I'll never be happy and always think that I've made a mistake.
 
hope you get a good lens either way, i have the sigma 10-20 and if i saw a pic like that i would assume i kicked the tripod or something, it is more than soft, it is blurry.
i must have got a good copy.
 
swag72 said:
If I don't do this I'll never be happy and always think that I've made a mistake.
Click to expand...

Could you repeat the shot when you get the Canon with 100% identical settings then post the results ??


It would be good to get a "real world" comparison :clap:





don't look at the results, if it's not £300 better than the sigma then :'(
 
I have a tamron 11-18mm that is super sharp, second to my sigma 105mm

But my sigma 17-70mm is really soft, and has a lot of ghosting at f2.8. I have emailed sigma today to finally try and get it fixed because it really isn't acceptable. Sigma should be tighter on their production of lenses because i'm always hearing of 'bad copies' and it sucks to own one
 
swag72 said:
Rich, you talk too much sense!! Jessops contacted, I will return the Siggy on Friday and they've kept back a 10-22 Canon for me!! That will teach me for changing my mind!

If I don't do this I'll never be happy and always think that I've made a mistake.
Click to expand...

I think I'll frame that quote Sarah, and use it as my sig :D

Your last line is so true. I just bought a 70-200L, yet I know that for 99% of the stuff I do, the sweet lil' EF-S 55-250 IS would do just as well, at less than a quarter of the cost. But it was a treat for me, and I just wanted the best I could get. The fact that I can't afford it and had to sell four other lenses to get it, makes it even better for some strange reason. Using nice stuff is so much part of the process :)
 
paleblue said:
Sigma should be tighter on their production of lenses because i'm always hearing of 'bad copies' and it sucks to own one
Click to expand...

without wanting to stray off topic too far, in contrast ive read quite a few stories of canon QC being off form also.

i agree with the above whoever said that people assume that as the sigmas are cheaper they must be lower quality, pixel peep looking for issues and inevitably find "problems".

i guess i must be lucky that all 3 of my sigmas are spot on :shrug:
 
neil_g said:
without wanting to stray off topic too far, in contrast ive read quite a few stories of canon QC being off form also.

i agree with the above whoever said that people assume that as the sigmas are cheaper they must be lower quality, pixel peep looking for issues and inevitably find "problems".

i guess i must be lucky that all 3 of my sigmas are spot on :shrug:
Click to expand...

Not off topic at all, and it was me that said when you go looking for problems you usually find them.

And in my experience when you're talking about quality control issues it is usually people's test procedures that are off form more often then the lenses.

I would like to hear more about them, and a few facts. Folks are likely to take unsubstantiated rumours as true. What are these "quite a few rumours"? Where did you read them? Exactly what was wrong, with which lenses, how were the problems discovered and what was done about them? If something was actually wrong with an expensive product, presumably people would have had it looked in to and fixed.

If you can give me any links, I'll see what I can find out, but honestly I know that in nine out of ten cases it is user error. Of course, no manufactured product is ever 100% perfect all the time, but if Canon's quality control has slipped badly, I think Canon would like to know as much as anybody.

I always check my lenses, usually before purchase and often of multiple copies. I do that because I used to believe internet rumours were true. I've bought more than a dozen lenses in the last few years and checked about 30 of various brands. Not a single duff copy. My last one was a Canon, three weeks ago. It's perfect.
 
hoppy - i cant remember exact details, various different forums etc. my point (probably badly made as usual) was really that like you say when people look hard enough for an issue they find problems with any lens including canon whether there is one or not.
 
Hope i'm not hijacking this thread, but just a quick test I've done to back up my claim:

sigmatest.jpg


First half was done manual focus. Second was autofocus (pressed halfway numerous times until confident it was was in focus), with the centre focus point selected aimed directly on the lock (the circle). Both are 100% crops, camera on tripod, 2 second timer. exif for both pictures: 17mm f2.8 iso200 1/200.

I know many will claim this as 'human error' somehow, but i realised when i first got the lens that the focusing wasn't right, and it reflects in the pictures when compared with my other lenses.
 
paleblue said:
Hope i'm not hijacking this thread, but just a quick test I've done to back up my claim:

<pic removed>

First half was done manual focus. Second was autofocus (pressed halfway numerous times until confident it was was in focus), with the centre focus point selected aimed directly on the lock (the circle). Both are 100% crops, camera on tripod, 2 second timer. exif for both pictures: 17mm f2.8 iso200 1/200.

I know many will claim this as 'human error' somehow, but i realised when i first got the lens that the focusing wasn't right, and it reflects in the pictures when compared with my other lenses.
Click to expand...

That's fine paleblue, reference your post #32 I presume, and your Sigma 17-70mm.

But you said the lens was soft, which it clearly isn't - it's misfocusing. Could be a camera fault from what you've shown, but assuming your other lenses are okay, then the lens requires adjustment.

Of course it shouldn't need that and you have every right to be miffed, but it's not the same as a 'soft copy' which is never properly sharp and seems to be a fact of life with a few lenses these days, thankfully rarely, which is why many cameras now have microadjustment for users to custom calibrate their kit. It's a two minute job for a technician to fix, but I guess you have to send it back to get it done. Maybe the camera also, just to be sure, which is a major pain :(
 
HoppyUK said:
That's fine paleblue, reference your post #32 I presume, and your Sigma 17-70mm.

But you said the lens was soft, which it clearly isn't - it's misfocusing. Could be a camera fault from what you've shown, but assuming your other lenses are okay, then the lens requires adjustment.

Of course it shouldn't need that and you have every right to be miffed, but it's not the same as a 'soft copy' which is never properly sharp and seems to be a fact of life with a few lenses these days, thankfully rarely, which is why many cameras now have microadjustment for users to custom calibrate their kit. It's a two minute job for a technician to fix, but I guess you have to send it back to get it done. Maybe the camera also, just to be sure, which is a major pain :(
Click to expand...

Yes sorry, I didn't mean a soft copy :bonk:. I do own a 450d which had this problem reported with focusing in some reviews, but as I said my other lenses seem fine. I think I will do a more in-depth test with other lenses just to be sure.
I emailed sigma and they said it will be a 2-3 week turnaround, and can be 30-90quid depending on the problem I have a few jobs to do so can't wait that long :bang: I may just sell it and get an L lens and be broke for a while.

Thanks for taking the time to answer and provide some insight :)
 
paleblue said:
Yes sorry, I didn't mean a soft copy :bonk:. I do own a 450d which had this problem reported with focusing in some reviews, but as I said my other lenses seem fine. I think I will do a more in-depth test with other lenses just to be sure.
I emailed sigma and they said it will be a 2-3 week turnaround, and can be 30-90quid depending on the problem I have a few jobs to do so can't wait that long :bang: I may just sell it and get an L lens and be broke for a while.

Thanks for taking the time to answer and provide some insight :)
Click to expand...

You're very welcome.

It's a bit harsh to get stung for that, for a problem that has existed since the day the lens left the factory. I always check these things as soon as I get something new, and so far I've never had a problem.

It's a damn nuisance, maybe you bought it used and it's out of warranty and that's why it got sold. Maybe the next owner will have the same problem... ? Sorry to lay that one on you, but there it is.
 
You must log in or register to reply here.
Back
Top