Lens test?

Lostkat · 28 Feb 2006 at 20:57

I've just bought a Tamron SP AF 17-35mm Tamron F2.8/4 LD with part of my yearly bonus to replace my Canon 300D kit lens. Ideally, I would have liked the Canon 17-40L, but I just couldn't justify the cost, so I went for this one instead.

I've read many excellent reviews. It's supposed to be a vast improvement on the kit lens, and an all round sharp lens. The only issue seems to be Tamron's variable quality, which means that getting a sharp one is a bit hit and miss. I spoke to the guy in the shop about my concerns, and as long as I return the lens in perfect condition, I can get a refund.

Anyway, I would like some advice on how to put this lens through its paces and compare it to the kit lens. I've taken a few shots indoors, but the lighting is so poor that the photos look terrible on both lenses. I'll probably need to do it outside. Any suggestions/links on what shots I can take to test various aspects of the lenses - sharpness/CA etc.?

Thanks in advance

dod · 28 Feb 2006 at 21:06

for focussing have a look at the chart here
http://www.photo.net/learn/focustest/

JBuk · 28 Feb 2006 at 21:06

i was just going to post a very similar question

ive got a 20d & 18-55 kit lense and ive just bought a 50mm f1.8
ive taken a few quick shots of the same subject at f20 15secs on both lenses , the 50mm 1.8 looks about as soft as the kit lense (should it be better?) or am i testing badly?

Sleepyd · 28 Feb 2006 at 21:22

JBuk said:
ive taken a few quick shots of the same subject at f20 15secs on both lenses , the 50mm 1.8 looks about as soft as the kit lense (should it be better?) or am i testing badly?

There's a world of difference between the kit and the 50mm no doubt about it - try shooting a real-life subject at F5.6 with both and see the difference.

jhmaeng · 28 Feb 2006 at 21:26

JBuk said:
i was just going to post a very similar question

ive got a 20d & 18-55 kit lense and ive just bought a 50mm f1.8
ive taken a few quick shots of the same subject at f20 15secs on both lenses , the 50mm 1.8 looks about as soft as the kit lense (should it be better?) or am i testing badly?

f/20 introduces diffraction which makes the image softer.

Most lenses on a DSLR tend to be sharpest at around f/8 or so - try testing at f/5.6 - f/8 range. Where you should notice the most difference between the 50mm and kit lens is actually somewhere between f/3.5 and f/5.6, because the 50mm f/1.8 should be nice and sharp by that range whereas the kit lens will still be relatively soft until you stop down to f/8 or so.

Sleepyd · 28 Feb 2006 at 21:27

I saw this on another forum - best use of the kit lens:

Truth be told it's not too bad at F8

SDK^ · 28 Feb 2006 at 22:01

Lostkat said:
Anyway, I would like some advice on how to put this lens through its paces and compare it to the kit lens. I've taken a few shots indoors, but the lighting is so poor that the photos look terrible on both lenses. I'll probably need to do it outside. Any suggestions/links on what shots I can take to test various aspects of the lenses - sharpness/CA etc.?

If you really have to do the test tonight then setup some good indoor lighting and photograph real objects at F2.8, F4, F5.6 and F8. But they are wide angle lenses so ideally I would wait until tomorrow and test with landscape shots

Also, as you are testing sharpness you should be using a tripod, mirror lockup with self timer.

Lostkat · 28 Feb 2006 at 22:19

SDK^ said:
If you really have to do the test tonight then setup some good indoor lighting and photograph real objects at F2.8, F4, F5.6 and F8. But they are wide angle lenses so ideally I would until tomorrow and test with landscape shots

Also, as you are testing sharpness you should be using a tripod, mirror lockup with self timer.

Yeah, I knew this was the best solution, but was just being impatient

Just going to have to wait for daylight. I can't be bothered to set up the studio lights just for a test, and I will be using it for outdoors shots most, so I should test it as such. Mirror lock-up, tripod etc. goes without saying. Thanks for talking sense

Will also test on the chart dod linked to.

Thanks for the comments/advice guys.

Sleepyd · 28 Feb 2006 at 22:31

SDK^ said:
mirror lockup

300D

My 2p:
Sharpness: I test first wide open against a group of detailed objects that are roughly the same distance away(50-60Feet in my case) to negate any depth of field considerations. There are some dockyards near my work which have boats lined up uniformly across one wall - they have writing on and small details such as rigging and blemishes that are on the paintwork which give a clear indication of what the centre and edge sharpness are like as they go across the frame bottom(and top if you re-compose) of the frame. I'd probably then test what the sharpness is like from 2.8-8. Dont go below 1/100th of a second to be safe.

CA: Tree branches / electricity lines against sky

I think those are the main concern with the Tamron.

You might also like to know that the 300D allows about +0.6EV worth of over exposure in the RAW files - you often find by reducing the EV by around 2/3rds of a stop it reveals the most detail in my experience.

Hope that helps

Lostkat · 28 Feb 2006 at 22:44

Sleepyd said:
300D

WASIA

Thanks for the tips

I was trying to think of things I could photograph in the garden with sufficient detail. There's not really much in the immediate area that I can do though. I might have to venture further afield. Perhaps down to the canal where there are the barges which have signwriting etc. on.

Trees against a light sky was already in my plan, as it's what I've had issues on with the kit lens before.

Samiad · 28 Feb 2006 at 22:56

You can test (crudely) against my test image :

This is at 35mm. I have only used one Tamron 17-35mm, but I believe mine represents 'a good copy' because I can't see how it could be much sharper.

edit - I should mention this is a straight RAW conversion, zero sharpening applied whatsoever.

morgan · 1 Mar 2006 at 08:20

JBuk said:
i was just going to post a very similar question

ive got a 20d & 18-55 kit lense and ive just bought a 50mm f1.8
ive taken a few quick shots of the same subject at f20 15secs on both lenses , the 50mm 1.8 looks about as soft as the kit lense (should it be better?) or am i testing badly?

if you think the best way to do a lens test is to use a 15 second exposure then I'd say the last thing you want to worry about is lens softness...jesus :rolleyes:

mods I think we need a beginners FAQ which gives answers to regular basic questions. this forum is going down the pan.

Samiad · 1 Mar 2006 at 11:32

Lol at "this forum is going down the pan"

Sleepyd · 1 Mar 2006 at 11:42

Agree.

Question 1: I want to buy a digital camera for the price of a cup of tea. What would you advise? I want at least 8MP and lots of manual functions.

Ahh we we're all beginners at somepoint.

AndyBorzi · 1 Mar 2006 at 11:51

morgan said:
if you think the best way to do a lens test is to use a 15 second exposure then I'd say the last thing you want to worry about is lens softness...jesus

mods I think we need a beginners FAQ which gives answers to regular basic questions. this forum is going down the pan.

Think you maybe bringing it down with comments like that!

JBuk · 1 Mar 2006 at 15:24

morgan said:
if you think the best way to do a lens test is to use a 15 second exposure then I'd say the last thing you want to worry about is lens softness...jesus

mods I think we need a beginners FAQ which gives answers to regular basic questions. this forum is going down the pan.

I asked a question , i did not say anywhere that i thought 15 secs was "the best way" , as i dont have any idea on how to properly compare 2 different lenses

Maybe an FAQ would be a good idea for beginners like myself, then I wouldnt have to ask questions that are obviously so beneath you

Lostkat · 1 Mar 2006 at 17:37

OK, enough of the bickering please.

morgan said:
mods I think we need a beginners FAQ which gives answers to regular basic questions.

What an excellent idea. It would add on nicely to the SLR buyers guide in the FAQ. I take it that you're volunteering yourself as composer for this?

matty · 1 Mar 2006 at 18:09

we toyed with the idea of a beginners guide, but that would take away a lot from the forum IMHO, i think it would here too. No question is a stupid question if someone learns something from it!

divine` · 1 Mar 2006 at 18:16

no one would read it anyway.

Lostkat · 1 Mar 2006 at 18:29

matty said:
we toyed with the idea of a beginners guide, but that would take away a lot from the forum IMHO, i think it would here too. No question is a stupid question if someone learns something from it!

I agree with you matty. I would hate to be on a forum where you get flack and rolleyes for asking a "silly" question.

I also disagree with the comment that these forums are going downhill. I have always found this forum to be populated by a huge mix of helpful people. Never before have I seen someone roll their eyes at a genuine question regarding lens testing until now. It saddens me somewhat.