Camera sackrider question

[DrSmile]

So I got the Canon XSI to try out for the school, very nice screen in the back, but I wouldn't recommend it for the serious photographer because it has limited settings (esp focus and exposure) and worst of all no top LCD. I had a heck of a time getting it to shoot sharp at F32 even with a top of the line 60mm Canon macro, I had to step it up to F16 to get decent results. My old Fuji S2 was wicked sharp even at F38 so I'm wondering why this doesn't work with Canon. I tried it on the Rebel and Rebel XTI bodies with the same results. Does this have to do with the CMOS sensor?

All pics were at 1/200 with lots of light from an adjustable flash and the camera set on manual.

[Nate Crisman]

Shitty lenses have lots of diffraction at small apertures, pretty often you will find a kit lens that is really sharp and makes nice pictures at F8 and F11, but is utter dogshit at F22 and smaller.

What lens are you using on this Cannon (Gasp) :wink:

I thought the XT, XTI and XSI (the rebel line) were CCD sensors and the CMOS was on the 30/40D 5D and up pro bodies?

[DrSmile]

It's a 12.2-megapixel CMOS Sensor. The lens is a Canon F2.8 EF-S 60mm Ultrasonic Macro, it's the best you can get for that focal length. I also tried the Sigma and Tamron (100 and 90mm) with the same results.

[Nate Crisman]

So you have used 3 different macro lenses and the whole Rebel line and can't seem to make sharp pictures at small apertures.

hrm I don't know what to tell you at all. Im sure you are used to macro work, so I doubt user error!

I have never heard anything about canon cmos sensors and issues with sharpness or macro focus.

[blcknspo0ln]

examples?

[DrSmile]

After doing some reading, apparently this is indeed due to diffraction at very small apertures. I don't know whether it's the lens design of the Canon vs Nikon or if it has to do with the way the sensor works (The Fuji SuperCCD is octagonal) , but there seems to be a big difference.

I'll get some samples tomorrow.

[DrSmile]

Here are comparison pics, Fuji S2 and Canon XSI with identical ring flashes and 60mm Canon/Nikkor macro lenses shot at manual 1/125 or 1/250 shutter speed flash sync. Included is a 1:1 crop of the Letter "D" on the pen (which is about 3mm in height). Both sensors are downsampling from 12MP to 6:

Fuji S2 at F19, you can see the depth of field issue requiring a higher F stop

http://mysite.verizon.net/johe/S2F19.JPG

Fuji S2 at F32, improved depth of field, slightly softer but acceptable

http://mysite.verizon.net/johe/S2F38.JPG

XSI at F32, good depth of field but SOFT (camera has internal sharpening set on high)

http://mysite.verizon.net/johe/XSIF32.JPG

XSI at F16, much better sharpness but unacceptable depth of field

http://mysite.verizon.net/johe/XSIF16.JPG

If you view the two middle ones side by side the difference is obvious.

[satanpez]

Not enough orange. Your white balance is off...

-Steve

[DrSmile]

Not enough orange. Your white balance is off...

-Steve

Remind me to kick you really hard in the nuts next time I see you Utah style!

[Hinz]

your monitor isn't calibrated correctly.

heheheh, you have a fuzzy "D"

[Nate Crisman]

wow...I honestly thought you were nit-picking at first without seeing any pictures, but that is pretty major.

Are you manually focusing? Using a tripod?

If your using autofocus....maybe your cannon lens/body has a backfocusing or frontfocusing issue. I can send you a file that you print out and shoot at max aperture @ 45* angle that you then look at your picture and measure the amount of focus error in your body/lens combo. (if you had a proper Nikon D3 or D300 you could input a custom focus point offset to make it PERFECT for that body/lens)

Then again, you said you used different canon bodies and 3 diff lenses...so :?

[Nate Crisman]

Not enough orange. Your white balance is off...

-Steve

Remind me to kick you really hard in the nuts next time I see you Utah style!

Make sure he's holding something expensive while getting hit nuts kicked so he can't even fall to the ground dramaticly either. :lol:

[DrSmile]

My first thought was that maybe the lens was front or back focusing. But I focus locked the camera and there is nowhere in the image where it is sharp at all. It's quite annoying.

[Nate Crisman]

HRMM..solution can the cannon stuff, go back to nikon. :shock:

[DrSmile]

I'm sticking with my S2, but the department needs something lighter for the "dainty" residents. The S2 is a paperweight.

[Nate Crisman]

I'm sticking with my S2, but the department needs something lighter for the "dainty" residents. The S2 is a paperweight.

Good thing they didn't buy D3s (think is a lead brick). They would have been so disappointed.

[NitrousPete]

I assume you need the entire image in focus and that is why you are stopping down. You will never get full DOF by stopping down, especially at macro ranges. You need movement. From here, you have two options: get a large format camera (not exactly small, portable, or easy to use) or one of these... http://www.lensbabies.com/

A much more expensive option is this: http://www.bhphotovideo.com/c/product/1 ... _TS_E.html (http://www.bhphotovideo.com/c/product/12132-USA/Canon_2536A004_Normal_Tilt_Shift_TS_E.html)

[Nate Crisman]

Yeah, obviously a view camera or a TS macro lens is the optimal answer.

Shiempflug ftw (I just love saying that).

But a lensbaby shouldn't even be in this conversation. That lens has about the optical quality of duct taping magnifying glass lenses to your camera. Utter dogshit as I have seen. They are designed to put things OUT of focus, not for aligning focal distances with the film plane. Thus they don't have really any quality control or advanced optical design,

Still, DrSmile shoudn't be seeing these soft images from that gear. and we aren't talkin about the area of the photograph that is out of focus due to shallow depth of field, we are talking about soft detail at the point of focus. None of his images are sharp at all.

[NitrousPete]

Nate, I've never really played much with a lensbaby. For $300 you would expect some kind of optical quality, even though it is more of an "artistic" lens than anything. I'm actually making a ghetto style one out of a 50mm lens I got on fleabay for $3. Looking back it doesn't have much film coverage and I probably should have found a cheap medium format lens, but as I only intend to use it for macro stuff that should help.

The Canon images look front focused compared to the Fuji. I wouldn't say the Canon is any less sharp at F16 than the Fuji is at F19, but the "D" is much closer to being in the center of the DOF on the Fuji. I wouldn't expect any DSLR to be sharp at F32, just way too much diffraction. It's possible he also has a bad sample of the Canon 60mm macro. I recently bought a brand spanking new 80-200 2.8 that looked like a Holga over 150mm. Swapping it out for a 70-200VR seemed to fix the problem :twisted:

Also, don't forget the Fuji is shoothing through far superior Nikon glass ;)

Anyway enough of this pixel peeping shit, this isn't dpreview.com. Is that your D200 for sale on Craigslist in Blairstown?

[Nate Crisman]

Nate, I've never really played much with a lensbaby. For $300 you would expect some kind of optical quality, even though it is more of an "artistic" lens than anything. I'm actually making a ghetto style one out of a 50mm lens I got on fleabay for $3. Looking back it doesn't have much film coverage and I probably should have found a cheap medium format lens, but as I only intend to use it for macro stuff that should help.

The Canon images look front focused compared to the Fuji. I wouldn't say the Canon is any less sharp at F16 than the Fuji is at F19, but the "D" is much closer to being in the center of the DOF on the Fuji. I wouldn't expect any DSLR to be sharp at F32, just way too much diffraction. It's possible he also has a bad sample of the Canon 60mm macro. I recently bought a brand spanking new 80-200 2.8 that looked like a Holga over 150mm. Swapping it out for a 70-200VR seemed to fix the problem :twisted:

Also, don't forget the Fuji is shoothing through far superior Nikon glass ;)

Anyway enough of this pixel peeping shit, this isn't dpreview.com. Is that your D200 for sale on Craigslist in Blairstown?

Heh, for $300 on a new lensbaby, your getting a $15 piece of glass, $50 of mount and brackets, and $235 of marketing expense. It really is a piece of shit, Iv used it (my old boss had one), it's a TOY lens at best. That company makes one single $300 item that no one really needs, yet has a full page ad in every photo magazine every single month.

Yes, I think "something" is not right with Rene's canon setup, either a front focus issue, bad QC soft lens?

Yes, my second D200 is on craigslist. Im picking up a D3 on Tuesday, it would never come out of my case again anyway :twisted: Then taking the hacksaw to my 10.5 fisheye :twisted:

[blcknspo0ln]

trade for D80 plus cash ? :lol:

[NitrousPete]

I think the D3 will be a great choice for you Nate. Doing indoor weddings in dim churches and crappy reception halls just begs for a high ISO king like the D3. My next SLR camera will probably be full frame too, an F5 ;)

One thing I have heard about AF systems, and Canon's in particular, is the AF sensors are much wider than the focus brackets. Apparently this was done to allow for a focusing screen that is not exactly centered. When doing a shot on an angle like shown, the focus sensor could be picking up on something at the extreme right of the sensor, which would be in front of the intended focus area. I always do focus tests with new lenses straight on, put the camara on manual focus, then move the test subject back and forward slightly without disturbing the camera to make sure the lens is focusing as desired. Don't even ask me about the (well used) 300 2.8 that I owned for less than 24 hours... horrible front focusing problem!

Why hack up the fish when the D3 has DX crop mode?

[Nate Crisman]

I think the D3 will be a great choice for you Nate. Doing indoor weddings in dim churches and crappy reception halls just begs for a high ISO king like the D3. My next SLR camera will probably be full frame too, an F5 ;)

One thing I have heard about AF systems, and Canon's in particular, is the AF sensors are much wider than the focus brackets. Apparently this was done to allow for a focusing screen that is not exactly centered. When doing a shot on an angle like shown, the focus sensor could be picking up on something at the extreme right of the sensor, which would be in front of the intended focus area. I always do focus tests with new lenses straight on, put the camara on manual focus, then move the test subject back and forward slightly without disturbing the camera to make sure the lens is focusing as desired. Don't even ask me about the (well used) 300 2.8 that I owned for less than 24 hours... horrible front focusing problem!

Why hack up the fish when the D3 has DX crop mode?

Yeah, while Im all hot for clean ISO 3200, usable 6400, and 12,800 doable for B&W, there are lots of other reasons I decided to step up to the top instead of getting 300s. Dual CF cards might keep me from getting sued in event of a card failure/corruption. Im going to shoot alot of pro motocross this summer and the 9fps will help me ALOT for that. And it's just a stronger body, rated for 350,000 clicks before it needs a rebuild. D300's are only 100,000. (I put 24,000 in 7 months, only 2 weddings). At the rate I shoot, a full season of weddings, Id be replacing 2 D300's every year and they would be sketchy quicker. (like driving around a 200k mile 1g that you know is going to break one of these days)

D3 in DX crop mode the 10.5 fisheye makes a 5 megapixel shot that looks exactly like on a D200....about 180* diagnal corner to corner. To be expected.

Put the D3 into FX mode with the fisheye "stock" and you get a werido image that the edges are all the lens hood.

If you hacksaw off the plastic lens hood fins, the lens is actually a bit wider, about 200* and makes a cool circular image.

Chopping off the hood gives you an option in fisheye, you can have a diagnal fisheye and a circular fisheye at the same time. I shop some FX mode (35mm film) last week with all my DX lenses to see how they would work on the D3 in 12mp FX mode.

18-200 is useless in FX, it makes a huge black vignette at every zoom setting

17-55 is suprisingly good. no vignette from 24-55mm and from 17-24 the vignette is not solid. This is a DX lens that is totally useable on an FX body. No reason to shoot 5mp DX mode when you can use it in 12mp FX mode. I usually add a vignette to nearly every picture anyway :lol:

The 10.5 dx fisheye was the really cool one, as I said, it becomes 2 lenses in one after cutting off the hood. Not like those little fins did much for flare anyway, just have to be careful not to set the lens down on the glass with no protection from the fins.

[DrSmile]

Sorry this isn't a front focus or a "bad sample" lens problem. We're talking about all the cameras in the department (we have 8). They all have dedicated macro ONLY lenses (Canon, Nikon, Tamron, Sigma) and the results are consistent. I spent a whole day taking sample pics, with me manually setting the focus in different position. A tilt lens is useless for my application as there is no way to get the lens inside the mouth so I'm stuck with a large F stop. The Canons just produce crappy results below F16. Oh well. Despite this the residents want the XSI so they will get it.

I would disagree that the D is more in the center of the focal trough on the Fuji at F19 compared to the Canon at F16. They are very similar and point (center) focused on the same spot on the pen. You can see the Fuji in fact is MORE front focused than the Canon in that picture. The problem with the Canon is that NOWHERE in the image is anything 100% sharp.

[Nate Crisman]

Sorry this isn't a front focus or a "bad sample" lens problem. We're talking about all the cameras in the department (we have 8). They all have dedicated macro ONLY lenses (Canon, Nikon, Tamron, Sigma) and the results are consistent. I spent a whole day taking sample pics, with me manually setting the focus in different position. A tilt lens is useless for my application as there is no way to get the lens inside the mouth so I'm stuck with a large F stop. The Canons just produce crappy results below F16. Oh well. Despite this the residents want the XSI so they will get it.

I would disagree that the D is more in the center of the focal trough on the Fuji at F19 compared to the Canon at F16. They are very similar and point (center) focused on the same spot on the pen. You can see the Fuji in fact is MORE front focused than the Canon in that picture. The problem with the Canon is that NOWHERE in the image is anything 100% sharp.


Ill take some pics with my 100mm macro at every aperture setting just to see what effect the diffraction has on sharpness. Be an intersting side note to see if the D200 sensor has the same issue of ultimate sharpness having a bell curve with regard to aperture