Canon EOS T1i / 500D DSLR Review

[Bryan Carnathan]

Discuss theCanon EOS T1i / 500D DSLR Review- tell us what you think of the Canon EOS T1i / 500D.

[alexniedra]

Great review, Bryan!





Just one question, though:


Increasing sensor density lowers the DLA (Diffraction Limited Aperture)
by magnifying the diffraction effect - making it more visible.
This number is the result of a mathematical formula that approximates
the aperture where diffraction visibly starts affecting image sharpness
when viewed at 100% full-size on a display (which is what a high
percentage of site users are doing to evaluate their gear)


I can totally understand that flaws in the optics (like diffraction) are more visible within a 15 MP image than say, a 12 MP image. But, why are do we compare 100% crops from the T1i with say, an 12 MP Xsi image? Shouldn't we observe these crops at 80% with the T1i, and 100% with the Xsi?


The same goes for noise - Shouldn't crops be observed at 80%, and then compared?


Sorry if this stirs debate - I'm sure Daniel Browning will be on this soon[]

[Benjamin]

Thanks for bringing a thoughtful review of the T1i to us so quickly, it was great!


I'm surprised to see the IQ improvement too, it just makes me think "ahh, how nice that would be if my 50D can do the same...". I'm still satisfied with my 50D for what I use it for, and we can expect an even better 60D or whatever in the future with less noise now []

[Jon Ruyle]

Another excellent review. I'm sort of tempted by the T1i for macro/astro (manual focus uses in which I often can't use a whole full frame field anyway).


What a long way the rebel line has come from my old (and much loved) XT.

I can totally understand that flaws in the optics (like diffraction)

I may well be misunderstanding you, alexniedra, but diffraction is not a flaw in optics. It is just a fact of life that the image of a point through a lens is not a point but a small disk (not really a disk but a diffraction pattern, but might as well think of it as a disk). The size of the disk on the ccd is proportional to the f number. That is, double the f number and the diameter of the disk doubles. When the disk is bigger than a pixel, you notice it. When pixels are smaller, you notice the disk sooner, and DLA is lower.


So, if diffraction is a flaw, it is a flaw in the Universe, not in the lens.


As much as I like the review and respect Bryan for his attention to detail, I mislike the following choice of words:

Increasing sensor density lowers the DLA (Diffraction Limited Aperture)
by magnifying the diffraction effect - making it more visible.

Perhaps it would be better to say "revealing" instead of "magnifying". Increasing sensor density does not change diffraciton, it only makes it easier to see. I know that seems picky, but there seems to be a misconception about this.


Sorry- Bryan already said he had given more attention to DLA than was warrented, and I agree he covered the subject sufficiently. Yet here I am talking more about it

[JesseY]

I've been looking to get into photography and have already committed to the EF-S 17-55 2.8 lens, so I've been waiting for quite some time for the reviews to hit for the T1i. Unfortunately, things are at 2 extremes right now.


Bryan, your review paints a pretty good picture of the camera -- good image quality, good noise handling, a durable shutter, and no mention of overly odd white balance or dynamic range issues. The only low point, which is unfortunate but probably not a deal breaker for me, is the slow AF system when taking continuous shots. Movie mode quirks notwithstanding, this seems like an excellent camera for the money. The mobility it offers is attractive as well.


However, another review came online today over at digitalcamerainfo.com. They paint a drastically different picture of the camera, particularly in areas that I care a lot about -- image sharpness,dynamic range, and white balance . Their resolution measurements were very, verypoor anddynamic range fell off drastically starting at ISO 400 and above.


They were routinely getting very soft images and things were so bad they asked Canon for a second camera which continued to perform below par. Now they did say they tried multiple lens but I'm not sure which ones they tried.


I still would reallylike the T1i as the 50d is sadly just out of my price range after dropping the cash for the expensive lens. Would you or someone else with a T1i be able to post sample pictures? Also, might it be possible to see how the camera performs on the ISO 12233 Crops when compared to say the 50d when using the same lens?

[Jon Ruyle]

JesseY, I was a little confused by the digitalcamerainfo.com review. They kept saying "the camera" wasn't sharp, but they're testing it with the kit lens. Do they mean to say the *lens* isn't sharp?


Of course, dynamic range is a different matter. I'm a little surprised by (and suspicious of) their result.


My rule of thumb is, when in doubt- Bryan is right. The other reviewer is wrong. []

[STL]

However, another review came online today over at digitalcamerainfo.com. They paint a drastically different picture of the camera, particularly in areas that I care a lot about -- image sharpness,dynamic range, and white balance . Their resolution measurements were very, verypoor anddynamic range fell off drastically starting at ISO 400 and above.

I would be interested to know at which apertures (f-number) they did their tests. As the T1i has a Diffraction Aperture Limit of f/7.6, testing the camera even at f/8 would reduce sharpness due to diffraction. On a side note : Canon's Digital Photo Professional seems to apply default noise reduction since a few versions. As the sharpness and the contrast both decrease when noise reduction is added, the issue they noticed in their review could come from this.


As a XTi owner, I think I'm gonna pass this one and maybe go with a used 40D when I'm gonna be able to afford it. The DLA is an issue for me, as I do macro shots and fireworks photography, which requires apertures between f/8 and f/11.


I think that it would be a good idea for Canon (and other SLR manufacturers) to stop cramming more pixels in sensors, as even though your picture is bigger, the increased diffraction can be a serious problem if you require a big depth of field. They should try to produce full-frame sensors at a lower cost instead, as the photosites are bigger than a 1.6 crop factor body for the same resolution.

[Jon Ruyle]

I would be interested to know at which apertures (f-number) they did their tests. As the T1i has a Diffraction Aperture Limit of f/7.6, testing the camera even at f/8 would reduce sharpness due to diffraction.

They did their tests at a variety of f numbers. The measurement of sharpness was number lines resolved across the whole picture. Diffraction is not an issue here.

I think that it would be a good idea for Canon (and other SLR manufacturers) to stop cramming more pixels in sensors, as even though your picture is bigger, the increased diffraction can be a serious problem if you require a big depth of field.

Since they tested lines per whole picture, more pixels could only do better, not worse, than fewer. Diffraction is not changed by increased resolution. Diffraction gets worse only when you stop down.

As a XTi owner, I think I'm gonna pass this one and maybe go with a used 40D when I'm gonna be able to afford it. The DLA is an issue for me, as I do macro shots and fireworks photography, which requires apertures between f/8 and f/11.

Macro would be one of the few cases in which I would think the new rebel would be superior to the 40D. It has better low noise performance and better resolution, both of which are important in macro since macro requires lots of light and since one tends to crop macro pictures. You'll be using manual focus most of the time, and probably won't care too much about super fast frame rate or responsiveness (thus much of the advantage of the 40d is negated). The rebel's superior screen will be of use for manual focus with live view.

[Bryan Carnathan]

Hi Everyone,


Alexniedra - The purpose of the 100% crops is to show you exactly what you get from the camera. I can add the XSi results scaled up to the T1i's pixel dimensions if that is wanted. They will be a little softer than the T1i results, but not dramatically different. Scaled down T1i results would be sharper.


Benjamin - Right - What will the 60D bring us? Likely full frame rate 1080p HD video at least.


Jon - Thanks! I'll change that word.


I'm looking to remove ALL confusion over DLA. I first included it to help people understand why their images were soft - to answer the question "Why is my f/22 photo soft even when I used a $2,000 L lens?" (to take the extreme example). I fear I might have still caused misunderstanding for some.


Jesse - That is a great lens - good choice in my opinion.


I'm not seeing any T1i image attributes that noticeably stray from the previous Canon bodies. Canon's auto white balance under tungsten has never been good, but it is easy to change the selected white balance in-camera or in-post. Though I didn't take a scientific measurement, I didn't notice any significant change in dynamic range. You can look for differences in the color block example.


The lens used for the testing can make a huge difference in the results. I have both the 50D and the T1i results loaded for the Canon EF 200mm f/2 L IS USM Lens (link to comparison) in the ISO 12233 Chart tool, but I discovered that one of the 4 flashes was not firing for the T1i test. I'm going to be replacing these results soon.


There are a lot of other factors that can make big difference in image quality. Shooting RAW can be one of them. Noise reduction is another. Sharpening is another (the ISO 12233 chart results have very little sharpening applied). Comparing one brand to other further complicates the process.


STL - While your plan is not a bad one, DLA will probably only be an issue for you if you are reviewing your narrow aperture results at 100% on screen or printing to a huge size. And even so, the T1i will give you more detail to work with. I think a low-cost full frame body would be a big seller. A refreshed 5D (add sensor cleaning, Live View ...) would be very popular.


Regards,


Bryan

[Jarhead5811]

I think I'll wait until the next Rebel comes out and get a T1i when it's in the $500.00 range. Then my XSi would be an excellent second body. I like that the batteries are the same on the two models.