|
BetterLekNextTime posted:The current kit lenses from Canon aren't so bad, but the ones from when the t2i was current (I bought one in 2010 or 2011 I think) are probably best avoided- back then the Tamron/Sigma equivalents were even more strongly recommended compared to the kit offerings. So just make sure you don't get whatever came with the camera you are buying. I see that the latest 18-55mm has gone from f/3.5 to f/4. I’ve heard that this was done to make a more compact lens. I’m sure that stop-wise the practical difference is negligible, but it’s still not a trade I’d necessarily make. That lens was really light already, and it’s not like it was huge to begin with. Seems like an odd move to me, unless it was done for another reason.
|
#
?
Feb 20, 2018 03:18
|
|
|
|
| # ? Jun 1, 2024 10:05 |
|
|
hope and vaseline posted:I don't think there's any primes besides the 17mm ts-e that'll give you an equivalent of 28mm on a crop body. If you're not married to the idea of a prime, the 10-18mm on the long end will give you the equivalent. It's a very nice affordable ultra wide. Which I'll forever dream with, what a beast... BetterLekNextTime posted:The current kit lenses from Canon aren't so bad, but the ones from when the t2i was current (I bought one in 2010 or 2011 I think) are probably best avoided- back then the Tamron/Sigma equivalents were even more strongly recommended compared to the kit offerings. So just make sure you don't get whatever came with the camera you are buying. I'll take care, they're usually bundled with the older models. Helen Highwater posted:A really good walking about lens that covers wide-angle up to mild-tele ranges is the 17-50 ƒ/2.8. Theres a Tamron and a Sigma version that are pretty much identical and can probably be found for cheap used as the new price for either is around $300 or so. That's going to give you an equivalent 27-80mm FoV with a fixed aperture throughout the zoom range and image stabilisation. ƒ/2.8 isn't blazingly fast but it's faster than the kit lens can go and it's fine for 99% of genera purpose photography. Good tips. Tamron and Sigma aren't very easy to find around here, but I'll keep they in the choices. Talking Tamron: Thoughts about the Tamron 18-200mm F3.5-6.3? There's one T2i with this lens on the local equivalent of eBay, but I can't seem to find much about it online. Waiting on the seller to know if it's the old or the new, image-stabilized version.
|
|
#
?
Feb 20, 2018 03:42
|
|
|
Superzooms like that usually have terrible image quality. Grab the 55-250mm STM if you want more range, you should be able to find it super cheap used.
|
|
#
?
Feb 20, 2018 13:00
|
|
|
The 10-18 is a good bet if you want a 28mm equivalent. The 24/2.8 is maybe the best choice if you want a sharp prime for general use, though. But you already have a lot of good primes. I've come across several copies of the 18-55 IS II over the years and never found any of them to be very good.. it's not that they weren't sharp (they weren't, but that didn't matter), but that they couldn't manage to reproduce their middling in-focus sharpness consistently across the frame. There was always a softer side to the image, unless one stopped down to apertures where diffraction made everything evenly unsharp. Maybe the new STM 18-55s are better. They have new optics and new construction. Try before you buy, if you can. Just take a picture of a tree or something at f/5.6 and see if the results are good enough. IMO being picky about your lens' sample quality (if possible) is the best way to get value for money when you're on a budget. Some units of the same lens will be better than others. There aren't too many badly-designed lenses in production these days (although the sigma and Tamron superzooms are exceptions), it's more that variation in quality of assembly and optical flaws can be very high among the cheaper stuff. You're almost certainly going to get better uniformity and sharpness from the old yashica lenses, but there's nothing in your existing kit that will give you a particularly wide angle of view on a T2i. Just leaving the less common stuff out of consideration, considering your local market. Sigma and Tamron make some good kit lens alternatives, but I don't see them in stores, even here in the US. Disregarding all the more exotic and expensive stuff out there, if I had a T2i, 10-18, 55-250, and a bag of good yashica primes, I'd consider that a complete system. Basically everything I'd need. ...Does the T2i viewfinder have a focus confirmation light for use with manual-focus lenses? I guess that's one other thing to consider. Otherwise, you'll have to focus the MF lenses in live view for focus-critical stuff.
|
|
#
?
Feb 20, 2018 14:48
|
|
|
Just ordered me the sigma 17-70mm f2.8-4, really excited to replace my aging 18-50mm f2.8. It'll be nice to fill in the gap between that and the 70-200mm, which I use very rarely, but having the option to get to 70mm without changing lenses will be really nice for a standard zoom. Does anyone have any experience with the sigma dock for autofocus adjustments?
|
|
#
?
Feb 20, 2018 15:03
|
|
|
SMERSH Mouth posted:The 10-18 is a good bet if you want a 28mm equivalent. The 24/2.8 is maybe the best choice if you want a sharp prime for general use, though. But you already have a lot of good primes. One thing I found with the 24mm 2.8 pancake is that my Tamron 17-50mm 2.8 (non-IS version) actually has better edge sharpness at 24mm than the pancake does. That's not to say the pancake is a bad lens. Far from it, it's more that the Tamron is surprisingly good. The pancake simply has limitations because its a much more basic lens type and there is only so much correction you can pack into a tiny form factor. It has plenty going for it still: it's light, it's fast, the autofocus is nearly silent (definitely can't say that for the Tamron) and it makes the camera look a lot less obtrusive which is useful for street photography.
|
|
#
?
Feb 20, 2018 16:33
|
|
|
rolleyes posted:One thing I found with the 24mm 2.8 pancake is that my Tamron 17-50mm 2.8 (non-IS version) actually has better edge sharpness at 24mm than the pancake does. Woah. That's pretty cool. Really need to get me one...
|
|
#
?
Feb 20, 2018 16:37
|
|
|
President Beep posted:Woah. That's pretty cool. Really need to get me one... There's often variability in the quality of lenses, particularly 3rd party ones like Tamron, so I'm not sure this will be the same for everyone. Maybe I got a particularly good Tamron or a particularly bad pancake, although the non-IS 17-50 has a reputation for being very sharp for the cost. To be completely clear, by edge sharpness I mean "at the edges of the image" rather than edges in the image. At the centre the pancake is at least as good as the Tamron in my experience, often a bit better. All of this is a little bit pixel-peepy though, not sure how much it would affect 'real world' images. rolleyes fucked around with this message at 16:45 on Feb 20, 2018 |
|
#
?
Feb 20, 2018 16:43
|
|
|
I think on an extreme budget, either of canon's 18-55 IS kit lenses will do you fine. The newer the better, but the two versions out there are identical optically. The only difference is the mark 2 has slightly better IS. Their image quality is decent, and the lenses are dirt cheap. For telephoto, the 55-250 is very good too for the price. Its biggest weakness is focusing on moving objects (birds, cars, etc) but it has the same upsides: decent images, cheap, and IS. The differences between the original and the mk2 are a bit subtle.. the original is sharper at wide open at 55mm but the newer one is much better at 250mm (the gap closes as you shrink the aperture). Sigma and Tamron definitely make good lenses but the stuff people are mentioning are $300 and up. You can get both the 18-55 and 55-250 for under $300.
|
|
#
?
Feb 20, 2018 17:34
|
|
|
Primo Itch posted:
I have this lens (the non-IS version) because it came as the kit lens with my 70D. I'm not a fan of it and I've barely used it since I bought the camera. What was said earlier about superzooms is right. It's very soft until you stop it down quite a lot at all zoom ranges and it's very slow to begin with. I am by no means a pixel-peeper but I was consistently disappointed by the results I got. A lot of chromatic aberration, not terribly sharp and not fast enough for anything you'd want any kind of reach for. It is quite small and light though - smaller than the 17-50 or my ancient 100-300.
|
|
#
?
Feb 20, 2018 17:37
|
|
|
xzzy posted:For telephoto, the 55-250 is very good too for the price. Its biggest weakness is focusing on moving objects (birds, cars, etc) but it has the same upsides: decent images, cheap, and IS. The differences between the original and the mk2 are a bit subtle.. the original is sharper at wide open at 55mm but the newer one is much better at 250mm (the gap closes as you shrink the aperture). My experience backs this up. I've got a first generation 55-250, and it does get kind of fuzzy at the narrow end when shooting wide open. Still markedly better than the EF 75-300 I borrowed though.
|
|
#
?
Feb 20, 2018 17:39
|
|
|
Now that this discussion has devolved into gushing over low-end Canon gear (which is righteous and true), I've got one more thing to add. My sequence of crop-body ILC ownership has gone, XSi, a6000, E-M10, X-T2. All of them had some worthwhile positive features. The a6000 had the best fine detail reproduction, optics notwithstanding. The E-M10 was super useful and easy to take everywhere. The X-T2 has the best high-iso output of any APSC camera I've used, and it's known for having excellent color rendition... but when I look back at the library of images I've taken with all these, nothing I've owned has had better color output than the XSi. The XS-T5 series rebels, and the other Canon DSLRs from the same generations, are still the benchmark for color quality (not accuracy) in my estimation. You can kind of see this for yourself if you go to the dpreview studio scene tool and see how various cameras render the color swatch grid. The T1i and XSi are too old to be on the list, but the 5D or T5i will display the same basic effect. Check out the red square. No other popular cameras (including the new-sensor-generation 80D and M5) have such a vivid and true rendition of red. By comparison, a lot of sony-made sensors display red as more of a "dark pink" or slightly brownish hue; others seem to include a little too much blue or yellow. Additionally, the blue square to the left of the red is a little oversaturated in Canon land, but I think it helps add to a perceptual distinction between blues and reds that generally increases the subjective appeal of the image for most people. What's weird is that while Canon seems to be using the same basic approach to color (vivid blues and deep reds) in their newer (80D and onwards) cameras, something has changed in the color rendition. Reds don't look the same. I don't own a newer Canon camera, so I can't say from subjective experience that their color rendition has gotten worse. It just seems like it's changed, but I can't say for sure. Anyway, this is all just to say that the T2i is a great camera and is still among the best for color rendition (even across multiple monitors, Mac, Windows, and prints...although always processed in ACR or LR), and color rendition is more important, I think, for how most people judge a photo, than image noise, dynamic range, or to an extent, sharpness. Great, now I have to not buy an old Canon. SMERSH Mouth fucked around with this message at 01:25 on Feb 21, 2018 |
|
#
?
Feb 21, 2018 01:23
|
|
|
I couldn't see a clear difference in the comparisons on dpreview, but I do have a t1i and an 80d so sperged out to try it myself.   left is 80d, right is t1i. Used a nifty 50 at f8.0 at iso 200 for 1/15 on both photos. Only adjustments in lightroom were setting white balance to identical values and boosting exposure. It's subtle but I do agree the t1i has richer colors.. but it seems like the 80d picks up more detail, like the scratches in the red bricks. But there's some ugly patchy poo poo going on in the blues on the t1i, kinda looks like sensor noise.. which I guess makes sense, the t1i has poo poo low light performance and fewer megapixels. Also I spent way too long on this, setting up to get the same shot twice is super fussy and I still hosed it up.. I didn't feel like getting my lights out to do it properly. xzzy fucked around with this message at 04:46 on Feb 21, 2018 |
|
#
?
Feb 21, 2018 04:44
|
|
|
Thanks for doing that- wow. That's shooting jpeg, right?
|
|
#
?
Feb 21, 2018 14:30
|
|
|
No I shot raw, I thought that would be better to avoid any jpeg filters the camera uses. Is shooting jpeg a thing that's done for hardware comparisons?
|
|
#
?
Feb 21, 2018 14:58
|
|
|
You didn’t get the same exposure and white balance soooo
|
|
#
?
Feb 21, 2018 18:02
|
|
|
You're gonna have to explain that because I set both bodies to the same settings and the images were taken within a minute of each other in the exact same conditions. And the white balance settings are identical in LR.
|
|
#
?
Feb 21, 2018 18:32
|
|
|
I wonder if some of that has to do with Adobe RAW's camera profiles. I think a better comparison would be using JPEG and the neutral color profile of the camera, or maybe Raw and canon's actual DPP software.
|
|
#
?
Feb 21, 2018 18:44
|
|
|
That's interesting and inspired me to do my own test. My idea of Canon having 'different' color rendition comes from a gut feeling after paging through years worth of my lovely photos. I've never tried to do an objective comparison before. It's not helpful for judging the case of whether the newest Canon DSLRs have significantly different color rendition from the older models, but here are some examples of the difference in color rendition between an X-T2 and 5D2. Processed from RAW. ACR defaults. Color balance set to 4300K in-camera. ISO400, f/4, 1/2, desk lamp. Included the Fuji 18-55 and a Helios-81 in the XT-2 test to show the difference in color rendition between the two lenses.  Out-of-camera JPEGs. Same 4300K WB. Same exposure.  It's easy to see the difference. But isn't this all arbitrary and completely fixable in Photoshop? I don't know. Here's the result of my quick attempt at bringing the Fuji red closer in line with the Canon red, using the Camera Calibration menu in ACR. The starting point for both images was a custom WB setting in both cameras, each based on the same grey card.  I'm not sure that a perfect match is possible using this tool. Adjusting the appearance of red also changes the appearance of green.
|
|
#
?
Feb 22, 2018 05:12
|
|
|
It could be a slight variation in light from interior bulbs that cycle. Lots of fluorescent bulbs are at like 60hz or something
|
|
#
?
Feb 22, 2018 19:47
|
|
|
My enthusiasm for comparing hardware plummets to zero at the point it involves installing Canon software on my computer or coordinating the current cycles in my light bulbs.
|
|
#
?
Feb 22, 2018 20:15
|
|
|
Just Lmao if u think u can make it in photography without being willing to be a sperg
|
|
#
?
Feb 22, 2018 21:26
|
|
|
I haven't thought that for years, the fact that everyone hates my pictures confirmed it a long time ago.
|
|
#
?
Feb 22, 2018 21:44
|
|
|
There's no RAW or JPEG or lightbulbs or lenses or UFOs overhead affecting anything. Its just different sensors, different BIOSes. Or rather the hardcoded procedures that dictate how signals from the sensor are handled. All your settings and profiles are just adjustments to these default values. Go ahead and dump them from the camera motherboards and inspect the code line by line if you want. The truth is out there.
|
|
#
?
Feb 22, 2018 22:03
|
|
|
The raw file (DNs off the sensor) needs to have profiles applied to it, period. The filters in the mask don't have identical transmittances, the sensor doesn't respond to all wavelengths uniformly, the sensor isn't completely linear, there is a dark current, etc. Some dude at Canon makes profiles for JPEG, someone else at Fuji makes their profiles, someone else at Adobe tries to duplicate those profiles and make their own. The profiles for all cameras are different. The people making the profiles are trying to balance between producing something that is attractive, that can handle a reasonable gamut, and is somewhat realistic, and . At the end of the day you're collapsing a spectrum into 3 numbers and there is never going to be a single "correct" way to do that. You could spend a bunch of time and come up with a model that makes a Fuji look like a Canon or vice versa, or you could just realize that there is no truth in photography and correct things to what you think is right.
|
|
#
?
Feb 23, 2018 00:02
|
|
|
Re: RAW file interpretation, I saw this video a couple weeks ago and it might explain a bit about the color variations between files, cameras and editors. https://youtu.be/AZdzcaaqaeY
|
|
#
?
Feb 23, 2018 01:34
|
|
|
Yeah this is why comparing color differences off ACR which is a 3rd party software trying to approximate the camera's actual profiles is inherently an erroneous way to make hard-line comparisons with color fidelity. Plus add the fact that who actually outputs their image straight out of acr, everyone has their own workflows to achieve their own desired personal aesthetic, that I feel like making statements like, entry level rebels reproduce skin tone more faithfully than the XXD lineup doesn't really make much sense.
|
|
#
?
Feb 23, 2018 01:43
|
|
|
That really just demonstrates why if you want to sperg out and compare two bodies, you want to change as little as possible between the scene setup and the exported jpeg. Yes the results will look different, but it eliminates variables and makes the comparison less complex. Granted, it still doesn't "matter" because if one thought the colors from a T1i are "richer" I would argue they could get the same effect with an 80d by nudging saturation and luminance settings until the perfect look is found. But his point about reprocessing old images to get a new result is totally true, with some qualifications. I recently tore down my own self-hosted site and gave in to putting it all on flickr and ended up re-exporting 7 years worth of photos. Most of them I updated to the latest LR process and it did trigger a bunch of adjustments. Wasn't scientific about it but it feels like the change was mostly a saturation boost and needing less sharpening/noise reduction to get a nice result. The dehaze slider in particular is a magic button for getting more pop out of a midday landscape. Certainly not enough to make a bad exposure a good exposure, but it does seem like it gives a little more dynamic range to play with. If someone had a catalog of thousands of images spanning the years there's no way it'd be worth the effort to reprocess them all.
|
|
#
?
Feb 23, 2018 02:01
|
|
|
I really try to not abuse it, but I love the dehaze function.
|
|
#
?
Feb 23, 2018 02:11
|
|
|
Just to make sure it's clear, I'll add that I'm not saying anything about which cameras render color more 'faithfully' ...whatever that means. I'm making a subjective claim: Colors from a certain era of Canon sensors are are generally attractive because - by default, maybe through hardware design, maybe through software manipulation, probably by a combination of both; I don't really know - they reproduce reds in a particular way, at least when compared using the two most common methods of delivering final images from camera RAW data: Adobe ACR and out-of-camera jpegs. My only technical angle here is that they may do so in a way that can be 100% reproduced in images from other cameras, using standard consumer image manipulation software (ACR & LR), but I kind of doubt that they can. At least, my own feeble attempts to reproduce the look using images from a Fuji camera didn't work. Maybe a better way to approach the question, rather than dicking around in Photoshop and going, "what do you think guys?" like I did, would be to ask, "are all sensors basically the same? They're just wafers that measure light intensity overlaid with color filters, right?" An answer to which was posted above, and that's really what I was looking for. SMERSH Mouth fucked around with this message at 02:20 on Feb 23, 2018 |
|
#
?
Feb 23, 2018 02:18
|
|
|
Canon about to produce a decent mirrorless camera? https://www.canonrumors.com/canon-eos-m50-more-images-and-specifications/
|
|
#
?
Feb 23, 2018 07:23
|
|
|
TheAngryDrunk posted:Canon about to produce a decent mirrorless camera?
|
|
#
?
Feb 23, 2018 08:07
|
|
|
evil_bunnY posted:Now they just need to make it mirrorless size and we’re in business! But then where would they put the name? What’s the point of using a Canon if you can’t show it off??? 
|
|
#
?
Feb 23, 2018 08:24
|
|
|
TheAngryDrunk posted:Canon about to produce a decent mirrorless camera? The M5 is really nice actually. This has some cool upgrades, but loses a lot of the external controls - I'm sure the M5MkII or whatever they call it will be sick. Now if they would just release some lenses evil_bunnY posted:Now they just need to make it mirrorless size and we’re in business! The M5 is basically the size of an XT-20 , and this looks about the same. Although to get any lenses other than a 35mm eq and some slow zooms you need to bolt giant EF lenses onto the thing. timrenzi574 fucked around with this message at 15:43 on Feb 23, 2018 |
|
#
?
Feb 23, 2018 15:41
|
|
|
The M5 actually looks pretty cool and seems well made. But yeah, the lens selection. Anyway, if the M50 does 4K as rumored, that's a pretty big departure from Canon's market segmentation practices up to this point. They've kept it as a feature on their 'pro' level and video-specific cameras so far. But 4K is kind of the standard now, even though we still largely watch HD video and most consumers don't have the kind of storage or computational capacity to do a whole hell of a lot of 4K video recording and editing. Personally, though, I've had a lot of fun and so far gotten some very nice results, shooting 4K with my own mirrorless camera. It's just short clips, though, and downsampled to 1440p for final output. I'd think that Canon's implementation will be geared towards that kind of more casual use, i.e. limited recording times and fewer supporting features like you'd see in an A7S or gh5. I wonder what kind of image quality the 4K video will have (assuming it's actually included). It's funny how Canon used to be on the bleeding edge of this 'DSLR video' stuff a decade ago and now they're not only one of the last holdouts against 4K in mid-tier consumer cameras, but even their 1080p output looks like poo poo, with moire, aliasing, and generally poor detail, even on their latest generation cameras. I mean, in the world of streaming video it doesn't really matter that much, but it's just funny that they've lost the spec sheet wars so badly when they used to be at the top of the heap. So it will be nice if the m5's theoretical 4K video quality holds up against its competitors' better than its immediate predecessors did in the 1080p realm. Even if it doesn't, out of the box, I wonder if there will be some Magic Lantern exploits that improve things. Heh.
|
|
#
?
Feb 24, 2018 02:17
|
|
|
SMERSH Mouth posted:So it will be nice if the m5's theoretical 4K video quality holds up against its competitors' .... Ahhhahhaa.
|
|
#
?
Feb 24, 2018 02:50
|
|
|
It seems like a lot of people on YouTube and whatnot like to gush over what a great camera the 80D is for vlogging. Are they just full of poo poo?
|
|
#
?
Feb 24, 2018 03:30
|
|
|
They're not wrong necessarily. They probably also don't know what 4:1:1 compression is and probably don't care. It's great for people that don't know any better and just want to make little videos as we all probably have when starting out. Fine for Facebook and student work. It's just that almost every major player has surpassed a majority of Canon DSLRs in video. mrlego fucked around with this message at 09:27 on Feb 24, 2018 |
|
#
?
Feb 24, 2018 04:24
|
|
|
4:2:0 blaze it
|
|
#
?
Feb 24, 2018 05:14
|
|
|
|
| # ? Jun 1, 2024 10:05 |
|
|
If nothing else, doesn't dual pixel mean the 80D has some very good video autofocus performance? Combined with Canon's track record of making cameras that 'just work', I guess vloggings find that the 80D is a good experience even if other cameras offer 4K, better quality or more bells & whistles.
|
|
#
?
Feb 24, 2018 05:51
|
|
