1. Reminder: Please use our affiliate links for holiday shopping!

Not sure if this is a fair comparison (E-PL1 and XSi)?

Discussion in 'Other Systems' started by Hyubie, Nov 23, 2011.

  1. Hyubie

    Hyubie Unique like everyone else

    Oct 15, 2010
    Massachusetts
    Herbert
    When I upgraded from P&S, I have not heard yet of :43:, so I jumped to the DSLR wagon via a Canon XSi. I then added some very well reviewed zooms and the nifty fifty, but really I treated it still like a P&S. The IQ was definitely better, but not in the level I see in a lot of images on the web. Not until I got an E-PL1 did I get more "serious", educating myself about composition, aperture, ISO, post-processing, etc etc etc, and the images got even better.

    Anyway, I held on to the XSi but have not used it ever since. However, I thought about it again today, and made a very simple test to see what I might be missing. (I haven't seen this specific comparison between the E-PL1 and the XSi before, too.)

    I used the nifty fifty on the XSi, and the 45mm on the E-PL1. I know the price difference between the two lenses is astronomical, but the 50mm is very well reviewed, and I was thinking the difference between the sensors should make up for it a little bit.

    (Plus, I don't plan on adding anymore to the Canon gear - I might even reduce - so I'm stuck with the setup nearest each other.)

    I shot both @f/2.8, in RAW, then used AutoTone and Auto WB in Lightroom.

    Canon:
    IMG_8648.

    E-PL1:
    PB234101.

    Conclusions:
    - Yes, I definitely miss the XSI's quick AF, even compared with the G3.
    - Colors are definitely richer on the E-PL1
    - Could it be true that the XSi is noisier (though I'm not a pixel-peeper by any means!)?
    - Is the XSi is a bit warmer? Although it doesn't matter much to me, I tend to like warmer images anyway.
    - I am not missing much since making the switch!!

    But the big question is, did I do a fairly just comparison? Or is there a really really big difference between the Canon 50mm and the Olympus 45mm so as to render my conclusions as unreliable?

    (Please let's not bring up :43: cannot do sports/action/what have you's - I just want to make a simple comparison.)
     
    • Like Like x 1
  2. Ned

    Ned Mu-43 Legend

    Jul 18, 2010
    Alberta, Canada
    No, I think your comparison was fair. We're looking at complete systems here, not just a lens or a body. As far as the XSi being noisier, to my recollection it should be, yes. However, it's also a significantly older camera so this is only natural.
     
    • Like Like x 1
  3. WT21

    WT21 Mu-43 Hall of Famer

    Feb 19, 2010
    Boston
    Since m43, I've also owned an Xti, two XSi and two 5D. (Can you tell I had a hard time giving up my DSLR).

    There is a qualitative difference in shooting a DSLR than shooting a pen -- in how you hold it, responsiveness, feel of the click, etc. Neither is better. They are just different.

    The XSi will be less noisy in base ISO and slightly less noisy in high ISO. The Pen will come our sharper. Color is up to your. Did you shoot RAW or jpg? The oly jpg will likely come out looking better, but the Canon RAW has a LOT more room to work in post, and the tonal gradients and changes you can do to Canon RAW outstrips Oly.

    Having said that, they are very close, and it's only really when you go FF (as with the 5D) that the differences are very stark, and not just minor.

    I didn't keep my EP3, but once I tried it, I KNEW the responsiveness of m43 would match or meet DSLRs, so that became a non-issue. If an EP4 (or whatever) ever comes out with a built in EVF, then that will help tremendously.

    I'm not convinced, after a lot of testing between my DSLRs and also NEX cameras, that APS-C really gives much of an edge over m43. Yes a little one, but not huge. I'm going to look at what Fuji does in the spring, but I really think the only way to threaten m43 is through a FF compact mirrorless.
     
    • Like Like x 1
  4. Art

    Art Mu-43 All-Pro

    Mar 13, 2011
    San Francisco, CA
    It is a fair comparison. Very similar eq. FL (80mm vs 90mm) and aperture.
    I was shooting with T1i and nifty fifty recently along with E-PM1 & Oly 45. T1i files are clean at low ISO. E-PM1 has quite a bit of shadow noise at base ISO but it's not seen at normal viewing sizes. Nifty fifty is unusable at f1.8. I found f2.8 is good for portraits and still shallow DoF. Oly has much better bokeh. You really need to watch your background with nifty or you'll get bubbles reminiscent of cheap CCTV lenses.
    T1i has very nice skin tonal gradations and its raw files respond to some serious PP in LR w/o apparent quality degradation. T1i's artificial light AWB and presets totally suck (embarassing I'd say). E-PM1 produces much more natural colors indoors.
    Overall I think it's much easier to achieve desired look via live view (which is also quite embarassing on T1i, more of a joke really). By using live view I know exactly what I'm getting ( WB, real-time DoF, exposure, color modes, face detect, etc.). I can't see consumer DSLR surving for much longer. E-PM1+Oly 45 is MUCH smaller and lighter than T1i+ nifty albeit more expensive. It's weird and unnatural shooting portraits via viewfinder losing eye contact with the subject. Welcome to the 21st century.
     
    • Like Like x 1
  5. WT21

    WT21 Mu-43 Hall of Famer

    Feb 19, 2010
    Boston
    Agree with Art 100%

    One note, though. The nifty is a COMPLETELY different lens on FF than crop. It's quite magical on a 5D IMO

    I'll try to post some examples later.

    Sent from my iPhone using Mu-43 App
     
    • Like Like x 1
  6. I think it's a very fair comparison, albeit with the XSi (450D) being about a year older. In fact I think that the 450D sensor is the closest in philosophy to the E-PL1 amongst Canon DSLRs. I actually have had a few 450Ds, one as my main camera for about half a year and I think maybe two others that I bought and resold because I wanted the lens/es that they came with. I recently was looking through the images that I took with them (there weren't that many - it was a lean period) and I've basically decided that I want to sell my 500D (T1i) and go back to the 450D. I like the 15MP sensor in the 50D but somehow I don't like it so much in the smaller body. The 12MP sensor in the 450D was reputed to have a very light AA filter like the current 12MP m4/3 cameras and you can see that in it's resolving power. It is by no means a noise-free sensor but I think that it has less noise than the E-PL1 in RAW. I find the 450D noise quite pleasant and if required it cleans up very nicely. I also generally feel that you can "work" the files from ther Canons harder than a m4/3 file in PP. And yes, the 50/1.8 can give 'orrible bokeh - pentagons, anyone?
     
    • Like Like x 1
  7. WT21

    WT21 Mu-43 Hall of Famer

    Feb 19, 2010
    Boston
    IMO, the T1i is the weakest of the Rebels. It has more MP, which greatly upped the RAW size, but with very little benefit. The T2i actually has a better sensor, justifying the larger file sizes.

    Some say the XTi sensor was better than the XSi and the T1i, because it had a bit more DR, and the RAW file sizes were smaller (12 bit instead of 14 bit), but without any noticeable loss.
     
    • Like Like x 1
  8. timallenphoto

    timallenphoto Mu-43 Veteran

    266
    May 20, 2011
    Kent, England
    My now very old 30D image quality is way better than my E-P1 but of course the Pen has other things in its favour.
     
    • Like Like x 1
  9. Agree on the T1i as I noted above. I don't know if it is something to do with focus accuracy compared to the 50D (since they theoretically share the same sensor) but having just taken the 500D (T1i) away on holiday for a week I don't have the same level of faith in it as I do the 50D. I don't have much experience with the XTi (400D). It's a model that I have owned briefly in the process of buying and trading lenses, but never used in anger.
     
    • Like Like x 1
  10. WT21

    WT21 Mu-43 Hall of Famer

    Feb 19, 2010
    Boston
    The XTi was a great camera, and I used one for about two years. Coupled with the 35/2.0, it was an inexpensive powerhouse for good quality shots. It's failings (besides using CF cards) was a low-res LCD, no spot metering (though a chkdsk hack came along that enabled that) and I think a weakness in auto-ISO (maybe it didn't have auto ISO???)
     
    • Like Like x 1
  11. Hyubie

    Hyubie Unique like everyone else

    Oct 15, 2010
    Massachusetts
    Herbert
    Thanks everyone! As always, this forum is quick to the rescue. :smile: I just really wanted to make sure I wasn't missing much for the type of pictures I take.

    One question though - can someone kindly illustrate how a Canon raw "stands up better to processing" compared with a :43: raw?
     
  12. WT21

    WT21 Mu-43 Hall of Famer

    Feb 19, 2010
    Boston
    Hi Hyubie,

    It would be hard for me to demonstrate, as it would take a series of pictures, and I've not done a true side-by-side comparison (same shot, trying to work in post). My perceptions (and that's all they are at this point) come from working with a couple thousand photos from each system. But maybe this explanation would help???

    If I adjust colors or highlights severely in a Canon RAW, I can do it to about 2-3 stops. If I adjust similar in Oly RAW, very quickly noise shows up in the shadows or in color (I can get maybe 1 to 1 1/2 stops). For instance, if doing a B&W treatment, and adjusting blue (for instance, to darken it for the B&W treatment), then the Oly RAW shows noise very quickly, where the canon holds together a bit better.

    Also, when adjusting for black levels or contrast, I find the the Canon RAW maintains smooth gradients between colors and from light to dark. The Oly edges quickly define, and instead of getting smooth transitions, I get steep drop-offs. Almost like posterization.

    This matters less when I don't mind a shot with some noise or when I want stark contrasts (like a grainy B&W treatment) or if I'm purposely shooting high key, because noise isn't going to show up anyway. I think the biggest issue with Oly is that their blacks/shadows aren't really black enough. I could deal with the limited highlight range, if the blacks were a bit more pure. My solution so far has been to clip the black levels to get rid of the noise, but that impacts the overall image sometimes in ways that I don't want.

    In general, this means that if I need to get my exposure much more precisely correct at capture with the Oly than with the Canon. However, with the Oly narrowed dynamic range, it also makes working with tricky lighting conditions harder. The Oly can produce beautiful shots, but in a more narrow range of lighting conditions, and is lest forgiving in post if you make a mistake when you took the shot.

    The APS-C crop has less of an advantage over m43 than FF. After working with NEX and Canon crop DSLRs, I decided that the m43 kit size and the lenses made m43 worthwhile, and that really the only next step that makes sense is FF CSC, and that would of course limit a reasonable kit to a range of wide angle to short portrait (maybe 85). After that, the kit would cease to be compact. So, even in a world with FF CSC, I think there'll still be a place for m43, and especially if they can improve the sensors and get the shadows a little more pure.

    This is all based on working with Oly output (and their older sensors). I've not tried the Panny cameras yet. I downloaded a couple of G3 RAWs from the internet, and I think they did give more working room, so I am thinking of trying a G3.

    Last caveat -- this is mostly screen viewing. Printing is less of an issue I find, as the limitations of printing even out the differences. For instance, I have a shot of my daughter in Rocky Mts National Forest at the 150mm end of my 14-150. The lens doesn't have a ton of contrast at that end, and the blacks of the Oly are not as black as I'd like, and that lens is not the sharpest at 150mm, but it looks fine in print, but not as much on screen.

    Would love to hear from others...
     
    • Like Like x 1
  13. meyerweb

    meyerweb Mu-43 Hall of Famer

    Sep 5, 2011
    I think most of the differences in the images has to do with the processing settings (or factory defaults) and the amount of sharpening applied. Apply a bit more sharpening to the Canon image and it would probably look as sharp as the Pen. Similarly, you could up the saturation or shift the color balance a bit and get the Canon image looking more vibrant.

    Without having access to the actual tree, I obviously don't know which is more accurate, but the purple ornament looks unnatural and oversaturated, to me, in the Oly image.

    Really, almost any modern m4/3 or DSLR is capable of taking excellent photos. Even with less than perfect lenses. And to get the best images out of any of them, you need to understand how in-camera settings and post-processing affects the image.
     
    • Like Like x 1