Ad
Ad
Ad
Pages: « 1 [2] 3 »   Bottom of Page
Print
Author Topic: Canon 40D And Nikon D300 Cameras  (Read 23600 times)
JohnKoerner
Guest
« Reply #20 on: January 28, 2008, 10:26:08 PM »
ReplyReply

Thanks for your comment.

How would you compare the autofocus of the 40D and the D300 in the field?

I have held each in my hands at the camera shop, and fiddled with them, but have not had the opportunity to test either one yet.

Jack

PS: If you don't mind my asking, what prompted you to sell the 40D?
Logged
cecelia
Newbie
*
Offline Offline

Posts: 33


« Reply #21 on: January 28, 2008, 11:59:29 PM »
ReplyReply

"How would you compare the autofocus of the 40D and the D300 in the field?"

Both are very, very good.  You can even go back and forth without too much confusion.  The D300 AF and the Auto ISO features are really outstanding--significantly better than the 40D.  The D300 AF is uncanny, and seeing the points it used to set the focus makes using AF much more intuitive - at a minimum, you immediatly know when to use manual focus.  I also think the AWB is better on the D300.  The D300 is 12 MPIX vs. 10 MPIX, and that is a plus.  However, the 40D is $600 cheaper...  

"PS: If you don't mind my asking, what prompted you to sell the 40D?"

Good question...I wanted to experience a full-frame camera.  I also began to realize that Canon and Nikon have very different strengths and weaknesses in their lens lineups.  I decided that a D3 or a 1DsMKIII would be too heavy, and also very expensive.  So my near-term compromise was 5D + D300.  If a 5DII comes out, I will either sell the 5D or convert it to IR.  A full-frame Nikon in the D300 form factor would also be very appealing.  I've acquired the amazing 14-24 lens, which I hope to use on both the D300 and the 5D with Mark's Nikon to Canon adapter.

I sold the 40D because two DSLR cameras is already one too many.  The 40D is a very nice system, and I can't really convince myself the 5D is better, except for 1 extra stop on the ISO dial.  Originally I though I'd rent new cameras to try them out, but it turns out that the losses incurred with a quick buy/sell are cheaper than renting a camera for a week.  

Hope this helps, and the good news is that you can't really go wrong with either of these outstanding machines.

-Cecelia
Logged
JohnKoerner
Guest
« Reply #22 on: January 29, 2008, 12:27:44 AM »
ReplyReply

Interesting. And thanks again.

I think I am going to learn to live with the slightly less-effective autofocus in the 40D (that is still darned good), and be take heart in the savings I will enjoy going with the Canon.

Still, the Nikon D300 sounds like a wonderful camera, and I am sorely tempted to go that way instead ... but everytime I go to B&H to buy one, when I put together the starter system I want (the D300 body + the Nikkor 28-70mm f/2.8D ED-IF lens and the Micro-Nikkor 200mm f/4D IF-ED lens) ... I keep getting stuck on the fact those 3 pieces alone total $1,200 more than the Canon with a similar lens configuration.

I keep rubbing my chin and thinking, if I went with the Canon I could add a 4th piece (namely an EF 70-200mm f/2.8L zoom) to my similar Canon starter kit, and have spent roughly the same amount of money as on the Nikon system, where I don't get an extra 70-200 lens. And this just stops me every time.

At the end of the day, I think I would derive more enjoyment out of that 3rd lens to add to my 40D, for the same monetary expenditure, than I would derive from a little better autofocus capability in the Nikon, but without an added 70-200 lens.

>Sigh< If Nikon were comparable in price, I probably would get the D300 in a heartbeat.

But I just can't get over the fact that I can get the Canon 40D (a wonderful camera in its own right) plus 3 L-series lenses for the same price I could get the D300 and only 2 nice lenses. I think that alone will be the kicker for me.

Thanks again

Jack
« Last Edit: January 29, 2008, 12:33:35 AM by JohnKoerner » Logged
DonWeston
Full Member
***
Offline Offline

Posts: 148


« Reply #23 on: January 30, 2008, 07:09:51 AM »
ReplyReply

Quote
Interesting. And thanks again.

I think I am going to learn to live with the slightly less-effective autofocus in the 40D (that is still darned good), and be take heart in the savings I will enjoy going with the Canon.

Still, the Nikon D300 sounds like a wonderful camera, and I am sorely tempted to go that way instead ... but everytime I go to B&H to buy one, when I put together the starter system I want (the D300 body + the Nikkor 28-70mm f/2.8D ED-IF lens and the Micro-Nikkor 200mm f/4D IF-ED lens) ... I keep getting stuck on the fact those 3 pieces alone total $1,200 more than the Canon with a similar lens configuration.

I keep rubbing my chin and thinking, if I went with the Canon I could add a 4th piece (namely an EF 70-200mm f/2.8L zoom) to my similar Canon starter kit, and have spent roughly the same amount of money as on the Nikon system, where I don't get an extra 70-200 lens. And this just stops me every time.

At the end of the day, I think I would derive more enjoyment out of that 3rd lens to add to my 40D, for the same monetary expenditure, than I would derive from a little better autofocus capability in the Nikon, but without an added 70-200 lens.

>Sigh< If Nikon were comparable in price, I probably would get the D300 in a heartbeat.

But I just can't get over the fact that I can get the Canon 40D (a wonderful camera in its own right) plus 3 L-series lenses for the same price I could get the D300 and only 2 nice lenses. I think that alone will be the kicker for me.


Thanks again

Jack
[a href=\"index.php?act=findpost&pid=170494\"][{POST_SNAPBACK}][/a]
Sorry for the late update, but I decided to shoot a couple of more [ high school basketball] with the D300 with lower NR settings then I had used before. I had wrongly assumed that coming from the older Nikon bodies, that the highest setting would yield  the best images. I had found when I did this, that detail severely suffered. Resolution in those cases ended up worse then the 40D with it NR on. With the d300 now set to low NR, the D300 makes images more on a par with the 40D at high ISO, but there is also a return of significant grain like noise. I guess you choose your poison....I will also add that this is when looking at images on a monitor at ridiculous sizes, like 66-200%. Also, when looking at prints, yes actual prints of all the images up to 20x40 or 24x32 inches, that at a respectable distance of 18-24 inches or beyond in viewing distance, that all prints looked great compared to images I made last year with a 5D. I think this goes to show that pixel count is NOT so important, processing firmware and software in camera has continued to move forward and these gains not pixel count make results much better then one woul expect from pixel numbers alone. I would suggest NOT looking at anything over what ever pixel size correlates with the largest print you are going to need or want to make, and sizes like 100%  or larger are really silly. Basics like having good light amount and quality will surpass any differences than pure numbers in all cases. A higher MP count alone, will not necessarily help. If you really want an eye opener as far as image quality alone, download as I have done and made sectional crop prints from what would be the latest uncomparable cameras, the D3 and XSi. These should not be compared for anything but image quality obviously, feature set, cost make it a silly comparison. I made prints from what would have been 30x45 in images. POINT - very little if any difference  in overall quality was seen when looking at these at even this size image. Yes this  is not a controlled or fair experiment, just a comparison of two random images supplied by the manufacturers of either camera, but I urge you to do it yourself, you do not need anything larger then a desktop printer and do sectional prints. My last comment, is buy the gear that feels comfortable and suits your shooting needs, style, concentrate on good technique and art if appropriate and let that be your guide.
Logged
BradSmith
Sr. Member
****
Offline Offline

Posts: 265


« Reply #24 on: January 30, 2008, 01:10:22 PM »
ReplyReply

Quote
My last comment, is buy the gear that feels comfortable and suits your shooting needs, style, concentrate on good technique and art if appropriate and let that be your guide.
[a href=\"index.php?act=findpost&pid=170943\"][{POST_SNAPBACK}][/a]

Don,
I absolutely agree with this comment.  I have always advised friends, when they ask the inevitable question of "what should I buy?", to decide on the general type of camera that you want, determine which are in that field, and then..... Hold them.  Play with the controls.  How do they feel?  Does the controls layout seem to work for you, etc.  Then buy the one that FEELS best to you.  

I'm following my own advice.  I'm just jumping into a DSLR system after years of 4x5 B&W.  I'm in the market for a pro-sumer class DSLR and lenses.  The major manufacturers all have the lenses that will meet my needs.  The 40D feels best to me, so that's what I'm getting.
Brad
Logged
DonWeston
Full Member
***
Offline Offline

Posts: 148


« Reply #25 on: January 30, 2008, 03:21:28 PM »
ReplyReply

Quote
Don,
I absolutely agree with this comment.  I have always advised friends, when they ask the inevitable question of "what should I buy?", to decide on the general type of camera that you want, determine which are in that field, and then..... Hold them.  Play with the controls.  How do they feel?  Does the controls layout seem to work for you, etc.  Then buy the one that FEELS best to you. 

I'm following my own advice.  I'm just jumping into a DSLR system after years of 4x5 B&W.  I'm in the market for a pro-sumer class DSLR and lenses.  The major manufacturers all have the lenses that will meet my needs.  The 40D feels best to me, so that's what I'm getting.
Brad
[a href=\"index.php?act=findpost&pid=171032\"][{POST_SNAPBACK}][/a]
Brad - one of the most sensable posts here, too much is gear driven by pure numbers. I have many large prints in my office, the two most complimented on by patients or family were shot with the cheaper end of the spectrum of equipment, one shot from 20 yrs ago in Jerusalem with a Nikon FM and tokina 70-200 zoom and a Canon Pro one P&S. Two of the most cheapest cameras I have owned in decades, both images far from perfect, although decently done, but win on content. Happy shooting with your 40d when it arrives...Don
Logged
JohnKoerner
Guest
« Reply #26 on: February 01, 2008, 08:42:14 PM »
ReplyReply

Quote
Sorry for the late update, but I decided to shoot a couple of more [ high school basketball] with the D300 with lower NR settings then I had used before. I had wrongly assumed that coming from the older Nikon bodies, that the highest setting would yield the best images. I had found when I did this, that detail severely suffered. Resolution in those cases ended up worse then the 40D with it NR on. With the d300 now set to low NR, the D300 makes images more on a par with the 40D at high ISO, but there is also a return of significant grain like noise. I guess you choose your poison....I will also add that this is when looking at images on a monitor at ridiculous sizes, like 66-200%. Also, when looking at prints, yes actual prints of all the images up to 20x40 or 24x32 inches, that at a respectable distance of 18-24 inches or beyond in viewing distance, that all prints looked great compared to images I made last year with a 5D. I think this goes to show that pixel count is NOT so important, processing firmware and software in camera has continued to move forward and these gains not pixel count make results much better then one woul expect from pixel numbers alone. I would suggest NOT looking at anything over what ever pixel size correlates with the largest print you are going to need or want to make, and sizes like 100% or larger are really silly. Basics like having good light amount and quality will surpass any differences than pure numbers in all cases. A higher MP count alone, will not necessarily help. If you really want an eye opener as far as image quality alone, download as I have done and made sectional crop prints from what would be the latest uncomparable cameras, the D3 and XSi. These should not be compared for anything but image quality obviously, feature set, cost make it a silly comparison. I made prints from what would have been 30x45 in images. POINT - very little if any difference in overall quality was seen when looking at these at even this size image. Yes this is not a controlled or fair experiment, just a comparison of two random images supplied by the manufacturers of either camera, but I urge you to do it yourself, you do not need anything larger then a desktop printer and do sectional prints. My last comment, is buy the gear that feels comfortable and suits your shooting needs, style, concentrate on good technique and art if appropriate and let that be your guide.



Thanks Don.

It seems that you and Cecelia have direct firsthand experience using these two cameras side-by-side, with Cecelia favoring the D300 slightly and you favoring the 40D slightly.

How do you find the autofocus on the 40D? Cecelia found the D300 to be far superior. If the 40D doesn't quite match the D300's autofocus, is the 40D's autofocus still acceptable? Also, how about in low light?

I intend to do some night-time macro work (moths and other nocturnal insects/spiders), and do you think the 40D would be up to that task? I intend to use it with a Canon EF 180mm f/3.5L Macro USM lens, and possibly a Macro Lite Adapter Flash Ring as well.

Thanks again.
« Last Edit: February 01, 2008, 08:44:25 PM by JohnKoerner » Logged
DarkPenguin
Guest
« Reply #27 on: February 01, 2008, 08:50:15 PM »
ReplyReply

http://www.popphoto.com/cameras/5062/nikon...on-workout.html
Logged
JohnKoerner
Guest
« Reply #28 on: February 01, 2008, 10:34:08 PM »
ReplyReply

Quote


Thank you for that.

That article pretty much said everything that has been the general consensus, but it illuminated my two specific questions exactly: (1) Autofocus and (2) Using in Low Light.

* Regarding autofocus, the article said, "During our tests tracking cyclists, horse riders and other fast-moving targets, we noted exceptionally swift (and roughly comparable) continuous AF speed with all sensors active."

* And regarding low light, the article said, "Surprisingly, the simpler Canon system was quicker than the D300's when acquiring initial focus in moderately low light (indoors or at dusk), an advantage augmented by Canon's super-swift USM (ultrasonic-motor) fast-aperture prime lenses."

And, finally, in RAW mode (which is all I will ever shoot), the images are virtually indistinguishable, and more importantly, when shooting continuously in 14-bit RAW mode the Canon kept its ability to do so up at a rapid-fire 6 fps, whereas the Nikon dropped significantly to a paltry 2.5 fps.

For my purposes, this is significant. And again, the price difference isn't just in the camera, it's in the lenses too. When I configured the systems I wanted (body + 2 lenses), the Nikon was $1,200 more in total, and really almost $1,500 more when you add to this the fact you also have to purchase additional software just to gain the ability to process the RAW images. (Canon supplies you with the necessary software.)

It seems like none of the D300's high-tech advantages "on paper" translate to any real advantes to the finished product, the photos, while actually some of its features actually offer disadvantages while working in the field. The live view that is so wonderfully-sharp in the D300, the article said, "Curiously, the Nikon (but not the Canon) flipped its mirror down every time we shot, disabling Live View until we reengaged it manually (by half-pressing the shutter release). That's the major reason we preferred the Canon for Live View; it also has a real-time histogram, exposure simulation, optional grid overlay and two quieter shooting modes."

So even though the Nikon has the brighter and prettier live view screen, at the end of the day it doesn't seem as functional.

Again, I appreciate the link.
Logged
John_Black
Sr. Member
****
Offline Offline

Posts: 263


WWW
« Reply #29 on: February 03, 2008, 02:10:40 AM »
ReplyReply

I recently sold my Canon 1Ds2 after several years and switched to medium format with a digital back.  The digital back files are great, but ergonomics & functionality compared to the 1Ds2 stink.  Became clear very quickly that I need a small dSLR to augment the medium format set-up, so I bought a Nikon D300 and a Canon 40D and putting them through their paces.  Here are some differences - in no specific order:

D300 AF system is very good, selecting an AF point is very simple and direct.  40D is nice, but the D300 is nicer and better.

D300's metering seems more willing to push the exposure to its edge; 40D seems to under expose around 1/3 of a stop.  On the Canon I tend to use EC adjustment more often.  D300 needs less EC.

D300 has a slightly larger viewfinder and 100% coverage, but no optional focus screens.  40D has optional focus screens.

D300's LCD rocks.  Canon needs to update their bodies asap.  

D300's direction pad works better than Canon's joystick controller.

D300's front dial is too far recessed in the grip, it should stick out a bit more.  Canon's placement of the front dial (just above the shutter button) makes more sense to me.

Switching metering modes, AF modes and drives modes on the 300D is a bit hard to do without actually looking at the camera.  With the 40D I can do this a bit easier because all the buttons are group above the upper LCD panel.

D300 has great build quality; 40D feels equally well bit - perhaps a bit more solid.

D300's on board flash seems better - exposures seem better.  Haven't used it much on either camera.

Had some odd WB from the D300; Canon seems slightly better (maybe), but tends to have slight magenta tendency.

40D does better on noise at lower ISO's.  On upper ISO it's a toss up and really depends on the light conditions - various conditions mask noise better than other conditions.

Canon's highlight tone priority works as advertised, but I'm not sure if I like it.  Nikon's D-Light is need (really just boosts the lower mid tones), but can save some post processing work.  Have both in conjunction would be really neat.

Canon files seem need to more post processing, but also take it better.  Nikon files need less work in the raw editor (using C1v4 for the Nikon, DPP for the Canon).  

The 40D seems to have a tiny bit more DR - may 1/3 stop.  It's a tiny, tiny difference.

The 40D shoots at a base ISO of 100; Nikon needs its extended ISO turned on.  40D does better at ISO 100 --- cleaner and more DR.

I'm going to stay away from sharpness because i don't have like quality lenses for both cameras.  For the Nikon I'm using the 18-200 VR whereas on the Canon I'm using Contax lenses via an adapter.  It's not even fair to compare.

I haven't totally figured out live view yet, but Canon's seems a bit better despite no AF.  I have no real meaningful opinion here.

Nikon has viewfinder has gridlines on demand - love that feature.

Nikon shoots 2.5 FPS @ 14 bits vs the 6.5 on the 40D.  That's stated in the specs, so Nikon isn't hiding anything, but people often talk about 6 FPS as it's uniformly available.

40D fits my hand better - it sit lower, more in the palm.  The D300 are a lip on the rear rubber where the upper edge of your thumb should sit.  The Nikon sits higher, slightly above the palm.  With the 40D I feel like I can really grip it and I'm not scared of dropping it.  D300 would probably feel more secure with the battery grip added.


I really like what Nikon has done with D300, but as said above - it's ~$1749 vs a 40D ~$1119 (I bought mine used for almost $200 less).  Bang for the buck goes to the 40D.  300D adds alot of features and with its AF engine, I think a sports shooter should spend some time with one.  For more general shooting I think the 40D works better.  It's a cleaner simpler design and doesn't get in the way of itself.  

Right now I can't choose between them, it's a very close race.  I really like the 18-200 VR for its range and stabilization.  If Canon had a 18-200 EF-S IS, I'd probably pick the Canon.  The money saved on the body could go towards a nice lens like a 35L (a nice fast prime for walk-around shooting).  Downside of the Canon is that to match the 18-200 VR, you need a 17-85 IS and 70-300 IS.  Canon has cheaper alternates such as the 18-55 EFS-IS and the new 55-250 EF-S IS, so dollar for dollar that matches the VR.  VR still wins in terms of having it all in one lens.

Eventually I'll probably buy a 1Ds3 or D3X and flip back to a high end dSLR.  Since the D3X doesn't exist yet, I don't which system I'll ultimately pick.  I'd prefer to pick the same brand today (ie the 40D or D300) today and purchases lenses which will work on the FF dSLR.  In the end there is no clear winner.  It's a tough choice.
Logged

JohnKoerner
Guest
« Reply #30 on: February 03, 2008, 08:12:36 AM »
ReplyReply

Interesting review John.

The one thing that puzzled me about your review was when you said, "I'm going to stay away from sharpness because i don't have like quality lenses for both cameras. For the Nikon I'm using the 18-200 VR whereas on the Canon I'm using Contax lenses via an adapter. It's not even fair to compare."

If this is true, and you used a sub-par lens and not a high-end Canon L-series USM, not only would it be unfair to compare sharpness, but wouldn't you say it would also be unfair to compare autofocus and ultimate end-product results as well? If you had a EF 70-200mm f/2.8L IS USM I think not only would this have made a difference in the sharpness, but also in the speed of AF as well as end-product quality on many levels. Wouldn't you agree?

That being said, it was still an interesting and helpful review on many levels. For instance, I too found that the build quality in both were excellent, but that the Canon just felt better in my hand. The Nikon has an annoying "lip" on it, at the ends of your fingers, that's true.

Thanks for sharing.
Logged
DarkPenguin
Guest
« Reply #31 on: February 03, 2008, 10:46:53 AM »
ReplyReply

Quote
Nikon shoots 2.5 FPS @ 14 bits vs the 6.5 on the 40D.  That's stated in the specs, so Nikon isn't hiding anything, but people often talk about 6 FPS as it's uniformly available.

The boys at Outback photo like the high speed for trying to do HDR shots.  Any significant difference between the 12 bit files and the 14 bit ones?
« Last Edit: February 03, 2008, 10:47:10 AM by DarkPenguin » Logged
John_Black
Sr. Member
****
Offline Offline

Posts: 263


WWW
« Reply #32 on: February 03, 2008, 01:24:21 PM »
ReplyReply

I've shot with Canon pro bodies for over 4 years (1Ds, !D2 and 1Ds2) and this is my second 40D too.  I've owned most of the L lenses at one point or another.

In regards to 12 v 14 bits; the upper highlights seem to have better gradients - smoother.  I'm not sure if the 2-bits make a significant visible difference.  There seems to be some added clarity, but saying that's because of the added bits is a subjective guess.  With the Phase back the extra bits pay off in post processing because those files can pushed, stretched, twisted, etc., without falling a part.  Part of medium format files' advantage is the added DR the lower tones and essentially no noise at the base ISO.  In a 1Ds2 v medium format bit comparison, I would say the difference is profound, whereas the 40D/D300 the value of the extra bits is harder to quantify.  

The P25 back has set a new standard (for me) for what a clean file should look like.  Whichever body comes closest to that mark, that's the one I'll keep.  The differences in auto focus, metering, buttons, LCDs, etc., are easy metrics for comparison purposes, but for my needs it really comes down to the ISO 100 image quality, the dynamic range and the color mapping.
« Last Edit: February 03, 2008, 01:46:31 PM by John_Black » Logged

DonWeston
Full Member
***
Offline Offline

Posts: 148


« Reply #33 on: February 05, 2008, 06:45:08 AM »
ReplyReply

Quote
Thanks Don.

It seems that you and Cecelia have direct firsthand experience using these two cameras side-by-side, with Cecelia favoring the D300 slightly and you favoring the 40D slightly.

How do you find the autofocus on the 40D? Cecelia found the D300 to be far superior. If the 40D doesn't quite match the D300's autofocus, is the 40D's autofocus still acceptable? Also, how about in low light?

I intend to do some night-time macro work (moths and other nocturnal insects/spiders), and do you think the 40D would be up to that task? I intend to use it with a Canon EF 180mm f/3.5L Macro USM lens, and possibly a Macro Lite Adapter Flash Ring as well.

Thanks again.
[a href=\"index.php?act=findpost&pid=171684\"][{POST_SNAPBACK}][/a]
John K. - glad to hear others feel similarly, anyway, shooting high school basketball in gyms is one of the hardest tasks you can ask of a camera, light amount and quality stinks to put it mildly. Both can vary all over the board, but without resulting to flash which is disturbing to the players, I find both bodies focus well, considering that expecting 100% focusing accuracy in these conditions is not realistic. At least not yet with todays tech. I have not used the newer bodies of either Canon or Nikon pro line, no 1d3 or D3. These kids are quick and shooting wide open with shallow DOF, even focus tracking has alot to keep up with. Am working at wide open apertures or nearly wide open and fast shutter speeds. If you are shooting macro at night, I assume with flash, light amount and quality won't be an issue and you will be shooting at lower ISOs and sync speeds and stopped down, no? This is more my macro experience anyway. Focusing should not be too big an issue? Live view?
John - I can not tell you for sure as I have not done this kind of work, I usually shoot my bug stuff during daytime, mostly with flash. For my sports work, I find the end results from both cameras very similar in the end. At present I am trying to decide which camera, system works better for me. Both cameras are excellent, and differences are subtle at most. Am trying hard NOT too get obsessed about these small differences in making my selection...not easy, some days I feel one way, the next the other...aaarrrggghhhh!!! LOL....
Logged
Tony Beach
Sr. Member
****
Offline Offline

Posts: 452


WWW
« Reply #34 on: February 09, 2008, 01:03:42 AM »
ReplyReply

Quote
However, now I'm not so sure. It seems that the lower noise of the D300 is at the expense of detail destroying noise reduction.
[{POST_SNAPBACK}][/a]


I found these test results interesting:

[a href=\"http://www.diwa-labs.com/photoalbum/view/?size=org&id=191656]http://www.diwa-labs.com/photoalbum/view/?size=org&id=191656[/url]

http://www.diwa-labs.com/photoalbum/view/?size=org&id=191599

http://www.diwa-labs.com/photoalbum/view/?size=org&id=191663

http://www.diwa-labs.com/photoalbum/view/?size=org&id=191588

http://www.diwa-labs.com/photoalbum/view/?size=org&id=191655

http://www.diwa-labs.com/photoalbum/view/?size=org&id=191601
Logged
John Sheehy
Sr. Member
****
Offline Offline

Posts: 838


« Reply #35 on: February 09, 2008, 12:29:02 PM »
ReplyReply

Quote
I found these test results interesting:
[a href=\"index.php?act=findpost&pid=173485\"][{POST_SNAPBACK}][/a]

They look like they are based upon different conversion styles.

These tests really should be done with literal RAW conversions, if its the camera one is interested in, and not the standard conversion recipes.

The bottom line is that Canon had lower noise in their cameras for so long that they did nothing about software noise reduction, and set a standard for third-party converters to leave the noise mostly untouched, while other companies like Nikon scrambled to hide their relatively noisier sensors and electronics in software noise reduction.  This carries on today, even though the noise characteristics of recent Nikons are much better, with the D3 being better than all but one Canon, and the D300 close to the 40D.
Logged
Tony Beach
Sr. Member
****
Offline Offline

Posts: 452


WWW
« Reply #36 on: February 09, 2008, 02:42:03 PM »
ReplyReply

Quote
They look like they are based upon different conversion styles.

These tests really should be done with literal RAW conversions, if its the camera one is interested in, and not the standard conversion recipes.

The bottom line is that Canon had lower noise in their cameras for so long that they did nothing about software noise reduction, and set a standard for third-party converters to leave the noise mostly untouched, while other companies like Nikon scrambled to hide their relatively noisier sensors and electronics in software noise reduction.  This carries on today, even though the noise characteristics of recent Nikons are much better, with the D3 being better than all but one Canon, and the D300 close to the 40D.
[{POST_SNAPBACK}][/a]

DIWA's testing indicates that the D300 noise performance is better than the 40D, and all the mumbo jumbo about different RAW converters and how they handle the files differently is unsupported speculation on your part.  My unsupported (but more believable to me) speculation is that third party converters profile files for DSLRs independently of the camera manufacturers, and Nikon has no more input than Canon does in this regard (nothing they include in the file to instruct their own RAW converter how to apply NR and sharpening is applied by third party converters).

Since the proof of the pudding is in the eating, and since I generally use Capture One for my own conversions, I downloaded some comparable files from the D300, 5D, and D3 from a DPR forum and converted them for myself.  I set the WB off of the wall and applied EC to match the EV as much as reasonably possible, then applied identical RAW converter NR settings to the D300 and 5D files, and identical sharpening settings to all three camera's files.  The D3 file required substantially less NR than the other files, the fact that there is a 2/3 of a stop difference in exposure might play a role in this; but regardless of that, the D3 was more like two stops better than the other cameras (subtract the extra 2/3 of a stop, and that adds up to just over a one stop improvement).

[a href=\"http://photos.imageevent.com/tonybeach/mypicturesfolder/sharing/ISO%201600_D300-5D-D3_Comparison.jpg]http://photos.imageevent.com/tonybeach/myp..._Comparison.jpg[/url]

What is most important in looking at these comparisons is that detail (and keep in mind that these are 100% crops) is not lost and very comparable looking at both the D300 and 5D.  If you shoot RAW then NR is optional, I have disabled it completely with every RAW converter I have used and generally apply it sparingly.  Reading forums on the internet I find that every camera eventually gets mishandled by photographers with questionable skills or agendas -- some cameras seem more prone to mishandling than others, and Nikon has apparently responded to that by following Canon's lead in applying a non-linear curve to the RAW data:  http://forums.dpreview.com/forums/read.asp...essage=26710014

Now in another round of conversions using ACR, the noise of the 5D file was less and the D300 file was noisier (again using the same NR and sharpening settings).  You can see crops from those results here:  http://photos.imageevent.com/tonybeach/myp...haring//ACR.jpg  What this indicates to me is that the argument that Nikon has some sort of inside track regarding NR being applied by third party converters is demonstrably false.

The shortcomings of ACR for my NEF files led me a long time ago to stop using it to convert my files.  You can see what ACR did to my D200 files compared to some other RAW converters here:  http://photos.imageevent.com/tonybeach/myp...comparisons.jpg  What's my point?  Well you buy a camera you're comfortable with and you learn how to use it, and if you shoot RAW to get the most out of that camera, you naturally use the best available RAW converter and set it up to optimize your results.  Unfortunately, I am still waiting for Raw Magick Lite to support the D300, but I have allowed my trial version of Lightroom to expire and have not updated CS2 in large part because I still consider the way ACR handles NEF files unsatisfactory (mostly it's a color issue for me, noise being much reduced with the D300 than what it was with the D200).
Logged
DarkPenguin
Guest
« Reply #37 on: February 09, 2008, 02:46:06 PM »
ReplyReply

Quote
My unsupported (but more believable to me) speculation

That's going on my white board right next to "Why can't we concentrate our resources across the board?""
Logged
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #38 on: February 09, 2008, 04:03:44 PM »
ReplyReply

Quote
My unsupported (but more believable to me) speculation

*Your* unsupported speculation is certainly no mumbo jumbo, it is worth of taking a look at.

Quote
nothing they include in the file to instruct their own RAW converter how to apply NR and sharpening is applied by third party converters

What makes you think that there is *anything* in those files regarding how to apply NR?

Quote
I set the WB off of the wall and applied EC to match the EV as much as reasonably possible, then applied identical RAW converter NR settings to the D300 and 5D files, and identical sharpening settings to all three camera's files

You find it all right to compare the noise in two images, where one of them has been created by *50% more light* than the other?

How was that mumbo jumbo?

Quote
the D3 was more like two stops better than the other cameras (subtract the extra 2/3 of a stop, and that adds up to just over a one stop improvement)

Care to explain, how you converted the difference in noise into EV?

Quote
If you shoot RAW then NR is optional, I have disabled it completely with every RAW converter I have used and generally apply it sparingly

You *can not* disable NR with ACR from version 4. A basic NR is included in the de-mosaicing algorythm, to the pain of quite a few phhotographers, who would rather see some noise but keep the details.

Pls note, that I don't have any reason to defend the 5D, a 2.5 years old camera against the most up to date cameras. My beef is only with the method of comparison.
Logged

Gabor
Tony Beach
Sr. Member
****
Offline Offline

Posts: 452


WWW
« Reply #39 on: February 09, 2008, 05:04:34 PM »
ReplyReply

Quote
What makes you think that there is *anything* in those files regarding how to apply NR?

I don't.

Quote
You find it all right to compare the noise in two images, where one of them has been created by *50% more light* than the other?

I worked with the files provided by someone else, my own approach would have been to use the same aperture, shutter speed, and ISO for all the files.  The person who created the files used the cameras' metering systems and let the chips fall where they may.

Quote
Care to explain, how you converted the difference in noise into EV?

Simple measurement of std. deviation in various color channels mostly taken from the wall by taking a specified number of pixels at a specific coordinate and matching it exactly to the other two camera's files, and by comparing files created from the same source at various ISOs.  The ISO 400 file from the D3 looked comparable to the ISO 1600 files from the 5D and D300 using Capture One (with all NR and sharpening disabled).  Since the D3 files were exposed 2/3 of stop to the right I subtracted that and came up with just over a one stop better performance if they had all been taken at the same shutter speeds, ISO. etc.

One issue that arises though is where in the scene can the images be said to be "equal"?  If I set the point in the shadows, midtones, or highlights I ended up with three different reference points with no way to exactly match all three camera's files -- so I chose a brighter section of the wall and used the D300 as the baseline so that none of the cameras would end up having negative EC applied.

Quote
You *can not* disable NR with ACR from version 4. A basic NR is included in the de-mosaicing algorythm, to the pain of quite a few photographers, who would rather see some noise but keep the details.

Perhaps so, but I would not presume that advantages any one camera over any other; since I personally don't use ACR and the initial comparison was with Capture One, it hardly seems relevant.  Looking at my converter comparison with the D200 used ACR 3.7 (which is still the version of ACR that converts D200 files, at least with the current version of Lightroom), I would have a hard time being convinced that the D200 enjoyed any NR whatsoever looking the file ACR generated.  One thing is certain to my eyes, there was abundant detail in all the ISO 1600 files from the D3, D300 and 5D and a big difference in both noise and detail as NR was raised and lowered using ACR as well as Capture One.
Logged
Pages: « 1 [2] 3 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad