MC Grain wrote:

Why don't you just cut to the chase an make the point you want to make. It sounds like you're just spewing the same old cliche thatit's the photographer not the camera that makes the photo - with the subtext being you're one of the photographers who can make better photos than everyone else with any gear. While I haven't looked at your port and therefore have no opinion on whether that's true, congratulations on getting to the point where you believe it is.


As far as what you've actually written, what you've demonstrated is a complete lack of common sense and critical thinking. You're making intentionally antagonistic smart ass comments about people's lack of high ISO photos in their ports. The thing is, everyone is excited about having a tool that now allows them to shoot that way. It's absurd and ignorant of the realities of time to think that people should already have a port full of photos with ISO images prior to have the tools the need to make them. Plus, some of the people here are looming at images prior to buying a camera that will allow them to shoot things they haven't been able to shoot.

One thing pretty much no one talks about is balancing flash and ambient when the flash is far too powerful. We have HSS for when the sun is to powerful, but there's no inverse for handheld when the flash is too powerful.

Flashes are designed for use in the 100-400 range and 1600 for bouncing. It's not easy to use a flash above that unless you stop down so much that you're better off at a low ISO.

Images that work at 1600 with a flash, need more than one stop of ISO to be shot without a flash. Rally it's more like 3 stops, so now that 12,800 is starting to be viable, you've got people talking about the possibilities that it opens for the future, and that is why they don't have ports filled with photos at high ISOs already.

I don't know whether this is all over your head or your grasp it all and are choosing to be insulting, but it seems that you don't have the self-awareness to understand how you're coming across. If you do, then you're trolling. Either way, we'd all be better off if you'd make your point once and be done.

Maybe start a thread called "why do you care about high ISO when you don't use it" and then people who want to debate a moot point can do so on topic.

LMAO what??

iseethelightman wrote:

Hmm dunno about all that. As for 2005 weddings, more than 20 photographers in my Rolodex were shooting 5D classics at 1600 and pushing 1 stop. They understood that pixel peeping isn't where the money is at. Capturing a great moment and printing it large and selling it to a client, or even better..their own portfolios to be shown. BTW noise hardly registers in a proper print all the way up to 3200 depending on size.

I have prints done that were shot at 12,800... Noise shows up much more on a backlight monitor than it does in print.

ACPhotography wrote:

I have prints done that were shot at 12,800... Noise shows up much more on a backlight monitor than it does in print.

Nice. What size?

ChanStudio wrote:
From the images and test that I have seen, I am pretty happy with the D800's DR and its pretty good ISO performance.  Personally, I wouldn't go over 3200 but if I do, I would make that into black and white image.

  The higher ISO you go, the lesser DR you getting and I like sensor with higher DR.

I am still waiting for D800 to arrive. How much longer do I have to wait?

Funny how we now talk about ISO instead of DR.  Is it because the 5DIII has nothing else to compare besides ISO?  joking.

Anyway, if you need higher ISO performance, go with Nikon D3s/D4 or Canon's 1Dx.  Why bother with the D800 or the 5DIII as both of these cameras aren't designed for high ISO performance.

I'm glad that I have high ISO if I need it.

Jerry Nemeth wrote:

I'm glad that I have high ISO if I need it.

I am sure but others might need better DR and more MP.  Majority of people would prefer better DR and more MP over high ISO (anything about 3200).  Few years back, ISO 3200 was considered as high ISO King and people would jump up and down with it.  Now we are talking about ridiculous ISO but losing out on DR.

  Ideally, if we have a camera that can produce constant DR regardless of ISO being selected, that would be great.

Jerry Nemeth wrote:
I'm glad that I have high ISO if I need it.

ChanStudio - OtherSide wrote:
I am sure but others might need better DR and more MP.  Majority of people would prefer better DR and more MP over high ISO (anything about 3200).  Few years back, ISO 3200 was considered as high ISO King and people would jump up and down with it.  Now we are talking about ridiculous ISO but losing out on DR.

  Ideally, if we have a camera that can produce constant DR regardless of ISO being selected, that would be great.

That's apparently not possible.

In any case, great high ISO capability doesn't reduce low ISO DR.

I just don't understand all these people complaining about cameras that can do things that previous cameras couldn't do, as if that's a bad thing. Autofocus? For amateurs! Video? Get a camcorder! Digital? If Ansel could do it with film, then so can you!

In the meantime, I'll look for photos where I previously couldn't find them.

Fred Greissing wrote:
Looks like the D800 is better than the 5D Mark III at ISO 12,800 when comparing scaled RAW files.



The inset is the d800 sample scaled to match the 5D sample.

Oh, sure, ha, ha.

Let's see the links to the original data and pics.  It has to be links to a real testing lab, and not some Nikon  blogger, because some people will believe anything.

Ruben Sanchez wrote:

Oh, sure, ha, ha.

Let's see the links to the original data and pics.  It has to be links to a real testing lab, and not some Nikon  blogger, because some people will believe anything.

Good Lord.. calm down... This is not intended to be scientific or anything... it's just resizing a screen capture from dpreview.

The good news is that both these cameras do a great job

iseethelightman wrote:

Nice. What size?

Mainly 8x10, I'm pretty sure larger would have looked as good since viewing distance is further away.

Fred Greissing wrote:
In camera jpeg better with the Canon 5DIII

http://www.cameralabs.com/reviews/Canon … oise.shtml

The impression I am getting is that with JPG's, the 5D III is about a stop better (although the noise tends to equalize when the D800 image is scaled down to the resolution of the 5D III).   The D800 RAW's seem to be just about equal within it's operating range.

That result doesn't surprise me.  Canon, using default settings, does a very good job with JPG's at noise reduction.  Indeed, the Canon engineers have already said, with the 1D X, to get the full benefit of high ISO, you will have to shoot JPG rather than RAW.

Again, when considering this, these tests were done using the defualt camera settings.  It isn't clear that they used the similar, aggressive noise reduciton features of the Nikon.

All that having been said though, I would be surprised if the 5D III didn't outperform the D800 in high ISO.  The cameras weren't built for the same purpose.  That is why you need to pick the right camera for the right job.

I am just surprised that they are within a stop of each other.  If the tests are accurate, I think Nikon did a darned good job of noise control with such high pixel density.  On the other hand, Canon deserves some kudos for a job well done also.

Leggy Mountbatten wrote:
http://www.youtube.com/watch?v=AfA2mTMt0u8

That settles it.

Leggy Mountbatten wrote:
http://www.youtube.com/watch?v=AfA2mTMt0u8

ei Total Productions wrote:
That settles it.

Yep. I wonder how many views it will take to pay for that camera.

ei Total Productions wrote:

The impression I am getting is that with JPG's, the 5D III is about a stop better (although the noise tends to equalize when the D800 image is scaled down to the resolution of the 5D III).   The D800 RAW's seem to be just about equal within it's operating range.

That result doesn't surprise me.  Canon, using default settings, does a very good job with JPG's at noise reduction.  Indeed, the Canon engineers have already said, with the 1D X, to get the full benefit of high ISO, you will have to shoot JPG rather than RAW.

Again, when considering this, these tests were done using the defualt camera settings.  It isn't clear that they used the similar, aggressive noise reduciton features of the Nikon.

All that having been said though, I would be surprised if the 5D III didn't outperform the D800 in high ISO.  The cameras weren't built for the same purpose.  That is why you need to pick the right camera for the right job.

I am just surprised that they are within a stop of each other.  If the tests are accurate, I think Nikon did a darned good job of noise control with such high pixel density.  ON th eother hand, Canon deserves some kudos for a job well done also.

It's nice that the D800 performs reasonably well at high ISO's and it's a nice feature to have. For me, I bought mine to shoot at 100 ISO, maybe 800 max if I need more dof/shutter speed. But mine will be shot either with strobes, good available light or off a tripod...

I did take it out last night to do some wide field astrophotography, wanna see how the stacking program handles 36MP...

ACPhotography wrote:
It's nice that the D800 performs reasonably well at high ISO's and it's a nice feature to have. For me, I bought mine to shoot at 100 ISO, maybe 800 max if I need more dof/shutter speed. But mine will be shot either with strobes, good available light or off a tripod...

I did take it out last night to do some wide field astrophotography, wanna see how the stacking program handles 36MP...

An astrophotographer on DP Review tested it and found it was doing signal processing at higher ISO's that makes it unsuitable for astro photos, at least above ISO 800. I think that was the cutoff, anyway. He was hoping for more faithful high ISO performance, so as to offer more options for his astrophotography.

ei Total Productions wrote:
The impression I am getting is that with JPG's, the 5D III is about a stop better (although the noise tends to equalize when the D800 image is scaled down to the resolution of the 5D III).   The D800 RAW's seem to be just about equal within it's operating range.

That result doesn't surprise me.  Canon, using default settings, does a very good job with JPG's at noise reduction.  Indeed, the Canon engineers have already said, with the 1D X, to get the full benefit of high ISO, you will have to shoot JPG rather than RAW.

Again, when considering this, these tests were done using the defualt camera settings.  It isn't clear that they used the similar, aggressive noise reduciton features of the Nikon.

All that having been said though, I would be surprised if the 5D III didn't outperform the D800 in high ISO.  The cameras weren't built for the same purpose.  That is why you need to pick the right camera for the right job.

I am just surprised that they are within a stop of each other.  If the tests are accurate, I think Nikon did a darned good job of noise control with such high pixel density.  On the other hand, Canon deserves some kudos for a job well done also.

I'm honestly not sure why people thought there would be such a difference. The d700 and 5dii were very close at high ISO (regardless of the fanboi comments) and even the d3s is only 3/4 stop better (according to everyones favourite site). The difference in MP was about the same as 5diii -> d800.

From what I can see, MP doesn't affect ISO performance all that much.

It's all about reducing the size of the image.

Leggy Mountbatten wrote:

An astrophotographer on DP Review tested it and found it was doing signal processing at higher ISO's that makes it unsuitable for astro photos, at least above ISO 800. I think that was the cutoff, anyway. He was hoping for more faithful high ISO performance, so as to offer more options for his astrophotography.

I can't tell you I'd do it a lot, but I wanted to see how it would work... I stayed at 800 ISO.

ChanStudio - OtherSide wrote:
Canon's current sensor technology has stalled.   Canon needs to put more R&D on sensors.  It can't make better ISO performance in RAW so it does it in JPGs.  Oh wait, what about DR?  Hmm..  I am hoping the 5DIII's DR would be at least 12.5.

As someone who owns both the MK II and the MK III, I can tell you definitively that high ISO performance in RAW is significantly better in the Mk III. I rate a camera's ISO performance by its "speed limit." In other words, what's the maximum ISO that would be useful to me?

The Mk II has a hard limit of 6400, because pattern noise becomes dominant above that ISO. It's the same limit as the 7D; above that, and the images turn to complete garbage.

For me, the Mk III's entire ISO range is now useful, depending on circumstances. That's a four stop difference. That's huge.

Before someone jumps in and claims I'm saying that ISO 102k on the Mk III is the same as 6400 on the Mk II: that's not what I'm saying. Just that the pattern noise is pretty much gone at the standard ISO's, and something I can deal with in Lightroom at ISO 51k and 102K. ISO 25K isn't even remotely usable in the Mk II; in the Mk III, it's quite respectable. I'd have no issues selling images shot at any ISO in the full range of the camera.

Leggy Mountbatten wrote:

As someone who owns both the MK II and the MK III, I can tell you definitively that high ISO performance in RAW is significantly better in the Mk III. I rate a camera's ISO performance by its "speed limit." In other words, what's the maximum ISO that would be useful to me?

The Mk II has a hard limit of 6400, because pattern noise becomes dominant above that ISO. It's the same limit as the 7D; above that, and the images turn to complete garbage.

For me, the Mk III's entire ISO range is now useful, depending on circumstances. That's a four stop difference. That's huge.

Before someone jumps in and claims I'm saying that ISO 102k on the Mk III is the same as 6400 on the Mk II: that's not what I'm saying. Just that the pattern noise is pretty much gone at the standard ISO's, and something I can deal with in Lightroom at ISO 51k and 102K. ISO 25K isn't even remotely usable in the Mk II; in the Mk III, it's quite respectable. I'd have no issues selling images shot at any ISO in the full range of the camera.

Be interesting to see some samples sometime if you get chance

Phil Drinkwater wrote:
Be interesting to see some samples sometime if you get chance

I think you've seen the only ones I currently have online. I have some shoots coming up, though. I plan to give it a workout!

Leggy Mountbatten wrote:

I think you've seen the only ones I currently have online. I have some shoots coming up, though. I plan to give it a workout!

Ahh I meant a comparison only if you have some free time.

I added a truck load of nr to the dpreview 5d3 ISO 12800 scene and was incredibly surprised to see a high quality result - easily good enough for a wedding whereas I don't like 6400 on the 5d2, so I do kind of agree with you.

Leggy Mountbatten wrote:
http://www.youtube.com/watch?v=AfA2mTMt0u8

All this tells me is that I should get a dedicated video camera.

c_h_r_i_s wrote:
What amuses me about all these geek experts doing these camera test is that they take a full frame pic then do a zooom in - blowup approx 1/50 of the image.

Wow look at the nose or pixel or whatever... WTF were you expecting to see !

My advice to them get the f**k in closer to the subject instead of having to blow up the image.
I'm not going to take a full length pic of a model just to do the final crop of her face.

well of course - but both images would look great sized down to 500px on the web. but if you're making big prints, then you want to know what the fine details look like.

Leggy Mountbatten wrote:
As someone who owns both the MK II and the MK III, I can tell you definitively that high ISO performance in RAW is significantly better in the Mk III. I rate a camera's ISO performance by its "speed limit." In other words, what's the maximum ISO that would be useful to me?

The Mk II has a hard limit of 6400, because pattern noise becomes dominant above that ISO. It's the same limit as the 7D; above that, and the images turn to complete garbage.

For me, the Mk III's entire ISO range is now useful, depending on circumstances. That's a four stop difference. That's huge.

Before someone jumps in and claims I'm saying that ISO 102k on the Mk III is the same as 6400 on the Mk II: that's not what I'm saying. Just that the pattern noise is pretty much gone at the standard ISO's, and something I can deal with in Lightroom at ISO 51k and 102K. ISO 25K isn't even remotely usable in the Mk II; in the Mk III, it's quite respectable. I'd have no issues selling images shot at any ISO in the full range of the camera.

Cool!

Leggy Mountbatten wrote:

As someone who owns both the MK II and the MK III, I can tell you definitively that high ISO performance in RAW is significantly better in the Mk III. I rate a camera's ISO performance by its "speed limit." In other words, what's the maximum ISO that would be useful to me?

The Mk II has a hard limit of 6400, because pattern noise becomes dominant above that ISO. It's the same limit as the 7D; above that, and the images turn to complete garbage.

For me, the Mk III's entire ISO range is now useful, depending on circumstances. That's a four stop difference. That's huge.

Before someone jumps in and claims I'm saying that ISO 102k on the Mk III is the same as 6400 on the Mk II: that's not what I'm saying. Just that the pattern noise is pretty much gone at the standard ISO's, and something I can deal with in Lightroom at ISO 51k and 102K. ISO 25K isn't even remotely usable in the Mk II; in the Mk III, it's quite respectable. I'd have no issues selling images shot at any ISO in the full range of the camera.

Found the same over the last day of using both..but I'll depart on the ISO 6400 thing. I had no trouble shooting my old 7D and 5D2 at 6400 and did so frequently. LR3 onwards makes light work of noise. Personally I live at a "hard" limit as you say of 3200.

On the 5D2 I preferred to shoot at ISO 1600, underexpose by 1 stop and bring it up in post. On the 5D3 I can just go ahead and shoot 3200 with the same result and no extra effort except to exposure.

For weddings and events 6400 to 25K is completely usable..then again I didn't have a problem doing that on the 5D2.

AF and ergonomics are improved. That alone is worth the buy.

Phil Drinkwater wrote:
Be interesting to see some samples sometime if you get chance

http://galleries.stevemelvin.com/5d_test/

Not an ideal test, but there are some things that can be gleaned from it.

Compared all three generations of 5D cameras, with links to images all resized to Mk I native size (4368 pixels wide).

ISO's from 200 to 102000 from all three cameras. Higher ISO's created by underexposing and pushing in Lightroom for the Mk I and Mk II.

Also did an image at ISO 100 underexposed by three stops and pushed in Lightroom to try to make it look like it was correctly exposed. It appears that the Mk II handles the push worse than the Mk I or Mk III, though none of them really passes this test. This is really the one thing I would like Canon to fix.

The Mk III seems to recover just a little more highlight detail than the others. On all cameras, the highlight recovery at ISO 100 and 125 was identical, and 160 was able to recover less detail. But 160 has more shadow detail. This lends support to the theory that 160 is a pull from 200, and 125 is a push from 100.

All shot with Auto WB, to see if they behaved differently. The Mk I and Mk II behaved identically, giving a color temp of 3550 and tint of +7. The Mk III gave a color temperature of 3250 and tint of +12. Perhaps the new meter is playing a role in this.

I used identical shutter speed and apertures for all three cameras; exposures appear to be identical over the standard ISO range.

The Mk III is clearly giving a lot more shadow detail at higher ISO's, not to mention much more pleasing noise. Note the cinder blocks to the left of the chair at ISO 100k.

All cameras are showing "amp glow" in the corners at extreme ISO's. Fortunately, the corners are usually relatively easy to deal with.

I tested, but didn't post, shots at 100, 125 and 160. I used a Fill Light of 80 to look at the shadow noise. On the Mk I and Mk II, 125 has the worst-looking shadow noise. 160 tends to look best, but not a whole lot better than 100. On the Mk III, the contrast changes, with 160 having the lowest contrast and most shadow detail. Again, 125 looks the worst.

The Mk I gives less-than-expected brightness at ISO 3200. I adjusted the ISO 6400 (equivalent) exposure to match the brightness of the ISO 1600 exposure.

Leggy Mountbatten wrote:
The Mk III seems to recover just a little more highlight detail than the others. On all cameras, the highlight recovery at ISO 100 and 125 was identical, and 160 was able to recover less detail. But 160 has more shadow detail. This lends support to the theory that 160 is a pull from 200, and 125 is a push from 100.

I've been shooting a little more to the right with the 5DIII. I need to remember to  shoot at the pulls. I have no problem giving up some highlight recovery for cleaner shadows.

Leggy Mountbatten wrote:
All shot with Auto WB, to see if they behaved differently. The Mk I and Mk II behaved identically, giving a color temp of 3550 and tint of +7. The Mk III gave a color temperature of 3250 and tint of +12. Perhaps the new meter is playing a role in this.

Did they all look the same?

Were you using Lightroom?

One thing to try is setting the WB on all three to the same Kelvin setting. When I did that with a 1D4, 5D2 and 7D, the numbers in LR were all different from what I'd set the camera to. I don't know if that is still happening, but you may be running into a symptom of whatever that bug is.

Leggy Mountbatten wrote:
The Mk III seems to recover just a little more highlight detail than the others. On all cameras, the highlight recovery at ISO 100 and 125 was identical, and 160 was able to recover less detail. But 160 has more shadow detail. This lends support to the theory that 160 is a pull from 200, and 125 is a push from 100.

MC Grain wrote:
I've been shooting a little more to the right with the 5DIII. I need to remember to  shoot at the pulls. I have no problem giving up some highlight recovery for cleaner shadows.



Did they all look the same?

Were you using Lightroom?

One thing to try is setting the WB on all three to the same Kelvin setting. When I did that with a 1D4, 5D2 and 7D, the numbers in LR were all different from what I'd set the camera to. I don't know if that is still happening, but you may be running into a symptom of whatever that bug is.

No, the Mk III was definitely cooler. You can see for yourself at the link.

I intentionally shot at Auto WB to see what the cameras delivered. Interesting to hear about this bug; that's the first I've heard of it. I'll have to take a closer look at the files.

Yes, I used Lightroom 3, processing the Mk III's files through the DNG converter, first. I haven't upgrade to LR 4, yet; Adobe rushed it out the door to force buyers of the new cameras to upgrade, and it's a bit too buggy to use in a production environment.