Can someone post 800, 1600 and 3200 ISO photos from 7D.......

Started Nov 25, 2009 | Discussions
OP Eugene Powers Forum Pro • Posts: 10,108
Re: Can someone post 800, 1600 and 3200 ISO photos from 7D.......

Yes, I was able to download them but now I am completely confused.

First of all thank you and download of all files took no more than 15 seconds all at once. This must be some kind of the record.

Now about my confusion.

All your pictures are extremely underexposed (at least 1.25 stops) yet there no more noise in them than in properly exposed pictures I have seen elsewhere. HOW??????

Do you have the latest firmware no one knows about it? Hell, these pictures have no more noise than my 5D2. I am extremely impressed.

-- hide signature --

Eugene

The only time a smaller sensor with the same pixel count is superior to a larger sensor (aka higher pixel density) is when you are focal-length limited.

Lee Jay

John Sheehy Forum Pro • Posts: 20,682
Re: Can someone post 800, 1600 and 3200 ISO photos from 7D.......

Eugene Powers wrote:

All your pictures are extremely underexposed (at least 1.25 stops) yet there no more noise in them than in properly exposed pictures I have seen elsewhere. HOW??????

HTP?

-- hide signature --

John

OP Eugene Powers Forum Pro • Posts: 10,108
Re: Can someone post 800, 1600 and 3200 ISO photos from 7D.......

Only if Reinhard gives me permission.
--
Eugene

The only time a smaller sensor with the same pixel count is superior to a larger sensor (aka higher pixel density) is when you are focal-length limited.

Lee Jay

John Sheehy Forum Pro • Posts: 20,682
Re: Can someone post 800, 1600 and 3200 ISO photos from 7D.......

Eugene Powers wrote:

Only if Reinhard gives me permission.

Huh? He already used, or didn't use, "HTP".

-- hide signature --

John

OP Eugene Powers Forum Pro • Posts: 10,108
Re: Can someone post 800, 1600 and 3200 ISO photos from 7D.......

He used some kind of the server. But these are his pictures and he sent them to me in the private email. If he reads this and gives permission I will post them on Megaupload.
--
Eugene

The only time a smaller sensor with the same pixel count is superior to a larger sensor (aka higher pixel density) is when you are focal-length limited.

Lee Jay

fanboy Regular Member • Posts: 118
Define "regular lights"

Eugene Powers wrote:

in RAW of the average room with regular lights, no flash.

What are "regular lights" ?

fanboy Regular Member • Posts: 118
Re: Can someone post 800, 1600 and 3200 ISO photos from 7D.......

Eugene Powers wrote:

Thanks. This is exactly what I am trying to do. There is a thread on 5D forum where most people hate 7D

That's not surprising. If I had a 5D, I'd hate the 7D too...considering how awesome the 7D is in comparison.

Phil Hill Senior Member • Posts: 2,757
Re: Can someone post 800, 1600 and 3200 ISO photos from 7D.......

Eugene Powers wrote:

no more noise in them than in properly exposed pictures I have seen elsewhere. HOW??????

That's not surprising.

Noise is very well controlled in the 7D. In fact, it seems to be only slightly higher than 5D2 samples I've seen, except you may be more likely to get banding from the 5D2. Some posted 7D samples were converted by a beta version of ACR that didn't deliver great results.

I can find a bit of noise at ISO 1600 in my 7D shots by looking for it. Nothing obviously noticeable below 1600, and it's only slightly worse at 3200. Just guessing, but I'd expect the 5D2 to show a bigger advantage at ISO 6400, and probably also at ISO 100 - 400.

The 7D seems to be optimized for 800 - 3200, although that might simply be the raw converter I've been using recently (Bibble 5 Pro beta). Or perhaps my threshold is higher, coming from the film world.

Phil Hill Senior Member • Posts: 2,757
HTP

By HTP, John probably meant Highlight Tone Priority.

OP Eugene Powers Forum Pro • Posts: 10,108
Re: Define "regular lights"

Whatever is in your room. Nothing staged.
--
Eugene

The only time a smaller sensor with the same pixel count is superior to a larger sensor (aka higher pixel density) is when you are focal-length limited.

Lee Jay

OP Eugene Powers Forum Pro • Posts: 10,108
Re: Can someone post 800, 1600 and 3200 ISO photos from 7D.......

Do you really believe this yourself? So why so many people returned 7D even those who tried them both at the same time. I think the answer is somewhere else.

I think they did not use right converter or something. So far I see nothing wrong with 7D samples I got and no one posted any proof or side by side comparison with definitive proof that 5D2 is better in IQ than 7D.
--
Eugene

The only time a smaller sensor with the same pixel count is superior to a larger sensor (aka higher pixel density) is when you are focal-length limited.

Lee Jay

Phil Hill Senior Member • Posts: 2,757
Re: Can someone post 800, 1600 and 3200 ISO photos from 7D.......

Eugene Powers wrote:

So why so many people returned 7D even those who tried them both at the same time. I think the answer is somewhere else.

The 7D comes close to the 5D2 in terms of noise, at least up to ISO 1600 or 3200, and it has even tamed the banding monster at those speeds. But there is more to IQ than just noise. I believe that some really were expecting a miracle from Canon, but physics dictates that a FF sensor is going to be better for some types of photography than APS-C.

Putting those issues aside, the 7D is better than the 5D2 in just about every way that counts. I'm a satisfied customer.

OP Eugene Powers Forum Pro • Posts: 10,108
Re: Can someone post 800, 1600 and 3200 ISO photos from 7D.......

Did you do a comparison between 7D and 5D2 yourself in term of IQ or you just read this somewhere?
--
Eugene

The only time a smaller sensor with the same pixel count is superior to a larger sensor (aka higher pixel density) is when you are focal-length limited.

Lee Jay

Phil Hill Senior Member • Posts: 2,757
Re: Can someone post 800, 1600 and 3200 ISO photos from 7D.......

Eugene Powers wrote:

Did you do a comparison between 7D and 5D2 yourself in term of IQ or you just read this somewhere?
--

The 7D reduces lens resolution by the 1.6x factor right from the start, so you can never get the full resolution of any lens when printing the full frame. Now those that crop in place of using a longer lens (birders, etc.) will benefit, but everybody else gets reduced resolution from the lens. Specially designed APS-only lenses can help to overcome this 1.6x disadvantage, but many are still shooting FF lenses on the 7D.

On top of that, the 5D2 has more than double the surface area to collect light. Again, physics come into play, giving an advantage to the full-frame sensor in terms of DR and reduced noise. More light = less noise, better color and cleaner images, all other things being equal.

Now remember, this is coming from a guy that really likes his 7D. I wouldn’t trade it for brand-new 5D2 even if it was offered. This is because the advantages of the 5D2 IQ while potentially real, are not meaningful for my purposes. Others might (and some do) make a different choice.

Based on what you’ve said you do with your cameras, though, the 5D2 might not be a better choice for you either.

But you can’t say that the 5D2 doesn’t have higher IQ for certain types of photography, at least not with credibility.

To specifically address your question, no I’ve not personally used a 5D2 because I prefer APS-C. But I’ve seen enough images from highly-qualified photographers and reputable testers to know that for certain things, the 5D2’s full frame sensor is the best tool for the job, just as medium format film was better for some things than 35mm. Conversely, 35mm was superior in other areas.

I do know that noise is not at all an issue for the 7D, despite some reports to the contrary. The lack of high-ISO banding is especially nice. And I’ve also found (from personal experience) is that to come really close to full-frame IQ you need to use lenses that are so good that they can afford to give up the 1.6x resolution factor. I’ve already replaced several lenses that weren’t delivering everything my cameras (50D and 7D) are capable of, and seen the improvement.

It’s interesting to note that the 1.6x APS-C factor is almost exactly the same as the magnification factor between 2 1/4 MF film and 35mm. I don’t remember anybody looking at a new 35mm camera and film combination, and speculating that it could reach the same IQ level as MF.

OP Eugene Powers Forum Pro • Posts: 10,108
Re: Can someone post 800, 1600 and 3200 ISO photos from 7D.......

Phil Hill wrote:

Eugene Powers wrote:

Did you do a comparison between 7D and 5D2 yourself in term of IQ or you just read this somewhere?
--

The 7D reduces lens resolution by the 1.6x factor right from the start, so you can never get the full resolution of any lens when printing the full frame. Now those that crop in place of using a longer lens (birders, etc.) will benefit, but everybody else gets reduced resolution from the lens.

Sorry, but this a total nonsense. 7D does not reduce lens resolution. In fact 7D uses the center part of the lens so you don't see everything from the edge where the resolution is lower. Resolution of the lens is measured by lines per mm
so no matter where you are n the lens surface it is still measured per mm.

Think of it as the rain measurement. No matter how big of the area volume per inch stays the same.

And so far 7D or 5D2 do not outresolve good lenses.

Specially designed APS-only lenses can help to overcome this 1.6x disadvantage, but many are still shooting FF lenses on the 7D.

Nope. APS-C lenses use less glass to save money (no one actually doing very good job on saving us money) but that is debatable. I think they just want to make more money by selling more lenses. No one proved that APS-C lenses are better on crop factor cameras than FF lenses. And a lot of people shooting FF lenses on crop factor cameras for this exact reason. I am doing the same thing myself. I had 40D and 50D which I recently sold.

On top of that, the 5D2 has more than double the surface area to collect light. Again, physics come into play, giving an advantage to the full-frame sensor in terms of DR and reduced noise. More light = less noise, better color and cleaner images, all other things being equal.

Really, so by your statement 1Ds2 (FF 16mp) should be much better than 5D2 (FF 21mp) or 1DS (FF 11mp) should be much better than 1Ds2?
That is why I sold 1Ds and 1Ds2 because they both were better than 5D2.
Oh yeh I guess upcoming 1D MK4 will be much worst than 5D2 according to you.
Well, it is all in the technology.

And the reason I sold 1Ds2 and got 5D2 is not the amount of pixels, though 21mp is nice to have, but the reason is because 5D2 has better noise control.

800 ISO on 1Ds2 was pushing it but on on 5D2 I am not worried to shoot at 1600. AS far as IQ both cameras are about the same.

Now remember, this is coming from a guy that really likes his 7D. I wouldn’t trade it for brand-new 5D2 even if it was offered. This is because the advantages of the 5D2 IQ while potentially real, are not meaningful for my purposes. Others might (and some do) make a different choice.

I am glad you like your 7D but you did not do full IQ comparison to 5D2. And those who are saying they like 5D2 IQ more than 7D besides their words do not give reasonable explanation WHY. That is the reason I started this thread in the first place.

Based on what you’ve said you do with your cameras, though, the 5D2 might not be a better choice for you either.

I am not saying that 7D (what I have seen so far) is better than 5D2 but 5D2 is definitely a very good choice for me and from what I have seen so far 7D might be perfect complimentary secondary body for me too and I don't even shoot weddings.

But you can’t say that the 5D2 doesn’t have higher IQ for certain types of photography, at least not with credibility.

Might be. 7D has pixel density of 47MP camera so with a good lens it might be better for a distant subjects unless noise destroys the picture.

To specifically address your question, no I’ve not personally used a 5D2 because I prefer APS-C. But I’ve seen enough images from highly-qualified photographers and reputable testers to know that for certain things, the 5D2’s full frame sensor is the best tool for the job, just as medium format film was better for some things than 35mm. Conversely, 35mm was superior in other areas.

Don't believe everything you see. All testers use specific setups to make tests.

That is why I asked to take pictures in the everyday situation without specifically setting anything.

I do know that noise is not at all an issue for the 7D, despite some reports to the contrary. The lack of high-ISO banding is especially nice. And I’ve also found (from personal experience) is that to come really close to full-frame IQ you need to use lenses that are so good that they can afford to give up the 1.6x resolution factor. I’ve already replaced several lenses that weren’t delivering everything my cameras (50D and 7D) are capable of, and seen the improvement.

First and again lets make this perfectly clear. There is no such thing as FF IQ.

It is all in technology. 40D for example had better IQ than 1Ds. They just trying to use better technology for more expensive FF cameras. Another example would be 5D2 and 1Ds3. 5D2 IQ is not better than 1DS3 but at higher ISO it has better noise reduction.

And your replacement of lenses has nothing to do which camera you are going to put them on. You just got better lenses, that is it. It might be other factors too.

For example I have 70-200 USM IS which was not very good on 1Ds2 yet it is much better on 5D2. Explain that.

It’s interesting to note that the 1.6x APS-C factor is almost exactly the same as the magnification factor between 2 1/4 MF film and 35mm. I don’t remember anybody looking at a new 35mm camera and film combination, and speculating that it could reach the same IQ level as MF.

Actually a lot of photogs here who are shooting with MF cameras are saying that we already reached low end MF but since I don't shoot MF I can't say that.

-- hide signature --

Eugene

The only time a smaller sensor with the same pixel count is superior to a larger sensor (aka higher pixel density) is when you are focal-length limited.

Lee Jay

Phil Hill Senior Member • Posts: 2,757
Where to begin?

Eugene Powers wrote:

Sorry, but this a total nonsense. 7D does not reduce lens resolution. In fact 7D uses the center part of the lens so you don't see everything from the edge where the resolution is lower. Resolution of the lens is measured by lines per mm
so no matter where you are n the lens surface it is still measured per mm.

Think of it as the rain measurement. No matter how big of the area volume per inch stays the same.

You’re mistaken about this. Yes, at 100% the resolution is exactly the same, and that's what you're probably thinking of. But all APS-C images require more magnification for a given print size, due to the smaller sensor. Just like film. The ratio of print size to sensor size determines the magnification, not the number of pixels. Remember, an optical image is projected on the sensor, and APS images require greater magnification for a given print size because the image projected onto the sensor is smaller. It would be the same if we were using APS film versus 35mm.

And yes, a growing number of APS lenses do have higher resolution, just as the lenses on some PS cameras are higher still. The 17 – 55 EF-S is an example that performs better on APS cameras than similar L lenses.

And then there are the lenses designed for Olympus 4/3 cameras that are known for higher resolution than traditional FF 35mm lenses. The greater magnification of the smaller 4/3 sensor demands it.

Also, the new APS-only Tamron 60mm outperforms any FF lens of similar focal length at f:2.0 that I've owned, including some of the best.

On top of that, the 5D2 has more than double the surface area to collect light. Again, physics come into play, giving an advantage to the full-frame sensor in terms of DR and reduced noise. More light = less noise, better color and cleaner images, all other things being equal.

Really, so by your statement 1Ds2 (FF 16mp) should be much better than 5D2 (FF 21mp) or 1DS (FF 11mp) should be much better than 1Ds2?

I'm at a loss here. Nowhere did I imply that. In the first place, all of those sensors have the same surface area. Second, I said all other things being equal, meaning similar technology.

Given similar technology, a larger sensor will definitely outperform a smaller one. Naturally, if you compare sensors from different generations (as in your examples), this might not always hold. But we were only discussing current models. And all of those sensors are the same size anyway.

Just because the benefits of full-frame sensors aren’t often realized in your photography (or mine), that doesn’t mean there are no benefits. In the film days (going back to the sixties) I had MF (2 1/4) gear, but I rarely used it, even though images were technically of higher quality. The reason was that for most of what I was shooting, the differences were not significant. But I also knew when a job did call for the big guns.

To see an example of where FF tends to beat APS-C, just find a local landscape photographer and ask him to show you some 20 x 24 (or larger) prints. It's easy to miss the subtleties when looking at 8-bit compressed JPEG images on the web, but you'll tend to notice it in the prints. This is why working landscape photographers typically prefer full-frame DSLRs.

reinhard Contributing Member • Posts: 514
Re: HTP

Hi,

Sorry for answering late. Got to bed early, and the timezone difference causes delays as well

You may use those pictures to your liking. No worries. There is no art or anything, just test shots.
Test setup was as simple as you could imagine:

  • Camera put on a table to avoid shake

  • Shutter release through Zoom Browser from Laptop

  • Lens EF-S 17-55 2.8, as this is the sharepest one I have, apart from my 70-200 f4L IS. But the room wasn't big enough to use the 70-200.

  • between the shots only ISO and aperture was changed. This of course also changed the shutter speed as the light (from a regular light bulb) did not change

-

  • Firmware is 1.1.0, to my knowledge the latest one.

  • HTP off

  • High ISO Noisereduction: Standard

The pictures are not processed, straight out of the camera. EXIF data should be with every picture.

If anybody else wants to get them: I'd need your email, or -if someone knows a platform where I can upload them unmodified to make them available for pulling, let me know.

-- hide signature --

Cheers,
Reinhard

Canon 50mm 1.8, Sigma 105mm Macro, Canon 70-200 4.0 L IS, Canon EF-S 2.8 18-55mm IS,Panasonic DMC-LX3)
http://www.pbase.com/reinhard
http://www.fotki.com/reinhard

Phil Hill Senior Member • Posts: 2,757
More on lens resolution, APS versus FF

When I say that lens resolution is reduced on APS-C, I’m talking about the resolution on the print, not the sensor. Of course the resolution at the sensor is the same, but it gets reduced by the extra magnification required to make a print. People often miss this on the screen, because it doesn’t exist at 100%, and if an image is resampled downward to fit the screen, so much data is thrown away that it doesn’t make a difference.

So nobody is going to be able to prove this to you with screen images, but you’ll see it if you look at some large FF prints, especially landscapes containing lots of foliage detail.

ChristineI Regular Member • Posts: 245
Re: HTP

reinhard
This is a free large upload site, its the one I have been using.
http://www.mediafire.com/about.php
--
Christine
http://ciphotography.freehostia.com/index.php

 ChristineI's gear list:ChristineI's gear list
Panasonic Lumix DMC-ZS40 (TZ60)
reinhard Contributing Member • Posts: 514
Re: HTP

Thanks, Christine.

Eugene will make them available through MegaUpload. But I'll check out Mediafire as well.

-- hide signature --

Cheers,
Reinhard

Canon 50mm 1.8, Sigma 105mm Macro, Canon 70-200 4.0 L IS, Canon EF-S 2.8 18-55mm IS,Panasonic DMC-LX3)
http://www.pbase.com/reinhard
http://www.fotki.com/reinhard

Keyboard shortcuts:
FForum MMy threads