Cameras in the next 3 years

[Antithesis]

Yeah, and the D300 will be available for under $1k, possibly even new like they did with the d200, sooner than you'd think. It is an awesome performing camera that gave people what they asked for: Decent High-ISO noise in a borderline professional grade body at a decent price point.

The good thing about the fast pace of digital is that the price drops like crazy on bodies. If you wait a couple years, you will be able to get that D700 for under a grand, or close to it. But the latest and greatest will so far surpass it, that you might second guess the purchase. I see things getting a little absurd when we can hand-hold a camera in relative darkness, and take pictures of stuff we can barely see with little noise.

I'm more curious to see what next race will be. ISO is already getting exceedingly good, so maybe dynamic range?

 

[KmH]

Figure the D300 to drop to $1000 new, about a year or so after the intro of the D400. That's about when Nikon will be wanting to shed the remaing stock of the D300.

 

[DScience]

Figure the D300 to drop to $1000 new, about a year or so after the intro of the D400. That's about when Nikon will be wanting to shed the remaing stock of the D300.

That will be nice. My D90 can be the back up and i'll pick up a D300.

 

[itznfb]

The professional Nikkor lenses you're unaware of will be in the lineup for a good number of years,because they are world-class lenses. 200-400 VR, 200 f/2 VR, 300,400,500,600 f/2.8 VR models,all superlative.

So do these 500mm f/2.8 and 600mm f/2.8 lenses really exist? This is like the 10th post I've seen in the last year mention such lenses. I know I've never seen any information on them.

you guys do realize that higher megapixels = less noise at higher ISOs, right? not that the total amount of noise is less, but when you have more pixels, the noise becomes so small that you can't see it.

You're backwards. Lower megapixels at high ISO = less noise. One of the reasons the 1D Mark III is 10mp. Also the reason why D3, D700, D300s and D90 all remained 12.1 and 12.3 mp.

 

[Derrel]

Sorry, that was a typo on my part: the 300 and 400 are the f/2.8 VR models, the 500 and 600mm models are the f/4 VR models; the reason you have probably seen 500 and 600 lenses referenced as f/2.8 is that MOST people group the super-telephotos as 300/400/500/600 and in 10 years you've probably seen that many people make typos...and you're just the kind of guy who probably keeps track of such mistakes.

Nikon has updated its super-telephotos as a "set" for many decades now; like the Ai series, then the Ai-S series, then the AF versions, then the AF-i models with the first in-lens AF motors, then the AF-S models, and now, finally, the super-telephoto lineup is is ALL unified with AF-S G VR mounts;200-400 f/4 zoom, 200 f/2, 300/f2.8, 400/f2.8, 500 f/4, 600/f4, and all designated as AF-S G, with Vibration Reduction across all six lenses. NO f/stop ring on any of them.

As for your comment, "You're backwards. Lower megapixels at high ISO = less noise. One of the reasons the 1D Mark III is 10mp. Also the reason why D3, D700, D300s and D90 all remained 12.1 and 12.3 mp."

Well, that's a common theory that doesn't take into account advances in microlens design and image processing hardware and software; the original Nikon D1 and D1h models both had a 2.7 MP sensor, but it later became possible to get to 12 MP with LESS noise and higher resolving power. Also, noise tends to be sublimated when printing, and the higher-MP cameras also produce VASTLY larger images; today's upcoming 18 MP Canon 7D has a BIG image that measures 5,184-pixels on the long axis,while a 2.7 MP D1h produced a 2,008 pixel image on the long axis; if one takes the image from the new, ultra-high MP 7D and down-sizes it to match the size of the image from the 2.7 MP D1h image, or a 1D 4.2 MP image, or an EOS 30D 8.2 MP image, the actual, visible noise in a PRINT from the 18MP camera's file will be visibly better,and more pleasing, than the noise level from a camera that has a lower MP count but much older technology.

If one pixel-peeps and looks at huge,high MP images on-screen, yes there is noise, but there is also more real,actual resolved detail,and if one actually makes prints, or down-samples the images, the noise with high-MP sensors is pretty bearable.

Improvements in the image processing can make a high-MP sensor look good,or great. The sensel (sic) shared by the Sony a900 and the Nikon D3x is the same sensel, but Nikon's additions to the sensel and their proprietary image processing hardware and software yields images with lower noise and better image quality than the Sony camera yields. The 1D Mark III's 10.3 MP APS-H sensor is a very good sensor, and it yields a nice file,perfectly adequate when the camera can manage to focus properly. But the reason it's 10.3 MP is that newpaper and magazine screen printing makes higher MP counts prety much useless--the halftone screens used in today's magazines cut in-camera resolution so much that I'd challenge anybody to go through back issues of Sports Illustated and look at the double truck that were shot with the 4.2 MP 1D or the 8.2 MP 1D Mark II or the 1D Mark II, or the increasingly numerous Nikon D3 spreads--newspaper and magazine halftone reproduction makes it virtually impossible to tell if the capture file was 4,6,8,10,or 12 megapixels. Even on a two-page spread in Spots Illustrated.

As a former newspaper shooter, I'm familiar with prepping and transmitting images for halftone reproduction; the 2.7 MP of the D1h was quite adequate,and halftoned, one can barely tell the difference between 2.7 and 12.2 MP captures. A 10 MP file transmits via FTP a LOT faster than a 24.6 MP file,and the halftone issue and the transmission via e-mail or FTP is the reason PJ/sports cameras range from 10 to 12.2 MP. Many sports guys shoot 700 to 1,000 frames at a football or baseball game, and a smaller file size helps with card swaps, editing,archiving, and transmitting and until Nikon invented 4-channel readout, smaller MP count sensors helped with image buffer AND card write speeds, leading to higher Frames Per Second rates on sports cameras. Getting 10 FPS at 10.2 MP was about all Canon could muster in terms of throughput off the card with adequate buffer for long sequences when they engineered the 1D Mark III to run at 10 fps.

The 1D Mark III also has a very good per-pixel level image quality. Great microlens array ,perfect AA-filter, big pixels wells for high sensitivity and resistance to overexposure,and a file that writes and flushes and transmits perfectly well for newspaper, or PJ uses. 10.3 MP is a perfect compromise.

If anybody needs a high-resolution camera, Canon is over 21 MP, Nikon and Sony at 24.6 MP. The 5D Mark II at 21.7 MP shoots a better file at 3200 than the original 5D at 12.8 MP did.

 

[Antithesis]

Whew, novel and a half. Your high MP argument is pretty damn sound. Though, I hate large files.

 

[KmH]

Another issue related to noise performance is the quality of the S/N ratio of the amplifiers between the image sensor pixels and the A/D converter.

After all the heart of a digital camera, the image sensor, is an analog device.

 

[Garbz]

If you want to get into the nitty gritty than everything up to the digital conversion affects noise performance.

The parallel read out of data, the layout of amplifying transistors under the sensor pixel, the grounding within the sensor, the quality of the grounding in the electronics, etc.

 

[robertwsimpson]

The professional Nikkor lenses you're unaware of will be in the lineup for a good number of years,because they are world-class lenses. 200-400 VR, 200 f/2 VR, 300,400,500,600 f/2.8 VR models,all superlative.

So do these 500mm f/2.8 and 600mm f/2.8 lenses really exist? This is like the 10th post I've seen in the last year mention such lenses. I know I've never seen any information on them.

you guys do realize that higher megapixels = less noise at higher ISOs, right? not that the total amount of noise is less, but when you have more pixels, the noise becomes so small that you can't see it.

You're backwards. Lower megapixels at high ISO = less noise. One of the reasons the 1D Mark III is 10mp. Also the reason why D3, D700, D300s and D90 all remained 12.1 and 12.3 mp.

ok let me qualify.

higher megapixels means less noise when you keep your picture at a reasonable resolution. if you zoom to 100% (which there is never a reason to do) yes, you will see more noise. if you frame a picture exactly the same on a 1dmkIII 10.1 mp and a 1dmkIII 21.1 mp, and look at the whole picture, guess which will have less noise... hmmm... let me think...

 

[itznfb]

The professional Nikkor lenses you're unaware of will be in the lineup for a good number of years,because they are world-class lenses. 200-400 VR, 200 f/2 VR, 300,400,500,600 f/2.8 VR models,all superlative.

So do these 500mm f/2.8 and 600mm f/2.8 lenses really exist? This is like the 10th post I've seen in the last year mention such lenses. I know I've never seen any information on them.

you guys do realize that higher megapixels = less noise at higher ISOs, right? not that the total amount of noise is less, but when you have more pixels, the noise becomes so small that you can't see it.

You're backwards. Lower megapixels at high ISO = less noise. One of the reasons the 1D Mark III is 10mp. Also the reason why D3, D700, D300s and D90 all remained 12.1 and 12.3 mp.

ok let me qualify.

higher megapixels means less noise when you keep your picture at a reasonable resolution. if you zoom to 100% (which there is never a reason to do) yes, you will see more noise. if you frame a picture exactly the same on a 1dmkIII 10.1 mp and a 1dmkIII 21.1 mp, and look at the whole picture, guess which will have less noise... hmmm... let me think...

That's apples to oranges. I have to assume you meant 1ds since the 1d doesn't come in a 21mp variant. You're comparing Full Frame to APS-H. Apples to oranges.

 

[robertwsimpson]

regardless, more pixels = more area to distribute noise over, which = less noise if you don't zoom to 100%

 

[Garbz]

The only problem is that a lot of software doesn't work like that, and noise is random.
The former comment is for example most image viewers will revert to nearest neighbour interpolation or something equally basic when displaying a zoomed out version. The simple reason is when me (the user) is flicking through 200 photos, I don't have the patience to wait for the image to resample. That was the biggest complain with Windows Picture and Fax Viewer. When scrolling backwards it's even worse. The Windows Picture Viewer in Windows 7 is faster but still slow compared to ACDSee even with colour management enabled in ACDSee. So you still end up with noisy pixels.

The noise is random issue is that while if you take the average of two adjacent pixels the noise is halved, but those two adjacent pixels will rarely follow this perfect ideal world. They may both be hot and thus no noise saving is gained from reducing the picture size by zooming out.

So yes it works, but it's not as good as having a much larger area to capture photons (such as with the D3). It's a question of signal conditioning to reduce noise while reducing definition, vs just simply increasing signal to noise at the source.

Oh just thought of something else too. There's a gap between pixels. It's small, but on a high pixel density sensor this further handicaps it in noise comparisons.

 

[schumionbike]

regardless, more pixels = more area to distribute noise over, which = less noise if you don't zoom to 100%

That's true but something like a Nikon D3 have less noise to distribute to begin with even though it does have a smaller area so there's two sides to it. You won't know which camera give better result at hight ISO until you see the prints from them. But you're right, it's not fair to compare the noise level of a 25 megapixel camera to that of a 12 megapixel camera if you're going to zoom in to 100% on both. That say, I think the Nikon D3 performed better than the D3X at high ISO level at most print sizes.