EOS 7D Mark II Spec List Surfaces [CR1]

[Marsu42]

ff just to good

So you'd rather use a 5d1 instead of a 70d ?

 

[pj1974]

I remember someone saying change the card slot to SD....the worse idea I've head this thread, cf all the way I want my buffer clear asap....

also usb3 or better yet gig Ethernet jack to get my files off.

I can't see GigE connectivity on a camera.... consumers like wireless... the throughput of wireless is laughable compared to GigE, but wireless is more convenient... so the poorer solution wins... USB3 has to come soon, at some point people will stop making chipsets that only go up to USB2... USB will win out over Ethernet because all you have to do is plug it in.... no configuration required, and a lot of people use laptops and tablets with no wired Ethernet connection... USB is a more universal solution than wired Ethernet.

I can't speak from experience, as I haven't made the leap yet, but I understand 802.11ac is actually starting to approach the real-world throughput of gigE. While I doubt they'd put an ac antenna array in the 7d2 for space and power reasons, they COULD get the throughput if they wanted.

Also, given that Apple's target market for laptops correlates pretty strongly with pro and prosumer camera buyers, I think a thunderbolt / USB3 combo would be awesome- USB3 for PC users, TB for mac...

This is a bit of a simplistic explanation, but here goes....

There are two basic types of network traffic, on type is where you set up "a pipe" and the data automatically streams down the pipe from one device to the other, the other type of data flow is a send/acknowledge data flow.... something like "here's a bit of data", answered by "I got it.... send me another"... and so on.. Most network traffic tends to be send/acknowledge and it takes time for the requests and acknoledgements to fly back and forth so the flow of data is slow.

When you connect with wire, data can flow both ways at once and this greatly speeds up the send/acknowledge protocols. On a wireless link you can only go one direction at a time and it takes time to turn the link around... plus you can have interference on wireless which causes re-transmissions and further slows things down. Processing of the data is faster on wire than wireless so there is less delay there too.

All this adds up...

In the end, you find out that it takes almost the same amount of time on a wireless link to move 20 bytes of data as it takes to move 1400 bytes of data so if you want any decent kind of throughput you need to be moving huge blocks of data in one direction and small amounts the other way..

Marketing people have a different perspective.... they compare the most favourable conditions of a wireless link to the least favourable conditions on a wired link... and that's how they come up with claims to be "almost as fast"

As a photographer, I suggest you do your own test... transfer a bunch of photos from a laptop to a computer over a wireless link, and then repeat the process over a wire link... send some tiny Jpgs and then try some RAW files and see the difference it makes to you on your gear...

I’ve written it before, and I’m repeating it here again – I really appreciate many of your posts on Canon Rumours forum, Don.
Thanks to you and many others (of course Dr Neuro deserves thanks as well for his great technical insights)

Your post above is spot on: technically accurate and helpful. I find that I rarely use wireless – though it can be handy at times (eg quick connect of laptop for 2 minutes). However for serious internet usage, data transfers, I stick with wired connections. So much faster and predictable (and I find, also more secure – ie less possibility of the occasional wireless drop-outs).

I have undertaken ‘tests’ of file transfer speeds for many years: using tiny JPEG vs large files (eg RAWs) and even vs much larger video files. I have done this to test HDD to HDD transfers, USB (back in the days of USB1.1, USB2.0 and now USB3.0. USB3.0 definitely ‘rocks’! I remember when I thought USB2.0 was ‘blazingly fast’… lol.

I’m looking forward USB3.0 being the new normal standard for all devices and hopefully included on the 7DmkII.(that might be my next camera, but in the meantime – I’m very happy with my Canon 7D. The CR1 rumoured spec list for the 7DmkII looks good (maybe too good to be true) – but I trust Canon will put out a winner when it does!

Regards,

Paul

 

[WPJ]

Additionally, with WiFi it is a shared medium, meaning all devices on that channel whether on your SSID or not share that bandwidth, your neighbor if configured for the same channel will also cut your own bandwidth. Additionally you may have all N or newer devices but if you have one old g or even B device is old ds from your kids, your spectrum turn down to the lowest common denominator B. Ouch yes your connection may still say 300mbps but if there is a b radio in site not on your network your AP has to throttle down to b for that device, sorry that's the spec.

there is tricks if your APs will deploy them to over come these issues but not on consumer devices, so unless you are running a Juniper, Cisco (not a Cisco best buy special) it other enterprise controller your out of luck.

 

[Pi]

ff just to good

So you'd rather use a 5d1 instead of a 70d ?

I would. The only reason I did not buy one a few years ago was the lack of MA.

 

[Pi]

Also, why is it that pixel size does not matter? I believe a lower resolution sensor of the same physical size would have proportionally larger photosites, which would exhibit less noise by gathering larger amount of light.

True for each photosite but not true for the image as a whole. What matters is the total light collected on the sensor. Different pixel densities mean different ways to sample the projected image, and a lower sampling rate is never better. Of course, there are also all those technological challenges.

Think about this: a 40mp sensor contains all the information a 10mp sensor can collect because you can always bin in software (ignoring the tech challenges for a moment). But it contains more information.

 

[WPJ]

Also, why is it that pixel size does not matter? I believe a lower resolution sensor of the same physical size would have proportionally larger photosites, which would exhibit less noise by gathering larger amount of light.

True for each photosite but not true for the image as a whole. What matters is the total light collected on the sensor. Different pixel densities mean different ways to sample the projected image, and a lower sampling rate is never better. Of course, there are also all those technological challenges.

Think about this: a 40mp sensor contains all the information a 10mp sensor can collect because you can always bin in software (ignoring the tech challenges for a moment). But it contains more information.

I'm just guessing but a full fram 40mp full frame camera will probably have the same size pixels as a 24mp crop sensor.

so the camp of my full frame captures more light really only captures more megapizels

 

[Skirball]

Also, why is it that pixel size does not matter? I believe a lower resolution sensor of the same physical size would have proportionally larger photosites, which would exhibit less noise by gathering larger amount of light.

True for each photosite but not true for the image as a whole. What matters is the total light collected on the sensor. Different pixel densities mean different ways to sample the projected image, and a lower sampling rate is never better. Of course, there are also all those technological challenges.

Think about this: a 40mp sensor contains all the information a 10mp sensor can collect because you can always bin in software (ignoring the tech challenges for a moment). But it contains more information.

I see what you’re saying, but that assumes that you’re combining photon counts from adjacent pixels, rights? E.g., if you grouped every four pixels together and counted the totals as a single pixel it would be equivalent to the 10 mp sensor with the same theoretical SNR. In that case, sure, more data is always better. But if you’re not summing the pixels then although you’d have four times as many with the 40 mp, the full well capacity would have to 1/4 of the 10 mp. Maybe not an issue at low ISO, but it’s going to limit you as you push higher, no?

 

[Skirball]

Also, why is it that pixel size does not matter? I believe a lower resolution sensor of the same physical size would have proportionally larger photosites, which would exhibit less noise by gathering larger amount of light.

True for each photosite but not true for the image as a whole. What matters is the total light collected on the sensor. Different pixel densities mean different ways to sample the projected image, and a lower sampling rate is never better. Of course, there are also all those technological challenges.

Think about this: a 40mp sensor contains all the information a 10mp sensor can collect because you can always bin in software (ignoring the tech challenges for a moment). But it contains more information.

I'm just guessing but a full fram 40mp full frame camera will probably have the same size pixels as a 24mp crop sensor.

so the camp of my full frame captures more light really only captures more megapizels

If my maths are correct I think the crop would be more around 15 mp.

I don't understand your second comment. Are you implying that a full frame doesn't capture more light than a crop sensor?

 

[Pi]

I see what you’re saying, but that assumes that you’re combining photon counts from adjacent pixels, rights? E.g., if you grouped every four pixels together and counted the totals as a single pixel it would be equivalent to the 10 mp sensor with the same theoretical SNR. In that case, sure, more data is always better. But if you’re not summing the pixels then although you’d have four times as many with the 40 mp, the full well capacity would have to 1/4 of the 10 mp. Maybe not an issue at low ISO, but it’s going to limit you as you push higher, no?

The well capacity would be enough to be 1/4 but the light falling on each pixel is 1/4 as well, so there is no problem.

so the camp of my full frame captures more light really only captures more megapizels

More pixels with the same light per pixel = more total light.

Again, the mp number is irrelevant. The noise is part of the image itself -it has a discrete nature. Different pixel densities sample it in a different way but a lower pixel count is no better.

 

[Skirball]

I see what you’re saying, but that assumes that you’re combining photon counts from adjacent pixels, rights? E.g., if you grouped every four pixels together and counted the totals as a single pixel it would be equivalent to the 10 mp sensor with the same theoretical SNR. In that case, sure, more data is always better. But if you’re not summing the pixels then although you’d have four times as many with the 40 mp, the full well capacity would have to 1/4 of the 10 mp. Maybe not an issue at low ISO, but it’s going to limit you as you push higher, no?

The well capacity would be enough to be 1/4 but the light falling on each pixel is 1/4 as well, so there is no problem.

Well yes, that was my point. You would have to reduce the FWC to 1/4 to maintain exposure. So, you increase the ISO two stops. Which would increase the noise at each pixel.

However, the thing I admittedly never thought about until reading this thread: If you then downsample the image in post to the same size as the 10 mp, do the pixels average out to give the same general level of noise as the 10 mp sensor? The more I think about it, at low ISO the answer has to be yes. But, if you’re really pushing the high ISO are the results so far off that it will screw with the final average? And I guess I don’t fully understand how the averaging would work considering the Bayer layer – statistically wouldn’t more noise mean more issues with the green channel, so averaging out the noise isn’t going to be so simple.

Sorry to ramble on mindlessly, it just got me thinking. I can’t believe I never thought of it that way, I was always of the camp of ‘so long as you have enough resolution for your uses, the bigger the pixels the better’.

 

[WPJ]

I see what you’re saying, but that assumes that you’re combining photon counts from adjacent pixels, rights? E.g., if you grouped every four pixels together and counted the totals as a single pixel it would be equivalent to the 10 mp sensor with the same theoretical SNR. In that case, sure, more data is always better. But if you’re not summing the pixels then although you’d have four times as many with the 40 mp, the full well capacity would have to 1/4 of the 10 mp. Maybe not an issue at low ISO, but it’s going to limit you as you push higher, no?

The well capacity would be enough to be 1/4 but the light falling on each pixel is 1/4 as well, so there is no problem.

so the camp of my full frame captures more light really only captures more megapizels

More pixels with the same light per pixel = more total light.

Again, the mp number is irrelevant. The noise is part of the image itself -it has a discrete nature. Different pixel densities sample it in a different way but a lower pixel count is no better.

pi, I'm not, saying the,crop would be better, but rather not as much difference as it is today.

 

[Skirball]

I see what you’re saying, but that assumes that you’re combining photon counts from adjacent pixels, rights? E.g., if you grouped every four pixels together and counted the totals as a single pixel it would be equivalent to the 10 mp sensor with the same theoretical SNR. In that case, sure, more data is always better. But if you’re not summing the pixels then although you’d have four times as many with the 40 mp, the full well capacity would have to 1/4 of the 10 mp. Maybe not an issue at low ISO, but it’s going to limit you as you push higher, no?

The well capacity would be enough to be 1/4 but the light falling on each pixel is 1/4 as well, so there is no problem.

so the camp of my full frame captures more light really only captures more megapizels

More pixels with the same light per pixel = more total light.

Again, the mp number is irrelevant. The noise is part of the image itself -it has a discrete nature. Different pixel densities sample it in a different way but a lower pixel count is no better.

Large FWC has no impact on high iso, only at base/low iso together with the number of the read out noise
QE ,Efficiency per unit area is the interesting part and offcourse the sensor size.
Latest sensors from Canon has a QE around 50% compared with old 5d and 25%

Doesn't larger FWC allow more DR? Of course, as I type this I'm thinking about the D800 its DR. Ah hell, I'm going back to using the green square mode and not thinking about this crap.

 

[Pi]

Well yes, that was my point. You would have to reduce the FWC to 1/4 to maintain exposure. So, you increase the ISO two stops. Which would increase the noise at each pixel.

I do not know what FWC is but the ISO has no real meaning per pixel. You convert the physical signal to a digital one. Yes, any given value in the RAW file would correspond to less photons, so in this sense, you amplify 4 times more. But then each pixel is involved with 1/4 the weight in the final image, which I always consider to be of the same pixel size, regardless of the sensor. Those two effect cancel each other, and the net effect is zero.

However, the thing I admittedly never thought about until reading this thread: If you then downsample the image in post to the same size as the 10 mp, do the pixels average out to give the same general level of noise as the 10 mp sensor? The more I think about it, at low ISO the answer has to be yes. But, if you’re really pushing the high ISO are the results so far off that it will screw with the final average?

No difference whether you are pushing the ISO or not. The benefits of more pixels is that you can do downsampling, when needed, in a much more intelligent way that binning 2x2 blocks. You can do better NR, etc.

All this is about the photon noise. The readout noise is another factor, and the ability of the manufacturer to make high mp sensors with good QE is yet another one.

 

[dolina]

I hope it comes by January. I also hope for a 5D Mark IV or 1DX Mark II.

 

[9VIII]

I am not a physicist, nor do I pretend to comprehend most of what some you fine folks say in that and other highly technologically advanced areas with respect to cameras. Those of you that debate these matters truly amaze me. Truly impressive depth of knowledge here, and I mean that sincerely.

That said, as a consumer of Canon products and basically an amateur enthusiast who just wants to get better and better at photography so I can take great photos in virtually all conditions, I cannot fathom why Canon would make a product such as the 7d2 "better" than the 1DX. I am not taking about "better value" mind you, I mean to say "better" period. To my logic, if Canon could do that, they would be doing it and THAT would be their flagship Professional camera body. But there are many people that I have heard / read both here and elsewhere who not only claim that the 7d2 is going to be the APS-C version of a 1DX producing similar images to the 5d3 but that it will cost $4,000 - $5,000 LESS than a 1DX and $1500 less than a 5d3.

Why on earth would Canon do this? The 7d2 will probably be a great camera (it better be) but if buying one eliminates the need for Pro level photographers and enthusiasts to have to buy any FF camera whatsoever then Canon would be committing business suicide. I can list a few reasons as to why they might do this, but NONE of them make logical sense from a business perspective.

I think the 7d2 will be an awesome camera, and it will probably be the perfect marriage to pro shooters or advanced enthusiasts currently using one of the 5 series or 1 series FF bodies. But "replace" those FF cameras it will not.

I think part of what may be missing from your equation is that it's virtually impossible to make crop sensor IQ as good as full frame. Maybe if you compare the 70D with the 1Ds you would get a better shot off the crop camera, but anything produced within a similar time frame is going to be drastically different. Full Frame is significantly better. No one looking for the best images overall is going to get the crop camera, unless they would be cropping anyway.
Even if the 7D2 matches the 1Dx in every aspect but IQ, people will still want the 1Dx. As mentioned, it would also indicate the impending release of a significantly better full frame sensor.
There might be a handful of wildlife shooters willing to pay 1Dx prices for a pro-level crop camera, but that market is probably around two dozen people. The number of people who can't afford a 1Dx and would love to give up some IQ for the same features in a less expensive package are, well, look at the popularity of the original 7D.

 

[Eldar]

To me a 7D would be a bird/wildlife backup camera to the 1DX. The most important thing for me, compared to what the current 7D provides, in order of priority:
1: AF. If it gets the 5DIII AF system, it will be a massive improvement.
2: High ISO performance. If it gets to 5DIII standard, I´ll be very happy
3: Dynamic range. I would love to see the DR number grow by at least 1.
4: Resolution. Provided I can get the above, I would want as much as possible. But I would not trade Noise/DR performance for more resolution.
5: fps. 8 is OK, 10 would make me happy
6: A good silent shutter mode, similar to or better than 5DIII. The 1DX machine gun can scare off grizzlies.

 

[renegade54]

I am not a physicist, nor do I pretend to comprehend most of what some you fine folks say in that and other highly technologically advanced areas with respect to cameras. Those of you that debate these matters truly amaze me. Truly impressive depth of knowledge here, and I mean that sincerely.

That said, as a consumer of Canon products and basically an amateur enthusiast who just wants to get better and better at photography so I can take great photos in virtually all conditions, I cannot fathom why Canon would make a product such as the 7d2 "better" than the 1DX. I am not taking about "better value" mind you, I mean to say "better" period. To my logic, if Canon could do that, they would be doing it and THAT would be their flagship Professional camera body. But there are many people that I have heard / read both here and elsewhere who not only claim that the 7d2 is going to be the APS-C version of a 1DX producing similar images to the 5d3 but that it will cost $4,000 - $5,000 LESS than a 1DX and $1500 less than a 5d3.

Why on earth would Canon do this? The 7d2 will probably be a great camera (it better be) but if buying one eliminates the need for Pro level photographers and enthusiasts to have to buy any FF camera whatsoever then Canon would be committing business suicide. I can list a few reasons as to why they might do this, but NONE of them make logical sense from a business perspective.

I think the 7d2 will be an awesome camera, and it will probably be the perfect marriage to pro shooters or advanced enthusiasts currently using one of the 5 series or 1 series FF bodies. But "replace" those FF cameras it will not.

I think part of what may be missing from your equation is that it's virtually impossible to make crop sensor IQ as good as full frame. Maybe if you compare the 70D with the 1Ds you would get a better shot off the crop camera, but anything produced within a similar time frame is going to be drastically different. Full Frame is significantly better. No one looking for the best images overall is going to get the crop camera, unless they would be cropping anyway.
Even if the 7D2 matches the 1Dx in every aspect but IQ, people will still want the 1Dx. As mentioned, it would also indicate the impending release of a significantly better full frame sensor.
There might be a handful of wildlife shooters willing to pay 1Dx prices for a pro-level crop camera, but that market is probably around two dozen people. The number of people who can't afford a 1Dx and would love to give up some IQ for the same features in a less expensive package are, well, look at the popularity of the original 7D.

Thank you for your reply, and I agree 100% with what you stated. Despite the logic of your statement I am still hearing in this forum as well as all over the fruited plain that the 7d2 is going to be "better" than a 1DX, or have the same IQ as a 5d3. Now while I am certain the more technical versed CR members could give me NUMEROUS examples of how the 7d2 may (upon actual release and having photos to view of course) can produced images comparable to the 5d3 in certain circumstances; but that would not translate to the 7d2 being so much as = the 1DX let alone "better" than it. It might be close, and of course it will have it's place as several other posters mentioned. Nonetheless, these iterations of cameras are still too close to each other (in respective ages) for a lower tiered body to suddenly overtake one much less TWO of his bigger, in every sense of the word, brothers.

 

[Loswr]

Despite the logic of your statement I am still hearing in this forum as well as all over the fruited plain that the 7d2 is going to be "better" than a 1DX, or have the same IQ as a 5d3.

This is the Internet, where you can find 'evidence' that humans are descended from extraterrestrials and 'proof' of the existence of unicorns that poop rainbows. Just sayin'.

 

[Chosenbydestiny]

Despite the logic of your statement I am still hearing in this forum as well as all over the fruited plain that the 7d2 is going to be "better" than a 1DX, or have the same IQ as a 5d3.

This is the Internet, where you can find 'evidence' that humans are descended from extraterrestrials and 'proof' of the existence of unicorns that poop rainbows. Just sayin'.

+1 Thanks to DXO the internet, the quality of information is at an all time low =P

 

[unfocused]

Okay, since I've been soundly mocked for suggesting a 1-2 stop improvement in ISO performance for a 7DII, I'm a bit stunned.

Is it the consensus of the techies on this forum that is truly impossible for any manufacturer to produce an APS-C sensor in which the noise level at 1,200 ISO is comparable to 400 ISO (a 1.5 stop improvement)? Or, for that matter, that ISO 800 could look as good as ISO 400 (a one stop improvement)?

That doesn't seem like an unrealistic expectation, but apparently it is.

If that is truly the case, then I have to wonder what is the point of any future upgrades? Yes, I understand there is more to a camera than the sensor and even with sensors there is the whole quagmire of dynamic range that I'd rather not get into. But, given some of the real world and highly impressive images I've seen on this forum and elsewhere shot with the 5DIII at, what for me as a former film shooter, are remarkable ISOs (3200, 6400 etc.) it just doesn't seem unreasonable to expect the 7DII to boost its ISO performance by even one stop.

If we are truly at the physical limits of sensor performance and can never get an additional fstop out of any future sensor, what does that mean for the industry? And, as a side note, given that Canon is poised to jump into the surveillance market in a much bigger way, is this a fool's errand if we have reached the limits of low-light performance?