Official specs: Canon EOS-1D X Mark II

Not me i dont want to have to sell 2 bodies just to buy 2 new ones. .. my bank account will hate me.

TeamSpeed wrote in post #17895857
Without a properly spec'd raw converter, I would be careful still on drawing conclusions. Even prior bodies were able to be processed when they first came out, but then 3rd part and Canon updated their raw software, and results came out better. The raw format itself may not have changed but how to translate some of the data inside each section might be different. However, I cannot see this getting any worse, only better, so who knows?

John is over on that forum too, so he will be able to give his thoughts here soon enough.

That comparison looks like the old Exmor vs Canon tests when Exmor first came out.

it's certainly a significant sign that "they've done it". I'd love to see this sensor technology in a 5D4, and even more so in a 5DsR2. I wonder if there are manufacturing hurdles to overcome with higher pixel density though.

TeamSpeed wrote in post #17895857
Without a properly spec'd raw converter, I would be careful still on drawing conclusions. Even prior bodies were able to be processed when they first came out, but then 3rd part and Canon updated their raw software, and results came out better. The raw format itself may not have changed but how to translate some of the data inside each section might be different. However, I cannot see this getting any worse, only better, so who knows?

John is over on that forum too, so he will be able to give his thoughts here soon enough.

That comparison looks like the old Exmor vs Canon tests when Exmor first came out.

The improvement looks very real to me. I don't think it will be worse with a new version of RAWTherapee.

frankchn wrote in post #17895848
There are some interesting new samples posted by a forum member at dpreview here: http://www.dpreview.co​m/forums/post/57265121

Specifically, they managed to process 1DX2 RAW files using RawTherapee and compared it to a 1DX, with both at ISO 100 and +5 EV in post -- http://3.static.img-dpreview.com …0b77543ee92496b​b1630a1c8b

Looks pretty promising to me .

Holy monkeys. That's a significant difference.

Yup. Look at the retention in the blacks and darks. Bout dam time

frankchn wrote in post #17895903
If the improvement holds up, I am interested in how fast Canon can push this sensor technology across its entire line up.

I do hope Canon pushes it to all new models introduced. When I got my 1Dx my wife got the 5D2 as a hand-me-down. She doesn't like the heft of the 1 series bodies so I'll be selling the 1Dx in time, but if the next 5 series is a beneficiary of these improvements then I'll seriously consider getting her (us) one...

jwcdds wrote in post #17895970
Holy monkeys. That's a significant difference.

it looks like.....

it came from a sony sensor

Canon did use a Sony sensor in the 1DXII, this is why they decided to go all hush hush on the sensor and pull back any notes about ADC on-chip... That explains this completely!

Just buy an adapter for it.

frankchn wrote in post #17895848
There are some interesting new samples posted by a forum member at dpreview here: http://www.dpreview.co​m/forums/post/57265121

Specifically, they managed to process 1DX2 RAW files using RawTherapee and compared it to a 1DX, with both at ISO 100 and +5 EV in post -- http://3.static.img-dpreview.com …0b77543ee92496b​b1630a1c8b

Looks pretty promising to me .

This is really bad news for Nikon.

If they had done this change with the 1DX, I wouldn't have thought so, but the longer Nikon/Sony had the low shadow, high DR advantage, the more they pull in with customers, perhaps new customers that are just getting into photography, etc. This will stop the bleeding however, and could garner some Nikon market share back to Canon.

TeamSpeed wrote in post #17896062
If they had done this change with the 1DX, I wouldn't have thought so, but the longer Nikon/Sony had the low shadow, high DR advantage, the more they pull in with customers, perhaps new customers that are just getting into photography, etc. This will stop the bleeding however, and could garner some Nikon market share back to Canon.

At least Sony can differentiate themselves by being mirrorless, but what does Nikon say now? If the sensor quality is now about equal, what do they do better than Canon?

TeamSpeed wrote in post #17896062
If they had done this change with the 1DX, I wouldn't have thought so, but the longer Nikon/Sony had the low shadow, high DR advantage, the more they pull in with customers, perhaps new customers that are just getting into photography, etc. This will stop the bleeding however, and could garner some Nikon market share back to Canon.

If Canon really wants to run with this, they should allow the user to decide what gain to use with each ISO. Right now, they only offer HTP, which basically says, use the gain the camera usually uses for half this ISO, giving me a stop more headroom, and a stop less footroom. With models that had shadow banding, HTP could be very risky at low ISOs. Now that banding is mostly gone, and footroom is much deeper, why not allow HTP-like shifts larger than 1 stop?

Just today I was reading someone's complaints that burst speed on the original 1Dx drops at very high ISOs - that could be because the files are large after compression, because of non-compressible noise bits. You can get rid of those bits and replace them with highly-compressible zeros, by allowing ISO 51K from ISO 6400 gain, or even lower, but then there will be slightly more noise.

For someone like me who shoots manual with auto-ISO mostly, I could set the camera up like so:

Setting on left, gain used on the right:

50 100
64 100
80 100
100 100
125 100
160 100
200 100
250 100
320 200
400 200
500 200
640 200
800 400
etc;

You get the idea - the gain can lag as you go up the ISO scale, and the higher the ISO, the more you lag, the more highlight headroom you get for surprises, and the smaller the RAW files become. You could set up a table for this, and switch between it and a simple, global EC offset, or normal gain/ISO relations for people who don't want to be bothered with this.

I could see this, and that kind of flexibility would be outstanding especially for those that shoot a variety of things. They could have different settings stored on the card and pull them up for what they are shooting. If they don't need the fps speed, then they could get better ISO at the high end.

I took the 100% crop from the DPReview thread, and the more you work on the 2 files (being one image), the right side cleans up very well and you get contrast/color, etc, but there is nothing you can really do with the right side, it never really cleans up.

This shows the flexibility of the cleaner shadows, you can take them quite a bit farther than you can with the noisy equivalent. I hadn't really wanted a 1DX2, but crap, now I really want one. I think I am going to get my 5D3 ready to sell this weekend just to free up some funds, and since getting the 7D2, I haven't really used the 5D3 much.

TeamSpeed wrote in post #17896121
I could see this, and that kind of flexibility would be outstanding especially for those that shoot a variety of things. They could have different settings stored on the card and pull them up for what they are shooting. If they don't need the fps speed, then they could get better ISO at the high end.

I took the 100% crop from the DPReview thread, and the more you work on the 2 files (being one image), the right side cleans up very well and you get contrast/color, etc, but there is nothing you can really do with the right side, it never really cleans up.

This shows the flexibility of the cleaner shadows, you can take them quite a bit farther than you can with the noisy equivalent. I hadn't really wanted a 1DX2, but crap, now I really want one. I think I am going to get my 5D3 ready to sell this weekend just to free up some funds, and since getting the 7D2, I haven't really used the 5D3 much.

use to be a marketing ploy from Acura; "if you drive the Acura, you will buy the Acura"

.......and I did