Comparing the EOS-1D X, EOS-1D X Mark II & EOS-1D X Mark III

[Profit007]

I just posted this in another thread, but I think it's also relevant here:

Like stills, there are different market segments. Those producing content have good options in the $12-25k range, or for under $10k there is the $2900 Pocket Cinema Camera 6K with no AF. It records straight to interchangable SSDs with better options than Canon Raw Light, 265, etc.

Then there is the content gathering crowd: documentary production, news, and live events like weddings. These folk need AF in a way that controlled productions don't. This numerically much larger market segment is never going to turn up at weddings with a pair of C500s, and Canon doesn't understand this.

The 1DxIII is only a worthy successor to the 1DC in that it was a Canon strategy to withhold tech and charge more of it.

Remember the 1DC was just a vanilla 1Dx with an extra heatsink, different firmware, and a $15k price tag.
It was a stunningly blatant cash grab from Canon which exposed that they really would switch off a feature in firmware unless you paid an extra $8.5k for it..... with all other hardware being the same. (And in this case it failed, sales were dismal.)

 

[Quirkz]

This comparison guide should also show the A9II.

* 24Mp,
* 20FPS,
* 693-Point Phase-Detection AF System
* real time eye tracking AF

In that case, lets bring in the m6 II as well:

* 32Mp
* 30 fps
* 5,481 point dual pixel AF system
* real time eye tracking.

I love numbers.

 

[SecureGSM]

... and, and.. M6 II is lighter and smaller

 

[Dragon]

Or its more likely that they would be using a cheaper NAND based memory buffer running on a faster NVMe memory controller, but still means a persistant state storage buffer. Canon couldn't claim a fixed "buffer" specification if they were relying on a users storage card and using a slower while more expensive removable storage spec like CFExpress for a dedicated internal buffer doesnt make sense to me (edit: perhaps there are some power consumption benefits with CFExpress). Either way it would mean the buffer would persist even after a power loss, interesting.

DPR did early testing using XQD cards and I didn't see any whimpering about the buffer. I think the buffer is real. What kind of memory is in the buffer is up for grabs. It could be LP RAM. It could be very high quality (I.e. log life) NAND. It even could be 3D Crosspoint memory from Micron. So far, Only Intel has released open market products with 3D xpt memory, but Micron is the manufacturer and they have the right to sell the stuff, so it is not out of the question. However, I do think that if the buffer was persistent, Canon would have made a big deal out of it, so most likely it is RAM for speed and ease of access (for as much concurrency in read and write as possible). In today's world that is only a few RAM chips.