Arduino UNO - Rev4

[PCB.Wiz]

And I wonder if the CAN is actually good for anything as neither of the variants have a crystal.

Yes, a strange omission, but it appears the simpler variant has provided footprints to fit a crystal and caps ?
Why not allow that on both models ?

[ataradov]

Those boards have so many "minor" differences, it is almost like they were designed by different people based on the basic description.

Without a transceiver on the board, I doubt there would be a lot of interest in the CAN here anyway.

[Rudolph Riedel]

I am getting 23.81MHz on the SPI and the datasheet gives a range for the MOCO oszillator of 6.8 to 9.2 MHz.
I have a doubt that this will work for CAN running at 500kBit/s and 1MBit/s seems impossible.
I only do not have a CAN transceiver breakout-board to test this.

Everything else should be fine.
CAN is one of the advertised features for the UNO R4 though.

[SiliconWizard]

Without a transceiver on the board, I doubt there would be a lot of interest in the CAN here anyway.

That's an opportunity for selling more "shields".

[ataradov]

There are already shields for CAN with SPI-based controllers. And those are proven to work. So, if you have to buy a shield anyway, might as well get the SPI one and not worry about the clock issues.

[westfw]

Sigh.  Am I just used to a different style, or is the Renesas documentation really awful?

[ataradov]

It is some of the worst I've seen. To the point where I'm again no longer interested in doing anything with Renesas parts. If I have to read and parse awful documentation, it might as well be some random Chinese part, at least then I probably got it for a good price.