Is the current EP4CE40F23C8 using the F484 package?
Does that mean it would be possible to mount a EP4CE115 on the LimeSDR?
Is the current EP4CE40F23C8 using the F484 package?
Does that mean it would be possible to mount a EP4CE115 on the LimeSDR?
Or to put it in a different way: will the LimeSDR support the biggest FPGA in the Cyclone IV E family?
@zener, it will support EP4CE30 only. The reason - bigger brothers uses GND pins instead of signal within the same package.
Thank you Zack.
But wouldn’t it have worked if the pins were allocated first for the 280/103 pins on the E115 device and then migrated to the 328/124 pins on the E30/E40? Of course you never know until you try to do the pin migration in Quartus…
@zener This is were we started. But almost all the pins are used in case of EP4CE40F23 device. Hence no chance to fit to the same package but less pins.
I searched for “set_location_assignment” and 226 PINs (including DDR2 assignments) were assigned in the project.
So, it looks like @zener made an interesting point as “wouldn’t it have worked if the pins were allocated first for the 280/103 pins on the E115 device and then migrated to the 328/124 pins on the E30/E40”
I would guess that the Blade RF SDR digital designers have had this option in mind to make the board compatible with 40K ( EP4CE40F23C8) and 115K ( EP4CE115F23C8) devices.
If this is feasible, couldn’t a new PCB revision be compatible with E115 device via migration?
It would be interesting to play with HW accelerators in the big brother FPGA.