CUDA never uses two GPUs

Hardware
CPU 4930K
NVIDIA 780 x 2
32 gb ram

Software
Windows 8.1 64 bit
CUDA 7.5(this installs NV driver 353.90)
VS 2013

Samples properly compile.
deviceQuery shows

deviceQuery.exe Starting...

CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 2 CUDA Capable device(s)

Device 0: "GeForce GTX 780"
CUDA Driver Version / Runtime Version 7.5 / 7.5
CUDA Capability Major/Minor version number: 3.5
Total amount of global memory: 3072 MBytes (3221225472 bytes)
(12) Multiprocessors, (192) CUDA Cores/MP: 2304 CUDA Cores
GPU Max Clock rate: 1020 MHz (1.02 GHz)
Memory Clock rate: 3004 Mhz
Memory Bus Width: 384-bit
L2 Cache Size: 1572864 bytes
Maximum Texture Dimension Size (x,y,z) 1D=(65536), 2D=(65536, 65536), 3D=(4096, 4096, 4096)
Maximum Layered 1D Texture Size, (num) layers 1D=(16384), 2048 layers
Maximum Layered 2D Texture Size, (num) layers 2D=(16384, 16384), 2048 layers
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 49152 bytes
Total number of registers available per block: 65536
Warp size: 32
Maximum number of threads per multiprocessor: 2048
Maximum number of threads per block: 1024
Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
Max dimension size of a grid size (x,y,z): (2147483647, 65535, 65535)
Maximum memory pitch: 2147483647 bytes
Texture alignment: 512 bytes
Concurrent copy and kernel execution: Yes with 1 copy engine(s)
Run time limit on kernels: Yes
Integrated GPU sharing Host Memory: No
Support host page-locked memory mapping: Yes
Alignment requirement for Surfaces: Yes
Device has ECC support: Disabled
CUDA Device Driver Mode (TCC or WDDM): WDDM (Windows Display Driver Model)
Device supports Unified Addressing (UVA): Yes
Device PCI Domain ID / Bus ID / location ID: 0 / 1 / 0
Compute Mode:
< Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

Device 1: "GeForce GTX 780"
CUDA Driver Version / Runtime Version 7.5 / 7.5
CUDA Capability Major/Minor version number: 3.5
Total amount of global memory: 3072 MBytes (3221225472 bytes)
(12) Multiprocessors, (192) CUDA Cores/MP: 2304 CUDA Cores
GPU Max Clock rate: 1020 MHz (1.02 GHz)
Memory Clock rate: 3004 Mhz
Memory Bus Width: 384-bit
L2 Cache Size: 1572864 bytes
Maximum Texture Dimension Size (x,y,z) 1D=(65536), 2D=(65536, 65536), 3D=(4096, 4096, 4096)
Maximum Layered 1D Texture Size, (num) layers 1D=(16384), 2048 layers
Maximum Layered 2D Texture Size, (num) layers 2D=(16384, 16384), 2048 layers
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 49152 bytes
Total number of registers available per block: 65536
Warp size: 32
Maximum number of threads per multiprocessor: 2048
Maximum number of threads per block: 1024
Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
Max dimension size of a grid size (x,y,z): (2147483647, 65535, 65535)
Maximum memory pitch: 2147483647 bytes
Texture alignment: 512 bytes
Concurrent copy and kernel execution: Yes with 1 copy engine(s)
Run time limit on kernels: Yes
Integrated GPU sharing Host Memory: No
Support host page-locked memory mapping: Yes
Alignment requirement for Surfaces: Yes
Device has ECC support: Disabled
CUDA Device Driver Mode (TCC or WDDM): WDDM (Windows Display Driver Model)
Device supports Unified Addressing (UVA): Yes
Device PCI Domain ID / Bus ID / location ID: 0 / 2 / 0
Compute Mode:
< Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 7.5, CUDA Runtime Version = 7.5, NumDevs = 2, Device0 = GeForce GTX 780, Device1 = GeForce GTX 780
Result = PASS

As you can see, both gpus are identified. But, bandwidth test shows only 1 gpu used

[CUDA Bandwidth Test] - Starting...
Running on...

Device 0: GeForce GTX 780
Quick Mode

Host to Device Bandwidth, 1 Device(s)
PINNED Memory Transfers
Transfer Size (Bytes) Bandwidth(MB/s)
33554432 10766.6

Device to Host Bandwidth, 1 Device(s)
PINNED Memory Transfers
Transfer Size (Bytes) Bandwidth(MB/s)
33554432 11381.7

Device to Device Bandwidth, 1 Device(s)
PINNED Memory Transfers
Transfer Size (Bytes) Bandwidth(MB/s)
33554432 210790.7

Result = PASS

Running the mandlebrot example and monitoring gpu utilization with CPUID HW monitor shows 1 gpu used - utilization goes up only on one card.

Interestingly, if I enable SLI, then the OTHER gpu is used. But the load is never shared.
How can I troubleshoot this?

The application has to explicitly target each GPU using cudaSetDevice().
Unless the application supports multiple GPUs it will just use what it considers ‘device 0’ by default.
Some of the samples in the CUDA SDK do support multi-GPU applications, but evidently not this one you chose.

And do not use SLI (remove the bridge and use the NVIDIA control panel to remove the SLI option) when running CUDA applications, as (last I heard) it can cause problems.

tnx. simple as that. the nbody simulation example does indeed work