A C++ based renderer that uses the Embree ray trace engine. Tungsten does not build on my AMD processor and gives an error of

CMake Error at CMakeLists.txt:39 (message):
  The target machine does not support SSE3.  At least SSE3 is required


-- Configuring incomplete, errors occurred!

As best I can tell, the code hs an OptimizeForArchitecture.cmake file that has hard-coded a set of architectures that are supported. I see “Zen” listed but not anything newer. So I suspect a bogus check that prevents it from building.

It does run on my Intel CPU so I am able to get the following metrics

elapsed              725.204
on_cpu               0.727          # 11.64 / 16 cores
utime                8433.549
stime                4.729
nvcsw                14841          # 8.50%
nivcsw               159852         # 91.50%
inblock              43192          # 59.56/sec
onblock              22744          # 31.36/sec
cpu-clock            8438307165810  # 8438.307 seconds
task-clock           8438363433623  # 8438.363 seconds
page faults          1916187        # 227.080/sec
context switches     178096         # 21.106/sec
cpu migrations       5143           # 0.609/sec
major page faults    127            # 0.015/sec
minor page faults    1916060        # 227.065/sec
alignment faults     0              # 0.000/sec
emulation faults     0              # 0.000/sec
branches             3271423353475  # 94.584 branches per 1000 inst
branch misses        194884295986   # 5.96% branch miss
conditional          3271423394947  # 94.584 conditional branches per 1000 inst
indirect             948203547518   # 27.414 indirect branches per 1000 inst
slots                26587693913018 #
retiring             11919863303231 # 44.8% (44.8%)
-- ucode             544757540566   #     2.0%
-- fastpath          11375105762665 #    42.8%
frontend             6358475987961  # 23.9% (23.9%)
-- latency           3394932472576  #    12.8%
-- bandwidth         2963543515385  #    11.1%
backend              3211618940012  # 12.1% (12.1%) low
-- cpu               1417917583347  #     5.3%
-- memory            1793701356665  #     6.7%
speculation          5133673255151  # 19.3% (19.3%) high
-- branch mispredict 4903411672670  #    18.4%
-- pipeline restart  230261582481   #     0.9%
smt-contention       0              #  0.0% ( 0.0%)
cpu-cycles           13873833874847 # 1.82 GHz
instructions         19579717111297 # 1.41 IPC
l2 access            186773491595   # 16.065 l2 access per 1000 inst
l2 miss              46766340321    # 25.04% l2 miss