WouterJRV

Ok thanks, raised an Issue at the Julia gh itself

With StochasticDiffEq v6.36.0 it should be working, right? Unfortunately, I'm still getting the same error.

`ERROR: LoadError: OutOfMemoryError() Stacktrace: [1] Array at ./boot.jl:408 [inlined] [2] Array at ./boot.jl:416 [inlined] [3] zeros at ./array.jl:525 [inlined] [4] zeros at ./array.jl:521 [inlined] [5] alg_cache(::DRI1, ::SDEProblem{Array{Float64,3},Tuple{Float64,Float64},true,Tuple{Int64,Int64,Float64,Array{Float64,2},Float64,Array{Float64,2},Array{Float64,2},Int64,Array{Float64,2}},Nothing,SDEFunction{true,typeof(f!),typeof(g!),UniformScaling{Bool},Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing,Nothing},typeof(g!),Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}},Nothing}, ::Array{Float64,3}, ::Array{Float64,3}, ::Array{Float64,3},...

Yes, tried it earlier and it gives a similar error. didn't read throuth the stack trace because I assumed it was the same, but apparently that's not the case `ERROR:...

Yes, with SOSRA I got `508.131326 seconds (18.10 M allocations: 825.235 MiB, 0.09% gc time)` and with SOSRI `762.754098 seconds (23.18 M allocations: 1.012 GiB, 0.07% gc time)`

Ok, thanks for your answer. It seems that there were two different explanations for the deviation between the SDE output and the 'true' solution: either it was something 'deeper' that...

Ok, it seems that this is an issue of too wide tolerances. Just decreasing abstol and reltol can give rise to instabilities as well. If I'm correct, the default solver...

Thank you for the quick answer. Does it makes sense that occasionally, numerical errors occuring when the tolerance is made less strict, actually avoid divergences? Naively, I would expect that...

Thanks for your answer > If you look at it as a function of real and complex parts, it becomes a commutative noise problem (at least, 1D is easy to...

So there is no way to get the same performance to get Julia near the same performance of MATLAB? This seems to be a bit surprising to me since MATLAB...