MicheleANE Posted July 19, 2018 Share Posted July 19, 2018 Hi,The possibility of multithreading the shading computation has been a very exciting news for us because we usually work on very large scale project that usually required quite long computation times.But, from tests we did internally, it would seem that all the threads used for the calculation allocate memory in the same 2GB space that is the limit of 32 bits architectures. This creates situations in which having more cores is a disadvantage because it drastically multiplies the used memory and , when it passes the 2GB limit, brings crashes.As an example, I tested this on a model that usually required around 300MB memory during shading computation using single core. Enabling multithread computation (8 cores/16 threads system with 64GB memory) you can see, after pressing the table button, both the percentage of used CPU (therefore the number of cores) and the used memory going up and up in a short time and when the memory almost reaches 2GB, the CPU usage drops to 0% and the process freezes and closing the software is the only possibility.Now, one solution would be to compile PVsyst in 64 bits so it can access the entire memory of the system that is then enough to support multiple threads. Is this possible? Are there other possibilities?Thanks for your consideration.Best regards Link to comment Share on other sites More sharing options...
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!Register a new account
Already have an account? Sign in here.Sign In Now