Overview Features Coding ApolloOS Performance Forum Downloads Products Order Contact

Welcome to the Apollo Forum

This forum is for people interested in the APOLLO CPU.
Please read the forum usage manual.
Please visit our Apollo-Discord Server for support.



All TopicsNewsPerformanceGamesDemosApolloVampireAROSWorkbenchATARIReleases
Performance and Benchmark Results!

160 MIpspage  1 2 3 

Thierry Atheist

Posts 644
22 Sep 2017 04:52


To really know how many million instructions per second are possible, the entire test should fit in the 32K of instruction and 32K of data caches.

Pure result.


Gunnar von Boehn
(Apollo Team Member)
Posts 6207
22 Sep 2017 09:32


Thierry Atheist wrote:

  To really know how many million instructions per second are possible,
 

 
Actually this is not important at all.
 
The real goal is to measure how much WORK can be done per second.
This is also the goal of the DHRYSTONE test.
The dhrystone test does NOT measure how many MIPS = million instructions per second your CPU does.
 
If you think that this is what SYSINFO tries to calc, then you are wrong.
 
The DHRYSTONE test runs a "defined workload" and compares the time taken - with how long a VAX computer need for it.
And it knew how many Million Instruction that VAX did execute for this.

So the MIPS printed in all these Dhrystone test is NOT the million instruction of your CPU - but how many MIPS a VAX would have needed to do this defined workload in the same time.
 
 


Mallagan Bellator

Posts 393
22 Sep 2017 17:40


So in other words, ASIC should be the final step for vampires, and computers like that will be faster than anything. Got it ;)

posts 43page  1 2 3