Buzz Benchmark
Posted: Wed Mar 18, 2015 7:38 pm
I just poured some more money into my computer, and I've finally gotten a respectable cooler so I'm starting to overclock, all with the intention of making my computer a better musical instrument. In the past, the methodology of stress testing your pc was to use a synthetic benchmark like intel burn test/linpack to push the cpu to it's absolute limits, but although these tests are still useful in varying degrees for determining stability, it seems more people are turning to real world benchmarks that test the cpu using the programs that will stress it in the way it will be stressed by the tasks you intend use it for!
I once encountered a proposed audio oriented benchmark using reaper and heaping it full of oscillators and samples and things, but I don't use reaper.
So I'd like to test my cpu for audio fidelity and stability and find the limits of what I'm going to be able to do in buzz using this machine of mine using a buzz benchmark.
So here's my question for all of you.
How should this benchmark be constructed? It needs to use many cpu intensive, but rock solid synths, should likely incorporate pvst, since we all use that and should probably be in the form of some kind of song so we can test machines becoming active and inactive (a failure point in my experience).
If we get a semi-standard bmx of this floating around it would be fun to compare results, possibly with audio samples. Another fun thing about it is that if there's any machine that you can't live without, you can put 10^10 of them in there and see what happens for you %D
Off the top of my head I just want to get eleventy-billions of Jacinths going into eleventy-billion Urani and then play it with an hd 2p spec filter~!
I once encountered a proposed audio oriented benchmark using reaper and heaping it full of oscillators and samples and things, but I don't use reaper.
So I'd like to test my cpu for audio fidelity and stability and find the limits of what I'm going to be able to do in buzz using this machine of mine using a buzz benchmark.
So here's my question for all of you.
How should this benchmark be constructed? It needs to use many cpu intensive, but rock solid synths, should likely incorporate pvst, since we all use that and should probably be in the form of some kind of song so we can test machines becoming active and inactive (a failure point in my experience).
If we get a semi-standard bmx of this floating around it would be fun to compare results, possibly with audio samples. Another fun thing about it is that if there's any machine that you can't live without, you can put 10^10 of them in there and see what happens for you %D
Off the top of my head I just want to get eleventy-billions of Jacinths going into eleventy-billion Urani and then play it with an hd 2p spec filter~!