Anderton
 
 
If it wasn't flawed, then I would think the results would be consistent and repeatable.
 
But beyond that, I don't think the way to measure the usefulness or performance of a DAW relates solely to how many plug-ins you can stack up. Think of it this way: You might like the way a car with a 1650 cc internal combustion engine handles better than a car with a 2000 cc internal combustion engine. Even though the latter is more "powerful," the one with the 1650 cc engine might have a lower center of gravity, tighter steering, and a transaxle that you can wind up to 65 MPH in third gear. Or it just might look cooler...or have a better sound system

And more so because plugins just don't hit the system very hard these days. We have tons of CPU resources. It is to the point you can stack convolution reverb as an insert effect if you like. It is just very rare that the issue is "Man, I just don't have the CPU power to load that effect I want!" Maybe if you are working on positively massive projects, but then you probably have a positively massive system which eliminates the problem.

Anderton

Goddard

Anderton
Yes, it's nice Sonar came out "on top"...but on top of what? A flawed test that has little, if anything, to do with the intended function of either piece of software?

 
Flawed in what way, may I ask?
 
(btw, in a later test run reported in that thread Sonar didn't come out "on top" anymore)  

If it wasn't flawed, then I would think the results would be consistent and repeatable.
 
But beyond that, I don't think the way to measure the usefulness or performance of a DAW relates solely to how many plug-ins you can stack up.

Craig, ok, glad to see it wasn't just an offhanded criticism being tossed out in a certain someone's direction over a past audio interface latency testing technique brouhaha...
 
Perhaps it's not really the DAWbench test itself which is flawed, but rather (as I strongly suspect) that GS poster's test setup and procedure. DAWbench results are ime pretty consistent across runs unless some change has been made to the platform or test setup between runs or something is screwed/screwing up.
 
The DAWbench test suite isn't solely about how many plug-ins (both DSP FX and VIs) can be run, although increasing the plug-in count is the method by which the test incrementally stresses a DAW system's performance. Audio interface driver efficiency (at different respective latencies) plays a significant role as well. The specifics of the test suite are given on the DAWbench site:
 
http://www.dawbench.com/benchmarks.htm
 
And as for measuring a DAW's usefulness and performance by testing with many plug-ins, well, let's just say that Cakewalk themselves have been known to do that!:
 
http://blog.cakewalk.com/windows-8-a-benchmark-for-music-production-applications/
 
And to promote based on it:
 
http://forum.cakewalk.com/Video-The-140Track-Cakewalk-SONAR-Session-m1836732.aspx
 
http://forum.cakewalk.com/SONAR-85-The-Fine-Print-m1841847.aspx
 
http://software.intel.com/en-us/articles/utilizing-intel-avx-with-cakewalk-sonar-x1
 
Anyway, if you or anyone at CW have suggestions/criticisms regarding DAWbench, I'd urge you/them to relate such directly to Vin/TAFKAT (he really does mean well, and I'm sure he would appreciate the input).