Soundcards vs onboard performance

Deff worth the cost if you have nice set of headphones etc

Very little I'm assuming. I ordered a creative soundcard with my new system, onboard is eh.

I personally have never seen hard numbers, comparing a sound card vs integrated sound and its cpu impact. I guess extensive Google searching is needed to try to find the benches you want.   Not sure its even been done because of its minimal impact.    

Like bprat22, I have not seen hard numbers, not since around 1995 or 1996 or so. Back then, with single core larger nm CPUs, integrated sound cards would reduce the clock cycle by ~5% dependent upon the bit rate of the sound. So approximately 5 out of every 100 cycles would be spent on sound, rather than other calculations. The on-board sound (for the few motherboards that had them) was generally of lower quality, far more so than today's onboard, so most opted for the dedicated in other to get better, more fluid Hertz patterns, etc. It wasn't really a performance issue, even for a single core.

So, in today's world of smaller nm, multi-core, layered cycle CPUs, integrated sound is an inconsequential speck in a clock cycle. I really think that in order to tell the difference, in say, a game, you would need to analyze the FPS at a level in the realm of 1/10th a frame or less. Of course this is conjecture, but I can say from personal experience with an older Core 2 Duo desktop I had, that removing the dedicated soundcard made no difference, not a single point, in benchmarks.

The best way to think of it, is that modern CPUs are rarely used at 100%. The only way that onboard sound would reduce performance would be if the CPU was at 100% and then needed to allocate resources to something beyond that 100%. 

But, a thorough Google search might reveal the differences in hard data you desire, but someone would need to be testing for that specifically, as it would be hard to control based on the small differences. Just be wary of the multitudes of people who claim to have gotten "x%" performance increases by buying XYZ soundcard, I have personally never seen anyone substantiate these claims, and I generally believe them to be buyer's justification, and observation bias.


Edited by Tidgxor - 27 Jun 2013 at 5:47pm

I have an extra Creative X-Fi (I can't remember which model specifically, would have to check) that I took out of another system. When my new system gets here, after I get some of the more important things out of the way I might drop the card in and disable the on-board to see if I can detect a difference. This all assumes Windows 8 properly supports the card since its coming out of a Windows 7 box.

I am trying to think of a good test to run with it. Something that utilizes audio during the benchmark (most of the ones like 3DMark don't utilize sound). About the only test that comes to mind might be the FarCry2 benchmark tool. Anyone have any other ideas on an easy-to-run benchmark?


Edited by Bucko - 27 Jun 2013 at 9:09pm