Click for Wireless Adapters and Bases!
Click for Wireless Adapters and Bases!
| Accelerate Your Mac! Bring in the Noise Mac Audio Column (Prev. Articles Listing) |
Do Different Players Strain Your CPU More or Less? By: Erik Lindahl Published: 5/14/2002 (Updated for reader comments below article and Panic's comments on Audion CPU usage) A few days ago MacAmp Lite was released for MacOS X. They state that one of their main features is that it uses less CPU-load than other available players. So, I asked myself if thsi is true. For the test three major music players for MacOS X were used: Apple's iTunes 2.04, Panic Audion 2.6 and MacAmp Lite 1.0. The system I used for this little experiment was:
My Findings:
Indeed, the comments [reader comments] are true -- Audion will run with significantly higher CPU when scrolling track/titles are turned on. This CPU load is only due to the overhead of updating the scrolling text frequently, which in Audion 2.6.1 actually requires a complete redraw of the face for each frame. The scrolling is turned off by default. The good news is that this has been optimized in the forthcoming Audion 3 so it shouldn't be so bad.
However, it's really doing our software an injustice to have the current
charts up there. It's not really fairly comparing the programs on a level
playing field. :)
I've asked Erik about this issue and he wrote:
To comment on what Will had to say: (Will's comments)"I just wanted to point out a possible oversite in the mp3 player testing. In Audion, if you turn off "scrolling" mp3 information in the interface preference tab it dramatically reduces the CPU load. Yes, this seems to be true. The fact is since only iTunes contiunes scrolling information even if it«s idle it did use some CPU. (Will's comments) For example, I'm streaming right now at about 18% CPU usage on a dual 450, but the test [Erik] ran states his average was 35% on a dual 500. 18% is much more inline with the other players. It also drastically affects the other tests, I would simply suggest noting that with the test data, or even retesting Audion. I did have scrolling OFF wile doing the test. I retested now again, getting similar results (this time however I did turn the Dock information off as well). Seeing from the terminal here is a clipping:
3543 Audion 2 28.5% 0:22.50 9 134 218 10.5M 11.8M 17.1M 79.3M 3527 iTunes 16.4% 22:27.79 9 147 202 8.28M 10.3M 12.3M 72.9M 3553 MacAmp Lit 10.0% 0:04.73 5 120 255 2.77M 9.93M 7.56M 69.9M As you can see... Audion is a HOG when it comes to RAM and CPU-usage. Turning EQ/Text scrolling off gave:
3543 Audion 2 20.3% 1:05.80 9 134 219 10.6M 11.8M 17.2M 79.3M 3527 iTunes 10.1% 22:52.61 9 147 207 8.55M 10.4M 12.9M 73.4M 3554 MacAmp Lit 7.7% 0:05.90 6 125 338 3.37M 11.2M 9.43M 72.2MAs you can see here I got similar figures to Will. The CPU-load would jump up and down a few %. The 35% is the PEAK I found for the particular APP (usually lasted for a second, then went back a few %). If will is getting a 18% CPU-load WITH EQ on there must be something wrong with my set-up or something. The test obove is wile streaming a 160kbs radiostation.
I used to like Audion too, however I've come like iTunes more and more.
Nicer/simpler interface, easier too use and administrate, uses less RAM and
is faster =) 
Fastest was in fact MacAmp Lite. The developer isn't bragging here. I found this player to often have way under 10% CPU load without any DSP effects on, where the runner-up iTunes usually was in the 10-12% range. Last was Audion using up to 20% of CPU power. In real world you probably want to turn on audio enhancements (DSP) such as an EQ [Equalizer]. This does in fact stress the CPU a lot more than not using them. Again, MacAmp Lite wins the CPU load test using around 10-14% of your CPU. Apple iTunes does a fair job, using a stable 15 where, as mentioned earlier, Audion has the highest CPU usage of 20-35%. 
In local MP3 playback vs streaming playback, CPU load varied from player to player. Itunes used a pretty even load where Audion needs a LOT MORE CPU for streaming audio. 
[ For info/links to MacAmp Lite - see the press release in Monday's (5/13/2002) news page.] Reader Comments on this Article:
"
Hi Mike,
For example, I'm streaming right now at about 18% CPU usage on a dual 450,
but the test Thad ran states his average was 35% on a dual 500. 18% is much
more inline with the other players. It also drastically affects the other
tests, I would simply suggest noting that with the test data, or even
retesting Audion. Another reader commented on MacAmp:
For me, since Sept, events in those lands between the Med and the Pacific are described so differently depending on the whom I'm listening to, that only after a little surfin' around do I feel I've at least some idea of what's happened now, maybe a few drops of clarity left over after boiling down all the broadcasted blarney... Oops, I digress. Unfortunately, the current PPC version of MacAmp is crippled nowadays, unable to use its tracker to update its lists--though all my existing bookmarks, if still valid, do work. But, were the tracker working right, it would quickly demonstrate why it's worth paying for despite the august presence of iTunes: When it comes to internet radio, MacAmp just lists many more stations than iTunes. I don't know why/how, but it does, bigtime. Or, it did, for me. If you ever have a free half hour (yeah right) I hope you get a chance to try out this new MacAmp vs iTunes. I think the pop music stations generally suck no matter what you're using to access them, but for foreign info and "world music" access, MacAmp's the best.
Jeez Mike, a lot of this almost reads like an ad of some kind....not
intended, I swear! I have no affiliation with MacAmp except for having
bought their program.
Audio Column Back Issues:
|
Your Source for the best in CPU/STORAGE/SYSTEMS/VIDEO card reviews, daily news, and more! and do not necessarily represent those of the site publishers. Read the site Terms of Use. All brand or product names mentioned here are properties of their respective companies. |