Is my PC powerful enough for DAW/VST composing?

[SpicedCider]

Over the last few months, I have been considering the purchase of a hardware workstation keyboard. I have an old Fantom-S that I got back when I was in high school (time flies...), and I was initially firm in my insistence on getting another hardware board until I heard demos of several VST sound libraries. HOLY CRAP!

 

I cannot believe how realistic, big-budget, and "ready-for-cinema" VST libraries such as Hollywood Strings, Goliath, and Symphobia all sound. Wow... I think that if I had even just those three libraries (or maybe trade out Goliath for EW Symphony Orchestra?) in my possession, I wouldn't need to invest in another sample library for years. I mean, how does it get anymore epic and movie-sounding than that?!

 

However, I realize that DAWs and VSTs are highly resource-intensive, and so I just wanted to post the technical specifications of my PC to gain an assessment of how adequate my PC is for composing with those components. Here are the copied-and-pasted specs (it's the HP 400f):

 

AMD Phenom II 945 Quad-Core Processor (3.0 GHz, 2MB L2 + 6MB Shared L3 Cache, 4000MHz System Bus);

 

8GB PC3-10600 DDR3 SDRAM memory (4x2GB) (expandable to 16GB);

 

1TB (Terabyte) 7200 RPM Serial ATA hard drive;

 

Windows 7 Home Premium 64-bit

 

 

 

So, what do you guys think? Will my "rig" cut it as-is? Or will I need a better processor and more RAM?

[zzzxtreme]

looks fine to me, although some would say a second drive would be beneficial.

[JeffLearman]

I started using Windows PCs for DAW back when it was 750 MHz with 256M of memory.

 

The question isn't whether your PC is powerful enough. The question is what are its limits. If you can run browsers and other applications without being annoyed, you can definitely do DAW work.

 

If you don't have enough CPU power, there are workarounds. For example, when you run out of CPU power to play all your tracks (with VSTi's, plugin FX, etc), you pick a subset of tracks and "freeze" them: render them to audio, muting the VSTi's and FX, and playing the rendered audio. Do that for (say) 8 tracks and you free up enough CPU power for nearly 8 more tracks of whatever it was you were doing.

 

A second hard drive is only important if you'll be recording lots of audio tracks at once. In that case it's nice to separate audio from everything else (program/OS/disk cache). But for a decent hard drive, you should have no problems recording up to 16 tracks at 44.1kHz/24bit format. (I have, on a laptop, without dropouts, and that was several years ago on slower gear.) It can also be useful if you'll be playing back LOTS of audio tracks, say, more than 24 or 30 (at 44.1/24b).

 

However, the amount of CPU power you need really depends on what plugins you'll be using. You'll soon find out which are the hogs and which aren't. For example, many years ago I used Arturia's moog emulator, which you could dial up a serious number of oscillators etc running at once without really even realizing it, and it would gobble up CPU like nobody's business for a single track. Again, that was on much slower, single CPU core processor, and by today's standards that might be a light CPU user. The point was while I had plenty of CPU power for FX, soundfonts, VSTi's like NIB4, and audio tracks, one track of synth could bring my CPU to its knees. It's still true today depending on what you load on a track.

 

Don't let what you have slow you down. Set it up and see what it's good for, and use that experience to judge what your needs are.

[JeffLearman]

PS: DAWs aren't particularly resource-intensive (if they're well-written). It's the FX and instruments that use the CPU power. That is, it's not the tool, but what you do with it, that eats up resources.

[SpicedCider]

Thanks for the explanation, learjeff. I suppose I'll learn with time and experimentation whether my rig is technologically sophisticated enough to serve as a DAW/VST workstation.

 

I am still in awe over how utterly realistic and natural modern VSTs sound. I haven't demoed a VST since 2005, so it's amazing to be able to conjure "that movie soundtrack sound" with the press of a key. Wow! Apparently, a vast proportion of the music I've heard over the last few years in movies and videogames that I thought just had to have been played by real people in real orchestras was actually "just" VST-based compositions.

 

It makes me wonder; with AMD's FX lineup of 6- and 8-core processors scheduled to be released over the next week, where will VSTs go from here? How will VSTs released in the future take advantage of these more powerful processors? I almost have to ask... how can they POSSIBLY sound anymore realistic? (Seems like a good idea for a new thread...)

[ChipCurtis]

 

I almost have to ask... how can they POSSIBLY sound anymore realistic?

 

 

People have been asking that since the beginning of synthesis & sampling. I don't think the question will EVER stopped being asked. And new product will NEVER stopped being made which claims to be the "end-all and be-all of realism".

 

Practically speaking, however, we've seen what happens when digital technology gets closer and closer to the elusive dream: People actually get fatigued by "too much realism" and revert back to the "warm" imperfection of earlier instruments which attempted this realism, or some facsimile of it.

 

Witness the uptick in the popularity of the sounds of the Mellotron, clonewheel organ, Rhodes piano, and many other instruments whose original purpose was to imitate strings, pipe organ, and acoustic piano.

 

When you infer that you are "finally happy with the realism of VST's", what you are saying is that you are impressed by the technology in its current state, not by some standard of "perfection" -- because there will always be better "perfections" down the road. Weren't we all impressed when we heard the "realism" of Kurzweil's K250 piano sound back in 1984? "How much better can it get" was what I heard from so many professional keyboardists back then, including Patrick Moraz when a friend and I interviewed him after a concert.

 

If you like what you hear from current VST libraries, then by all means go for it. But to ask the question "how can they POSSIBLY sound any more realistic" is like saying "how can Einstein's theories POSSIBLY be improved upon?" The future always brings new problems and perceptions of problems that can be further and further refined down the road.

[SpicedCider]

I see what you mean. I think what I'm so amazed by with the current generation of VSTs is that, whereas in the past, I could definitely tell when a composer for a game/movie had used samples, I literally can't definitively assess whether I'm listening to samples or the real thing nowadays.

 

I think that in my own mind, I always considered a synthesizer that could perfectly emulate real instruments to be the perfect composing platform -- you know, the "ideal" that synthesizer companies had always worked to achieve. Now it seems like they're finally there (although it wasn't achieved by the synthesizer companies). Yeah, I've been out of the loop. :-)