Power Technologies iDFX software, worth $40?

[Code Monkey]

Friday, iLounge ran an article in their news section about a new iTunes add-on, iDFX. I got into a back and forth with the progam's developer and, since iLounge would rather keep potential advertisers happy instead of me, turns out once again I got tossed into the "miserable user" queue (or whatever their article comment software calls it) where I "mysteriously" get errors whenever I go to post a new comment, but "mysteriously" only comments on subjects I've been a bit cantankerous about.

Since I don't want to have completely wasted a bunch of time creating the following demo, I figured I'd post it here:

http://www.keepandshare.com/doc/view...d=1452693&da=y

There's three short 40 second mp3s in the zip file.

One is a true back and forth between the DFX processed audio and the non-processed audio like Paul originally showed in the YouTube video. The other two are "fakes". One is all DFX processed audio, but with alternating portions reduced in volume the identical amount the DFX processing boosted the volume. One is all non-processed audio, but with alternating portions amplified the identical amount the DFX processing boosted the volume. See if you can tell which is which and if it really sounds like there's a real difference between the processed and non-processed audio once the trickery of non-matched volume levels is accounted for...

[ScoobZ]

My guesses:

Sample 1: all DFX processed audio
Sample 2: back and forth between the DFX processed audio and the non-processed audio
Sample 3: all non-processed audio

I'll state my preference after I find out how I did. :-)

[Code Monkey]

Alright, it took one week to get one reply and only four people ever downloaded it, w00t! Informed consumerism at its (pathetic) best

I'll give it 24 hours to see if that changes and then I'll let you know.

Thanks for taking the time to actually give it a look

[ScoobZ]

Originally Posted by Code Monkey

Thanks for taking the time to actually give it a look

NP. I d/l it on the 13th but only just today got to check it out.

[ScoobZ]

The suspense is growing. :-)

[Code Monkey]

Originally Posted by ScoobZ

The suspense is growing. :-)

Oh yeah, meant to answer this yesterday and brain farted after I loaded the page it was on...

You, both by merit and that sweetest of words, default, win the illustrious CM no-prize with 100% correct.

My question is which did you prefer and how big of a difference do you attribute to the DFX processing beyond the volume boost that can be achieved any myriad number of ways for free?

[ScoobZ]

Originally Posted by Code Monkey

Oh yeah, meant to answer this yesterday and brain farted after I loaded the page it was on...

You, both by merit and that sweetest of words, default, win the illustrious CM no-prize with 100% correct.

w00t!!! I am a winnah!!! lol

My question is which did you prefer and how big of a difference do you attribute to the DFX processing beyond the volume boost that can be achieved any myriad number of ways for free?

I preferred sample #3, the non-processed audio.

In samples #1 and #3 I was easily able to discern that the volume boost was just that, a straight across the board boost with no other variance between the segments. I was also easily able to discern the difference between the two samples, especially at the high end.

Volume boost aside, the DFX treatment brought in an overly bright, tinny element that I don't particularly care for. It seemed somewhat unnatural to me, if that makes sense.

[Code Monkey]

Originally Posted by ScoobZ

Volume boost aside, the DFX treatment brought in an overly bright, tinny element that I don't particularly care for. It seemed somewhat unnatural to me, if that makes sense.

That's exactly what I found as well. Spectral analysis showed that outside of the volume boost, the only universally noticeable difference is they dope/spike the audio with high frequency noise, the so-called "restoration" aspect, and that makes everything sound artificial to my ear. Then, if you zoom down to the micro second level of detail, you can see that they've also done a slight amplitude increase between the rise and fall in the wave form, I guess that's their "improvement of dynamics" - it's so slight, though, I doubt it's audible. I've attached a screen shot of two wave forms, the top is the unprocessed, the bottom is the processed - to the right where the audio gets more complex you can better notice the slight amplitude increase the processing does.

But really, that's it, a volume boost, some high frequency noise, and an extremely slight boost to amplitude changes in the wave form.

[ScoobZ]

Originally Posted by Code Monkey

But really, that's it, a volume boost, some high frequency noise, and an extremely slight boost to amplitude changes in the wave form.

Well, I suppose there is a portion of the listening population who find these changes pleasant and welcome. I can't be counted among them though.

Thanks for the pic, too.

[Code Monkey]

Originally Posted by ScoobZ

Well, I suppose there is a portion of the listening population who find these changes pleasant and welcome.

I'm skeptical: First, I'm not how many would notice the difference at all when levels are matched, it is a subtle difference. Then, when the software demo plays audio back to let you compare, what you hear is the rise and fall of the volume boosted processed audio compared to the original unboosted audio. Unless you already know enough to match the levels outside of their carefully staged comparison, your average mark is going to presume it not only sounds better, it sounds a lot better due to the "louder sounds better and more clear" principle. I'd be interested to know the results from an honest comparison, though.