Hmmm... What about VST instruments?

That would be great, thanks! IIRC, ModPlug Player played this module right, but I can't be sure, since I stopped using it quite a while ago (it's your fault, for writing XMPlay ).
Anyway (and I may be totally off here), wasn't there a threshold for the speed/bpm value below (or was it above?) which the module was considered VBlank? You could just pre-scan the pattern list at load time for those values (I'm assuming you already do it to calculate the song length, so it shouldn't be expensive at all).

XMPlay doesn't play this module correctly, for example (old timings).

So much for the "balls-on accuracy", heh?
Seriously, I believe it's a bug, since previous versions were perfect.

<edit>
Bug description: First six channels are mute.
</edit>

<edit2>
Oh, boy... Guess I should have scrolled a bit down. Anyway, Ian, why don't you ++ the least significant digit of the version number even for these trivial fixes?
</edit2>

Pure eye-candy, but... How about implementing this visualization mode for modules?
(I'm mainly referring to the channel/effect matrix.)

Most of those formats (all except Octamed I think) can be replayed by Winamp plugins. Check the support site.

I'm asking for internal format parsing, so it could use XMPlay's internal mixing routines. Those plug-ins (ModPlug/MikAmp) use their own mixers (redundant), and their accuracy leaves a lot to be desired (especially as far as OctaMED and Composer 669 are concerned).

Hi there, Ian!

One more thing I find XMPlay lacking (as a module player) is it's relatively limited module format support. My personal collection features some relatively exotic formats (Farandole Composer, Composer 669, Oktalyzer, OctaMed, Scream Tracker 1.0, Asylum/DSMI, Ultra Tracker, D-Lusion X-Tracker) that still cannot be played with XMPlay.
The documentation for the file format and effect conversion for all formats above, except for Asylum/DSMI, is available on Wotsit's Format.

Thanks, and sorry for being such a nag...

I'm not sure what you mean by "tunable" (I just see a "FIR" switch in Winamp), but sinc interpolation is a FIR filter.

Yes, I'm aware of that. Maybe it was an older version (I stopped using Winamp quite a while ago). I remember it had an advanced configuration dialog for the filter, with a combo-box (where you could select the windowing function, IIRC) and two number input boxes (one for the cut-off coefficient, but I don't remember what the other parameter was).
I requested this because I found that filter to resample at the highest quality I ever heard. Don't get me wrong, the sinc interpolator is very nice. It's just not perfect for very low quality samples (fails my Rhesus Minus' "Daisy Chain" test ).

As well as SSE2, it's also optimized for 3dNow. I didn't bother with optimizations for older technology, as I figured those CPUs wouldn't be fast enough anyway, but included the vanilla FPU version so that it could at least still be used when writing to disk on those CPUs.

My P3 @ 650 MHz still has enough computational power to play every module I have on my hard drive (and I'm talking about complex ITs, like Skaven's "The Goblin Returns"), as long as I'm not doing anything else.
I'm assuming you're using floating point sample data for internal mixing, so that would exclude MMX optimization. But perhaps SSE could bring a performance boost, even if only by allowing concurrent MMX/x87 execution (freeing up the FPU registers).

Regarding the sample rate options... it'll be in the next update

Thank you! I really appreciate it!

Edit:
There's something I noticed in resonant filtering. There's a rather long filter sweep at the end of the 7th order of "Drifting Onwards", by Jeffrey Lim (the module that came with Impulse Tracker 2.14, as a resonant filter demonstration). I can't accurately describe it, but XMPlay doesn't sound like Impulse Tracker, as if the resonance became "uncontrolled" (too loud?) at very low cut-off frequencies.
Speaking of resonant filters, are they implemented using SIMD instructions?

I'll be repeating myself, but since my request for 3.2 was too late...
I only wish for a tunable FIR interpolator (similar to the Winamp module plug-in) and 88200 Hz sampling rate (or higher, as long as it's a multiple of 44100 ).

Also, IIRC, the sinc interpolator is only SSE2 optimized. Is there any reason not to optimize for MMX/SSE?

My wishlist? Just a tunable FIR interpolator (a la Winamp MOD plug-in) and 88200 Hz sampling rate...

If you're planning on doing anything serious (writing an audio CD, for example), I recommend you to write the file at the highest possible sampling rate (that is a multiple of your target sampling rate, for oversampling purposes) and sample resolution, with NO dithering/noise shaping.
Then, use a good audio editing software (I use Cool Edit Pro, now Adobe Audition) to do the post-processing (nyquist pre-filtering, downsampling do the target sampling rate, nyquist post-filtering, sample resolution conversion/dithering/noise shaping).

Granted, XMPlay does an impressive job on 8-bit sound at 44100Hz+ with dithering and noise shaping enabled (sounds quite 16-bit-ish, nice shaping curve, Ian! ), but if you're going for the best possible quality (or you're just a paranoid control-freak like me) you might as well bring on the heavy artillery.

Greetings!

On the Ogg Vorbis topic: Would it be too much trouble to implement the "vendor" and "nominal bitrate" fields when displaying the general stream/container information?

Thanks!

I've compared Modplug Player against XMPlay, as far as resampling quality goes, and I still find MPP to have the best resampling engine I've ever heard (I used a tough one, "Daisy Chain" by Rhesus Minus, to test the quality). Granted, XMPlay has a FAR more accurate module playback...
Are you planning on improving the resampler quality? Perhaps using a higher order filter (8-point, 16-point or even higher), or trying other interpolation algorythms (I particularily like the sound of Gaussian interpolation, but that's a matter of personal taste ). That's the only thing that keeps me from erasing MPP from HDD for good!

It's not a secret that some MODs actually sound better with
no interpolation. Chip tunes are a good example on this.
No interpolation sometimes also gives a more natural sound
and it is always good to keep the sound as "original" as
possible.

Actually, they sound "better" with interpolation (go study some signal theory if you have doubts ). But since the original AMIGA wouldn't do interpolated resampling, musicians eventually found ways to turn that handicap into an advantage, by making some weird sound effects...
Interpolated resampling always makes modules sound better, but sometimes not like the original composer intended!

Yes, I meant the final output , but after giving it some thought, I believe you should implement something more useful than dithering... People interested in mastering some MOD compilations on CD would just render at the highest possible quality (interpolated, 32-bit at 96KHz) and use some good post-processing software (Cool Edit Pro, anyone? ) for the final conversion stage.

Now for the suggestion: how about supporting 88.2, 176.4 and 192KHz sampling rates? As you most certainly are aware, multiples of the base frequency produce no aliasing on downsampling, which should be simplified to a trivial "discard every other sample" (excluding filtering).

OTOH, maybe I'm just paranoid... :laugh:

[Mathematically]

Yep, the most noticeable difference over linear interpolation is that snares and hi-hats are generally sharper.

Well, my reference is ModPlug's 8-point interpolation (cubic polynomial? I'm not sure about the algorithm...) Linear interpolation was never an option, for obvious reasons...

Not sure old MODs (with poxy sample quality) are going to sound "better". Better resampling means less artifacts, which, in the case of low quality samples, is not necessarily how they were meant to be heard

They'll always sound better... Mathematically better, that is!

A Blackman-Harris window does a better job of removing the ringing in this file's case... so I've changed it to that now

Hehehe! I was going to suggest that windowing function, since I believe it's the one that yields the best tradeoff between time/frequency resolution!

Oops! I've been itching to ask you these, and almost forgot! What's the accuracy of your calculations? Since you said you optimized the interpolation routines for 3DNow!, I presume you're using 32-bit floats. Two questions: first, is the mixing routine dithered? (Spreading the quantization errors should increase the dynamic range a bit...) Second: I'm currently using a P4-based machine (OMG... Did you know they botched down the standard non-SIMD FPU of this beast ON PURPOSE?! ), so that means "No low CPU usage for you!", at least for a while, right?

I heard the samples, and all I can say is WOW! Impressive results indeed! I never thought it could sound this good, especially with almost pure waveforms (that XM from WAVE is a very nice example ). The snare also seemed extremely clean with most (all?) of it's high frequencies, and absolutely no audible interpolation artifacts. The low-pass filter from the Vorbis encoder could have something to do with this, but I pretty much doubt it (and I do spot compression artifacts miles away, believe me ). Can't wait to give it a run on some really nasty modules (Daisy-Chain, by Rhesus Minus, comes to my mind... ).

PS: What windowing function are you using?

Cheers!

I didn't know about this method, so i googled around a bit and found a very nice site explaining the concept of bandlimited interpolation. Unfortunately(?) I also found this page revealing some problems with the same algorithm (mainly related to the Fourier transform). I don't know how those ringing artifacts translate to digital audio processing, but (at least graphically) they're quite ugly! But I assume your sinc interpolator is nicely band-limited... Cheers!

Well... Perhaps that could be (not too terribly ) useful for you, me, Ian and for the rest of the 10% of the XMPlay user base we're included in, but I believe the other 90% couldn't care less. Please don't get me wrong, I'm quite aware that convolutive DSP techniques yield the best sounding results, but I don't think a convolution plugin would be that useful for the average user... Call me nuts: when I go to a cathedral or a relatively "wet" concert hall, the first thing I think is "God, I wish I'd be alone with my laptop and a couple of microphones for a while!", but I don't see people running around capturing (or synthesizing, for that matter) impulse responses...

I believe new module loaders and higher (studio+) quality resampling/mixing routines should be top priority. I mean, don't we have enough PCM/bitstream players?! Do you REALLY need another bloated piece of (compiled) code?!
Also, for practical purposes, I don't see any reason why a player should have a skinning system. Eye candy? Come on, people! I'm sure you have something better to do than to sit all day changing skins! Load a playlist, minimise that thing and go do something useful! (Some coding, for example! )
Development focus must shift towards improving what made XMPlay famous in the first place... Let's make it THE BEST module player, not another average generic player. Besides, being a software developer, you (should! ) know how "featuritis" can be a deadly disease!