I've been finding that overall OpenAL is working better for me than DS, but with the downside of the loop hiccup.
(and apologies if all this is covered already, I didn't see it anywhere)

After putting the hokum_turbine_int file through a filter to remove the high-pitch component (really starts to grate after a while), I noticed that the sound loop hiccup was less obvious, so had a closer look at some of the other sound files, and adjusted a few, and the looping glitch is much less noticeable now (though not absolutely cured).

I also changed the length of a couple of files where they were the same length and it looked like they might be played simultaneously - the idea being that this would avoid multiple loops ending (and glitching) at the same time though effectiveness of this was at best inconclusive, since the most noticeable glitch seems to be on the turbine_int loops, and in particular with the US helicopters (i.e. a much more continuous sound). And no noticeable glitches with the hind at all.

So I took one of the other sound files, mangled it quite a bit and got a new comanche_turbine_int sound that doesn't suffer nearly as much as the original. It doesn't sound much like the original either, being more like a muted blast-off rumble than a turbine, but since I've never been in a real helicopter I have no illusions as to knowing what it is supposed to sound like. On the other hand, it removes an annoyance and you can still just about hear changes in engine pitch. As yet I've only tested this particular one with the comanche, and it may or may not work so well with any of the others.

And one fix : the hokum_rotor_int sounded exactly like the hokum_turbine_int (I don't *think* I over-wrote it) so a low-pass filter on the rotor_ext to create a replacement one.

Of interest to anyone? Let me know and I can post/PM a link to the modified files.
[FWIW : done using sox and audacity]

Sure! Publish them!

OK - files available at http://www.flymouse.demon.co.uk/cohokum/

NB there are two zip files - one with pcms that can go in the 'audio' subdirectory, and one with wavs so you can have a chance to listen first. Several sound identical, because the original source files were the same, so it's not an error.
Also worth noting that what works here (headphones only) may well be frighteningly bad through speakers!

No warranty, take backups, look before you leap etc

And apologies for the delay - I would have put these up sooner but it's been years (literally) since I last played quake and I forgot to set a frag limit so we only stopped when the first person waved the white flag to run off to the Small Room.

Very nice sounds!

Glad you like them!

I dare say there may be room for improvement but there is a lot of trial and error involved - several attempts sounded fine on their own and even when testing in the audio software but once you get into the actual game and have everything else going on around you along with engine pitch changes it just sounded like I was driving a demented lawnmower. Which can be amusing, but not ultimately a useful result...

Obviously 'improvement' is a very subjective term but I have also been looking at possibly adjusting some of the identical turbine sounds to get slight but noticeable differences depending on what I am flying (except the Hind which is (a) already quite distinctive and (b) an excellent adventure in its own right), as even between machines of the same type there would surely be *some* difference in what the engines sound like.

Or I could just be digging a very deep hole for myself.

Many tanks!

I was already wondering if there was a solution to the deafening turbine sounds in the Comanche...

I have made further attempts at rotor_int and turbine_int sound adjustments and the hiccup seems mostly hidden on most of them though you still get it from time to time on the rotor slap sounds.

Two files, tagged with '20120206', at URL a few posts above - one with ready-to-use pcm and the other with mp3s of each pair to 'preview', and the traditional reminder about backups in case you don't like them.

Note that some of these are a bit quieter than before but I used them without changing the INI sound settings and it wasn't a total disaster so maybe only minor tweaks would be needed.

So - comments, suggestions, odd clicks, too much 'eee', not enough 'whuppity whuppity'...?

I'm glad that you're working on it.Always annoyed me too,Hiccups still is, but much much smaller.
The more it annoys,that overall the sound is perfect in EECH.
Each correction is welcome, look forward to more

I tried going a bit further with those last ones but found that the final stage (conversion back to 8-bit, I can't find any 'native' 8-bit sound editors) was adding in too much hiss so I have tried a different approach - creating a few experimental sounds from scratch - hiss still appears in the final stage but not so much as to be a problem.

The results - whilst being far from perfect and probably not to everyone's taste - are better than I thought they would be, and I have uploaded a few of them for anyone who want to see what they are like (zips tagged with '20120210' [Edit2: NB duff mp3 versions with silence-weirdness now fixed and re-uploaded). The filenames are a bit obscure and obviously have to be renamed to be used but I hope it is clear what they should be - they are all for '_int' sounds, just set the name for whatever you want to fly. NB rotor and slap must be matched, i.e. both should be e.g. 'trem24' as this is the beat rate.

FWIW Currently I am using:
rotor_brn_trem25g4.pcm as hokum_rotor_int.pcm
slap_wht_trem25cont90.pcm as hokum_slap_int.pcm
turbine_brn_trem212_bs.pcm as hokum_turbine_int.pcm
which seem almost good enough to me though I haven't gone through the whole range of mix-and-match yet.

The turbine ones could definitely be better, the slap is probably a little loud and there's an occasional small crackle but it seems better than the hiccup (I can only speak for a headphones-only setup) though this set is primarily experimental and therefore - in theory - easily fixable. That said, a more advanced sound system than mine could well reveal all sorts of hidden horrors!

On the plus side, I now no longer have to turn the volume down for just this one game. I did find that the voices (especially co-pilot) were now comparatively (too) loud, so I extracted them, put them through the mangler and poked them back into the sound.dat file, which seems miraculously unharmed. A bit of extra hiss as a by-product but it's all radio talk so I can probably forgive that...

Comments and/or suggestions gratefully received, especially if I'm now heading in entirely the wrong direction!

Latest attempts now uploaded to URL above, tagged with '20120225', raw pcm and mp3 'preview'.
Again, all very much in the 'works for me' category.

I ditched the attempt at 'created' sounds, as these mostly tended to end up as one of three types:
a) demented lawnmowers
b) metal scraping along a piano-wire
c) coconuts

So, back to adjustments of the 'proper' sounds, noting that 'adjustment' may not be equivalent to 'improvement'.
Any excessive hiss should now be much reduced - thanks to an old wave-edit program I dug up - which seems to work in 8-bit mode with little or no internal conversions.

The files contain adjusted rotor/slap/turbine sounds for:
ah1t, blackhawk, havoc, hind, hokum;
and adjusted rotor/turbine sounds for:
apache, comanche.
No further changes to the ch3/ch46e/ka29/ka50/kiowa sounds which are in the '20120206' files.

And I'm in two minds as to whether less noise in the Hind is necessarily 'better', though it's not a huge difference.

A couple of things I did notice (though some/all may be in the 'just me' category):
- the ah1, ka29, kiowa don't seem to use their 'turbine_int' sounds (but I did these anyway)
- the hind and blackhawk slaps were a bit quiet - a tight turn would sound like you had gone into stealth-mode (now made noisier)

On a side note in relation to speech, I had another go at reducing the volume with a somewhat brutal divide-by-2 of the original data which solved the hiss problem... by replacing it with distortion! Though (perhaps oddly) it seems to 'fit' in context - crackly radio, and the more people that are speaking at once, the worse it gets - so I'll defend that as being 'added realism'...

[ sorry, I seem entirely incapable of making short posts like 'new files, go get them' ]

I've now uploaded an 'all so far' zipfile (dw_all_pcm_20120312) which contains my preferred versions of the ones I adjusted.
A couple of them are earlier versions (or recreated in the same way) which I felt were better than my subsequent efforts.

As these mostly work well for me, I'm not at this point looking at doing much more to them, unless anyone has specific comments and/or requests.

I stopped trying to adjust the volume of the speech samples since even after I had got around the whole unsigned-but-actually-signed PCM data thing and got some nice clean volume-reduced versions, it all went horribly wrong once they were mixed in with everything else (even reversing the polarity didn't work) and there's only so much distortion you can take, even with my made-up excuse of 'realism'.

So instead I turned (almost) everything else up with:
canopy_sounds_amp=450,90,375;15,15,100;120,80,100;10,10,100;240,10,100;20,20,100;180,20,100;180,10,100; # canopy sounds amplifier controller
(and turned main volume down, other games up)

Where there's a will, there's an Or.

Did somebody try it at home?

Most of the downloads were for the first batch, so either not many were impressed with the first attempt or their problems were completely fixed and they saw little need to download the final 'all_pcm' set.

I know which one I *want* to believe