Hello all - new to the forums here...

Question regarding the mpc WinAmp plugin... What is the default bitrate if the dither option is not checked? I searched, but can't seem to locate that info. My hardware is 24bit and I use 32bit capable DX plugs in WA2 through the AdaptX DSP plug, and have a dither plug already in my chain...

Along the same lines, with the proliferation of 24/96 hardware, wouldn't it be a good to allow the option of 24 or 32 bit decoding?

TIA

EDIT - should have mentioned this relates to the mpc plugin...

Hmmm... I'm not really proficient at C yet, but after reviewing the source files it *appears* that the output of the plugin is hardcoded at 16 bit and the dither option simply specifies whether to truncate or dither the internal 32bit down to 16bit... I have queried the author by email for more information.

Anybody able to confirm or comment?

Just curious, it seems to me that 24bit is becoming the next standard and other applications like the MAD decoder and foobar offer 24 or 32 bit output...

Boy, I think I'm losing ground rather than gaining any understanding here

Just finished reading several long and detailed threads regarding the K value and Headroom settings for the plugin, and now I'm more confused...

Are the recommended settings based on the assumption of 16bit hardware? The reason I ask is that it seems that one of the help files for the Headroom slider indcates that if you have 24bit capable hardware you should use k-20+... Which is very confusing since the plugin *appears* to only offer 16bit data (see above posts). The sticky for recommended settings suggests a value of K-14 but really doesn't offer what that is based on. To be fair, I'm still learning about all this mpc (but I like what I hear!). The several threads regarding headroom and K values mention a headroom level of 83db, but isn't that based on 16bit? I thought 24 bit had a value of closer to 100db?

My soundcard is an EgoSys WaveTerminal 2496 and I do a lot of home studio audio work with Cubase etc. IMO it is a very transparent card and although I have not conducted any ABX tests with MPC, I can clearly hear the difference through WinAmp on MAD 24bit decoded mp3s (Lame 3.92 -altpreset -extreme) versus MAD 16bit.

IIRC, you're right, 16 bit output is non-optional!

UPDATE - The author was kind enough to respond to my email and confirm that the plugin is hardcoded for 16bit, although he comments that he has plans for additional features when time allows.

Still a bit confused on the Headroom issue, but I'll continue research.

I'm new to the mpc codec, but I already prefer the quality of the 16bit playback of mpc to the 24bit playback of mp3's encoded with Lame -alt-preset-extreme. Good work!

The headroom should be like this for all sound cards - first lets forget lossy as that muddies the water a little (yes there are some that argue a 24 or 32 bit output from a 16 but mp3 file sounds way better, I am not convinced...), lets go onto lossless.

If you have a 16 bit lossless file and play it 16 bits through a 24 bit sound card, the output should be identical if you had a DSP program that took the 16 bits and offset it to 24 bits. The sound card does that for you, otherwise if you played an 8bit file you would never hear it, so you are already driving the sound card as best as it can.

The only benefit of 24 bits is if you have an EQ on (again very marginal), the other benefit is if you have something that uses ReplayGain (to all those who have emailed me, I will get around to adding it right across the board) because you can scale up the volume when it is required and it prevents clipping.

Apart from that I see no good in taking 16 bit data (a mpc file) - offsetting to 32 bit then dithering down to 24 bits.

** These views presented are Spoons and not necessary that of the whole world **

I agree with that assessment. I think it may be in my case that perhaps the DX plugins I have in the chain (varies, but usually along the lines of MaxxBass, BBE enhancer, Waves Q8 EQ and Waves L1 IDR from my studio applications) seem to produce a more transparent sounding mix from my lossy files, particularly the more 'needy' ones. I will not proclaim myself as any type of expert, but I can usually AB the 24 mp3s played through my system fairly easily. Whether I would be able to do the same for the mpc files I cannot say, I only would be interested to try.

Thanks for the comments... As I said I'm new to this mpc and I'm just getting my feet wet and am open to all advice/suggestions.

Have you tried blind? That makes quite a difference. There's still the possibility that at 24 bit mode the output level of the card is different that at 16 bit mode, making easy to perceive a difference. But I think this would be quite strange.

MPC and MP3 don't store data as integers and decoding 32 bit vs 16 bit does make a numerical, yet not audible, difference.