[Patrickhat2001]

I'd been frustrated by the sound of my rig as of late. Things just didn't sound right the detail wasn't there and the tonal character of things was just wrong--everything sounded stuffy and digitized (for a lack of a better word). I couldn't figure out what was wrong but I had a hunce it must have to do with my source. Having just got the EMU 1212 two weeks ago I was wondering how I could have been so impressed by this card only a short time ago because now my rig sounded like crud. Then I realized something: for some reason "use album gain" had been selected for replaygain within Foobar (which I normally have set on "disabled"). I immediately disabled replaygain and was greated by the incredibly detailed and transparent sound that I remembered getting out of the 1212 when it first arrived. So the lesson here is don't turn on replaygain unless you absolutely have to, it only decreases the sound quality (by quite a bit actually, my rig just jumped from blah to totally freakin' awesome again!) and make sure you don't accidentially turn it on either.

Has anyone else found replaygain to decrease sound quality? I don't recall such a thing being mentioned before.

 

[NightRaven]

Really. It is my understanding that replaygain is just straight digital attenuation.

 

[phamducduc]

Strange, i use replaygain for track and the sound is still sweet. Placebo effect maybe? Maybe I'll test out your hypothesis in the later future.

 

[Publius]

Patrick, extraordinary claims requite extraordinary proof, and this is so extraordinary to almost sound like a troll. Replaygain should be working entirely in the double-precision domain, and as long as that's working properly, there should be no quality loss. I've used RG with no problem for quite a while and I detect no quality difference when using it.

Some quick checks to make sure it's not something else.

1. Dithering should be OFF
2. ReplayGain preamp should be zeroed out
3. Output data format should be 32 bit
4. Remove all DSP effects, except maybe resamplers
5. Use WaveOut (if only to verify that it's not ASIO's or DirectSound's fault)

Unfortunately it's a little hard to put the onus on RG through ABXing, since foobar uses replaygain for level matching in foo_abx. One thing I'd suggest is to save both the original file and the RG'd file out as wavs, open them in editors, scale the RG WAV up to the original amplitude and compute the difference of the two files.

 

[Patrickhat2001]

Yikes! I don't mean to come off as anything resembling a troll. I'm just telling it as I hear it. I don't even pretend to know how half of these things work. I didn't realize I was making a grand statement!

I also don't have any files that are specifically encoded to use replaygain I just had it turned on by accident in Foobar. I always want all of my music to be at it's intended volume setting.

I'll do some comparisions in strait waveout to see if I still hear the same differences. But I really should be going to bed now, oh well....

 

[Patrickhat2001]

Alright, I just spent twenty minutes comparing wavs, MP3s and songs played directly off an audio CD and the differences are still there. I must say, though, that when using regular ol' waveout the differences between replaygain switched on or off were difficult to hear at times (because of the inherent lose in clarity when running the sound through the Window's kmixer) although they're still there (at least, of course, to my ears). All of my settings are identical to what Publius suggested, except, of course, I usually use kernal streaming or ASIO instead of waveout/directsound (and I disable resampling).

Again, this is just what I'm hearing, regardless if there should be a difference or not.

 

[averydonovan]

I don't notice any quality loss when using replaygain. And replaygain only affects the playback of files with replaygain info, it shouldn't affect any files without replaygain info at all, so I am baffled as to how you can possibly be hearing a difference in the output. I replaygain all my songs and albums so I can not have to constantly mess with the volume control.

 

[Kuja]

Head-fi needs a little injection of the below methinks

Quote:

8. All members that put forth a statement concerning subjective sound quality, must -- to the best of their ability -- provide objective support for their claims. Acceptable means of support are double blind listening tests (ABX or ABC/HR) demonstrating that the member can discern a difference perceptually, together with a test sample to allow others to reproduce their findings. Graphs, non-blind listening tests, waveform difference comparisons, and so on, are not acceptable means of providing support.

Then again, it strikes me that if Headfi adopted similar rules, there would be very little being discussed

 

[Publius]

Ah, sorry for biting your head off like that. I don't doubt that you do perceive this, but this needs to be analyzed very carefully for all sorts of good reasons. It may indicate a very serious problem with foobar, or it could convince a lot of people to run off and do things they really shouldn't need to be doing. In any case, thank you very much for continuing to look into this.

In theory, the only nonreversible effect replaygain might have on playback is increased quantization noise, which could be masked through dithering (even though you're not really suppsed to use it with 24 bit cards). I'd check to see if the quality improves if dithering is enabled.

Also if you haven't scanned anything through replaygain, then by definition, replaygain is not in your processing chain. What files are replaygain exactly affecting?

Finally, which version of foobar are you using?

 

[evil-zen]

It may just be the sudden increase in volume that makes music clearer and detailed.

 

[Skarecrow]

Clipping out the yin-yang doesn't make my music sound better.

 

[Patrickhat2001]

I'll visit hydrogenaudio.org soon and see what I can do about setting up some ABX testing for myself today, if I get a chance. I'm not really a believer in the necessity of ABX testing nor do I believe that it is the best way to do comparisions (I believe the fear of a placebo effect is really overblown--if you hear differences that are clear to you for a good period of time, like say a week, I think it's safe to say that those differences really are there regardless of how you biases may come into play).

 

[ted_b]

Patrick,
I agree. I don't like replaygain either. I think it hurts the dynamics and overall sounds processed. I mostly listen to albums (flac files are albums in my collection, I ripped them as "create image and cue file") so rg'ing isn't as much of a temptation anyway.

My gosh, this thread reminds me of the "burn her, she's a witch" scene from Monty Python and the Holy Grail. I'm sure Patrick will give us more details, but to start out and call him a troll, and that he must be doing something very wrong is quite harsh, especially for a forum that began as a subjective comparison of headphones. There are very few empirical experiments, double blinds and Nobel prize winning thesis's here. Most of the give and take on this forum is derived from a "just sounds better", then responding with "why, what's your system" then replying " here's what I'm doing" kind of thing. It works, usually. If the culture is now more scientific, then I'd suggest we just join hydrogenaudio.org and call it a day.

 

[Riordan]

Quote:

Originally Posted by Patrickhat2001 Yikes! I don't mean to come off as anything resembling a troll.

you didn't. not the tiniest little bit.

Quote:

I didn't realize I was making a grand statement!

again you didn't.

publius, kuja:
calling someone a troll for giving his experience and asking questions is rather rude. citing other sites' rules that (for good reason!) have no relevance on head-fi is not helping either.

this is NOT a scientific community, nobody makes scientific claims, nobody needs to prove or disprove anything.

 

[breez]

Patrickhat2001:

Let's try this. Right click on an item on the playlist, select Convert -> Run conversion and output to Wave (PCM, fixed-point). Then do it again with Replaygain enabled in the Diskwriter settings (accessible through preferences or the right click menu Convert) and do a bit compare between the files. Assuming your files do not have the replaygain info/tags, files should be identical, but anything could happen.