![]() ![]() |
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
71.110.241.250
Finally did some level-matched A-B comparison between Foobar older version 0.8.3 with Otachan ASIO plugin and new 0.9.45 with supplied ASIO plugin.
There's no way to use the Otachan ASIO with new Foobar. Song files and computer settings were otherwise identical from same computer/audio setup.
Shockingly, the 0.8.3 with Otachan ASIO plugin sounds significantly better than newer version with built-in ASIO plugin. The effect is not really subtle, with Otachan having MUCH more resolution, detail, "pop," and definition in the upper-midrange to midrange area. New Foobar sounds pleasant and very forgiving in this area, but there IS a new veil there also. It's really too bad...
Follow Ups:
Thanks to Steve, I was able to compare Otachan 51 to 47a in Foobar 0.8.3. All my comments below are based on 47a.
I still don't know what setting during compiling of these ASIO plugins that affect the final sound, but yes 51 sounds different from 47a everything else being same.
51 sounds a wee bit more "exact" with a slight bit more edge-definition and leading edge transients, especially in treble/upper-mids. These are not night-and-day differences, mind you, but they're there.
47a seems ever so slightly softens the leading-edge and emphasize that "glow from within" quality of midrange. Along with what my system does, I think I prefer 47a, but others may prefer 51, especially depending on what kind of music you listen to.
ASIO is just a protocol for sending data from one place (e.g. Foobar2000) to another (e.g. an audio driver).
If you are hearing differences between two plug-ins that support ASIO output, one of the plug-ins is probably doing something other than just repackaging the original audio data. Given how many iterations of Otachan's plug-in there have been, and how dependent they are on processor features (something that is important for plug-ins that are manipulating a lot of data very quickly) I would not be surprised if there is a bit of DSP magic going on in there.
It may not necessarily be the code itself but the options selected in the C++ compiler for processor and SSE instruction set compatibility.
Sadly it all makes a difference.
I went digging a bit to find out exactly what is so magic about this Otachan ASIO plug-in and found some interesting information. Aside from a quote from Otachan that there should be no difference in sound between the various versions, and documentation that says all it is doing is passing the raw data to the ASIO driver, I found the original changelist from Otachan himself.Claims about differences between various versions can now officially be filed under the placebo section of the internet, as far as I'm concerned:
Use this:
http://babelfish.altavista.com/babelfish/trTo Translate this:
http://otachan.com/foo_output_asio(exe).htmlfrom Japanese to English.
Ive read these claims from Otachan myself. I hear differences in every version of ASIO, so I dont believe it. I dont use it anymore now, so its a non-issue for me. No ASIO is better than the best ASIO I have heard IME.
Hi Scrith.
I see you received my response email.
Well, I got surprised when Otachan said me last year that there shouldn't be differences in quality between builds, but the theory always differs from practice.
This is what I have been saying and recommending for years. 0.8.3 is the only player besides Jriver to use on a PC with USB interface. The SRC upsampler for 0.8.3 is also better.
Steve N.
Dear Jon L,
I would like to try this.
I am using FOOBAR 9.5, but I doubtr there is much difference between this and the 9.45.
I am using the HAGTECH USB between the computer and the DAC. Also, 1 am using the USB-ASIO.com driver.
What components do you recommend?
Thanks for the help,
Rick McInnis
Rick
What did you do to get USB ASIO to work? I am using a HagUSB and .95, but I can't get it to work with USB ASIO.
db
I am NOT an expert on this stuff!
I built my computer as specified by cics's paper (available here, seach for it).
I had never "built" a computer before (it was so simple using the word "build" seems to aggrandize the effort).
It did seem to require "something" other than simple installation but I have no idea what it was. I kept installing FOOBAR and "it" and uninstalling and eventually it all worked. It could have been that I had unplugged the USB cable from the computer when I was having trouble.
FOOBAR will not play with the cable unattached. It just blinks and sits there!
I am sorry I cannot give you good advice.
I have read somewhere that 0.95beta has the best SQ and to not use 0.945.
What are your thoughts compared to Foobar 0.8.3 with Otachan ASIO?
I believe I'm using the 47 Otachan ASIO with foobar 0.8.3.I have only tried the latest Winamp with the stock ASIO. Can you point me to where the Winamp Otachan ASIO is?
Winamp does sound promising, but something funny is going on b/c like 15% volume on Winamp correlates to almost max volume on Foobar. I'm beginning to think max volume on Winamp is actually way more than 0 dB.
Winamp is also presenting me with problems I can't seem to solve. For some reason, my install won't let me play WMA Lossless files. I know it "should." I also have issues with its poor playlist/library management that takes too long to load songs and refuses to display all the ID3 info I want, mainly "album name." No one seems able to tell me how to fix these issues, even at the Winamp forum...
Jon - you should try ASIO 51 for Foobar 0.8.3. 47a is a bit soft-sounding. If you dont have it, I will email it to you.
BTW, all of these pale in comparison to the "driverless" firmware that is available now, like that used on the Benchmark DAC-1 USB.
Steve N.
the way data is supplied to its input - and that, to my knowledge, what all these ASIO and KS techniques are used for?
In other words, how does DAC's firmware eliminate the need for ASIO plugin for media player?
The firmware is for the USB interface. It eliminates the need for a custom driver and eliminates the damage that Kmixer can do. 16/44.1 data is converted to 24/44.1. 24/96 data is not changed. It uses the low-level Windows kernel, so no ASIO is needed. Very low latency so no pops or ticks. You can actually do work on the PC while you play music.
so no need to worry about it.
It works just fine with EMU drivers, but somebody here complained that it gives +6dB with some other sound cards.
Look under Winamp5 (link below) - I was mistaken, latest .exe version is actually 0.70.
Strangily, 0.70 exe would not work in Winamp 5.5, but I tried Otachan 0.67 dll, which works! Now the Winamp volume slider is disabled, and the volume level is similar to foobar at a given output.
Sonically, I still prefer Foobar 0.8.3 wit Otachan 0.47 dll over Winamp with 0.67 dll. Foobar still has more detail resolution and focus. In fact, Winamp/0.67 dll sounds more similar to foobar 0.9.45 with stock dll.
I hope Otachan ASIO is developed for foobar 0.9.45 so we can see if that's all the difference or not.
BTW, my problem with Winamp not playing WMA losslss files is solved with Otachan 0.67 dll asio. I don't know why, but that's how PC Audio goes, I guess...
all 3 options that are packed in zip file (I believe they are named something like ESS1, ESS2)?
It works for me with EMU 1212m and EMU 1616.
Yeah, I tried all that came in the Zip, there was one called "normal" and one called SSE2. Both .exe's dont' work. Maybe my Lynx 2B card doesn't like .exe files or something..
The problem is that Foobar and Winamp have different fullscale volumes, so when the volume controls are bypassed with an ASIO output plugin, the levels are different.
Foobar is 0dbFS(100%) at full slider - so bypassing the volume control is a non-issue. Winamp is 0dbFS(100%) at half slider, and +6dbFS(200%) at full slider, so bypassing the level control gives a net +6db gain. (Winamps volume seems to be modelled after the windows volume control, which is also designed to be full scale at half slider)
People that prefer Winamp+Otachan over Foobar+Otachan might be guilty of "picking the louder one".
Carcass said earlier he did not get this phenomenon using his particular audio interface, which is possible, but he did not reveal a detailed test method. I get this phenomenon for both interfaces I currently have: M-Audio Revolution 7.1 and Creative X-Fi Xtreme Music. And now, others are getting the same result, which does not surprise me.
Christian Budde has written an ASIO ouput plugin for Winamp that retains volume control functionality. This plugin is useful for experiments with Winamps volume control and ASIO. It also supports my theory about Winamps full scale level of +6db, since I get 0db gain with Christians plugin when set to FIFTY PERCENT SLIDER in Winamp, and the expected 6dB(100% gain) = 200% volume with full slider.
For the record, kernel streaming does the same thing. Now I know why Winamp was causing clipping on my Behringer using Kernel Streaming via a SPDIF output. +6db is why.
What I am saying is that Foobar+Otachan should give the same level as Winamp+Budde if Budde is set to 50% slider (aka 0dbFS aka 100% volume.) which also proves the theory that Winamps volume is 100% at half slider.
Non-believers can non-believe, but my experiment is valid, and gives the same results for two different interfaces, so I would need to see some really compelling evidence before I would be inclined to doubt my findings.
Cheers,
Presto
to the same track, switching between Foobar 0.9.x with default (not Otachan) ASIO and Winamp 5.35 with Otachan ASIO. The loudness was the same, to my ears. BTW, I'm pretty sure (not at home now - can't check) that volume controls became disabled in both Foobar and Winamp.I could best describe the difference as Winamp sounding "smoother" - less sibilance on vocals, softer - but not less extended - highs. I don't think this kind of preference can be attributed to difference in loudness.
One of the tracks used was "Embraced By Darkness" by Saturnus, that has
high amount of continuous high-frequency content (cymbals) starting at the beginning of the track.I still think that all these loudness anomalies (or lack thereof) are results of interaction with particular pieces of software, in my case EMU drivers.
Hey Carcass93:
When you said there were no level disparities, I thought you did an actual level check.
All you need is a test tone that is -6 or -12dbfs. 0db is no good because if Winamp IS adding 6db gain, you wont see this on a VU meter unless the range of the VU meter goes over 100% (over full scale). When you play the test tone from EITHER foobar or winamp you should get the same "VU levels" for either situation.
I was getting different levels with Winamp (+6db) which is what had me quite concerned.
Cheers,
Presto
Do you remember which ASIO build you are using for Foobar 0.8? The "best build" on Otachan's page is 0.51.2. Not my favorite...
On June and July I improved the quality but I didn't update the web cause I'm preparing something really nice.
The other day I listened to the 0.51.2 "best" build and it was garbage (compressed, distorted, ...) compared to some builds I´ve been testing these months.
In the meantime I've added the 0.51.7 standard build cause I feel shy being the 0.51.2 the best build. If you want the experimental build I've been using from July, ask for it by mail and I will send it to you.
The 54.1.and 54.2 .exe versions seem to be louder, possibly more dynamic than the "best" 51.2. At least to my ears.Version 47 is probably the most balanced with best timing. My "best"
Is there any way to make those run under Vista ?
I used to like 47a, but that was before I eliminated all audible jitter in my hardware. Now it sounds a but muffled to me. 51 is better IMO. No ASIO at all is even better, but you must have specific hardware to support this.
Not having any instruments to measure jitter I can only speak in general terms. In my search for jitter-reduced playback I strive for:
1. Best dynamic range
2. Best coherence, particularly during low level playback where you can still hear EVERYTHING
3. Best timing, I can almost feel musician's intentions.
4. Overall "very well coupled" sound across the spectrum, no softness, no hardness, superb bass integration and just a feel of getting everything from the material.I have heard the above only few times but it stuck in my head as the holy grail in digital. Surprisingly the hardware was not very expensive.
Can you describe jitter-free sound?
The ultimate in clarity and pinpoint imaging. Very dynamic and 3-D. Extended without being harsh.
Once the jitter is inaudible, then all you hear is the HF sibilance from your preamp and amp etc.. With the right preamp, this can be minimized as well. Then you are on the track to a really live sound, just like being in the studio with the performers.
Steve N.
Steve:
Quote: "Once the jitter is inaudible, then all you hear is the HF sibilance from your preamp and amp etc.. With the right preamp, this can be minimized as well. Then you are on the track to a really live sound, just like being in the studio with the performers."
Using the ASIO dll Carlos sent me (thanks Carlos) and Foobar 0.8.3 (or j river for that matter), behind quiet passages like the opening quiet bass solo of First Dance on William Parker's William Dances, I hear a slight hiss, something like tape hiss of old days. Is that what you are talking about?
db
No, that is probably tape hiss. I have a lot of older recordings with this.
What I am talking about is sibilance that rides on top of the music, masking some of the HF detail. This is produced by non-optimum designs in preamps and amps, particularly when op-amps are used. It has the effect of compression and noise addition. It cannot be heard unless music is playing. Its a type of distortion. I'll be demonstrating a new preamp technology at THE Show during CES that eliminates most of this. It's an eye-opener.
Were you able to run his 1.43 versions? What machine do you have?
My both computers (P4 1.9G and amd64 2.0G) froze and I had to reboot.
Analog tape hiss (without other side effects) is in my book the thing to strive for in digital playback.
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: