You are viewing comment page 2 of the post. View the post contents here.
Tip: Press Ctrl+F to search for a specific keyword or keyword combination.
We've noticed that you're using an AdBlocker
It's not just you, over 66% of our site's visitors are blocking the ads.
Please disable adblock for this website and refresh this page if you:
find the content useful
want us to create more useful content and software
want tech support through the comment section
The ads are placed so that there is minimal interference with page reading. There are no pop-up, pop-under or sticky ads.
Alternatively, you can support us by making a donation.
I do my listening with some Sennheiser HD650s and a decent headphone amplifier. After this tweak now listening to music under Windows sounds similar to how it sounds under Linux (Ubuntu, disabling Pulseaudio, using ALSA and VLC configured to output through ALSA, direct hardware, no resampling)
I wondered why it was so, why Windows sounded weird relative to Linux (believe me, the 650s resolve REALLY well, you can easily pick up those minute differences) and this is it, probably, I don't think the Linux driver does the same stuff behind the scenes.
I haven't tested with speakers (I have a 2.1 setup I sometimes use instead of the 650s) but I've always preferred Linux's output when using them, so I suppose I'll come to the same conclusion.
Thank you VERY much for this tweak.
It's good to know, thank you for the feedback. The Xonar drivers for Linux were built from scratch and were not done by C-Media, that would explain not having included such a "feature".
I see. Then this explains the difference. Thanks again for this. If you could add the tweak in the installer as a checkbox for a future Unixonar revision I'd be more than grateful.
I forgot to put it in my first post, I use a Xonar DX.
Hi Carve
I report and confirm that the sound quality improved, using video and music playback on youtube sound seems crystal-clear, the higher, mid, low frequencies that are more homogeneous at a high volume than you can keep it over 80% without distortion; Using the software Player K-Media Player Classic, the same result and even with PowerDVD 17
I did not test with video games yet.
Before Tweak, however, it was more difficult to make adjustments on the treable and bass and individual channels vol level of the Gigaworks set. The output seems to have improved for me.
My settings: Speakers: Creative Cambridge GigaWorks S750 (modified amp 850 watts rms) + Central speaker Polk 150 Watt, in Cmedia Console equalizer set of soft-rock presets and Dolby HT v4 enable on movie set, set 7.1 analog on (HDAV daughter Card 120 Db), Db gain setting for channel front side and rear 11 and 12 ch center and 9 sub ch drivers v1.81a R2
Hi,
I've got a Xonar STX with Beyerdynamic DT 880 600Ω.
Applying this tweak is amazing, sound is crystal clear no matter video games, movies or music
Thanks 🙂
How did you achieve these low dpc values in general ?
Im getting ~ 0.3ms on highest execution time running the same test that you did. (same driver, same card, same player, different soundfile)
You are referring to Windows 10 results?
It could be something with how your motherboard handles the hardware resources. You should take a look at The case of DPC latency: Part 4.
Yeah im on windows 10.
I already had hpet disabled and applied most of the tweaks from there, im on a Z77 board (Asus Sabertooth Z77) with an Intel I7 3770 maybe i can improve it with different chipset drivers / bios settings.
No way, cant get it lower than 0,25ms i guess i just have to live with that i even put the card in another slot.
Are you sure that your values are 100% correct as my values are 4 times higher than yours?
Highest DPC routine execution time (µs): 267,801471
My results 100% are correct. I got "Highest DPC routine execution time (µs): 70.489780 (STXII.sys...)". I have a DFI LP P35 motherboard. Could be something with your motherboard. Maybe you can check the results of your Xonar card in a different system.
Lol, I dont think anything today compares to DFi mobos. Maybe some extra ASrocks could get close. 😀 But doubt that.
I tend to agree, DFI made some great motherboards and are surely missed by the enthusiast community.
Implementing "disable SpeakerCompensation" completely cured a problem that was driving me crazy. Thanks!
The image is now clear, clean and flat. Whatever SpeakerCompensation was doing, it was profoundly bad and it affected the entire range, absolutely not just the twinkly tip-top.
All my material (some of which I've been listening to since the 1960s, so I'm pretty familiar with it) sounded like it had a vinyl sheet over it, especially with equalizer-flat. I struggled to trim it out but never succeeded, and no way could I run without the equalizer. The vocal ranges especially were distorted strangely and unpleasantly, and nothing subtle about it. I can now run with the equalizer disabled.
This is the minimal "correction" that made the card listenable, but still not right. I was just suppressing the offensive range.
30 60 120 250 500 1k 2k 4k 8k 16k
+3 -1 -2 -3 -4 -4 -4 -3 -1 +3
I listen to bird-sound and other natural sounds, recorded (high-res) by me and others more than music and I never game.
1.80a Driver, Essence STX II card, Windows 7x64, ASUS Z97-PRO motherboard, i7 4790 CPU (Devil's Canyon, nice chip), good cables.
I use Audacity for analytic manipulation but mostly Winamp for listening.
Swan M200MKII self-powered speakers (on my desk), (These are well broken-in and (amazingly) clear/transparent. I think the manufacturer took unusual care to match and trim the internal amplifier to the cones and domes, and there is no case resonance that I can hear. All to say, I am confident that when I am hearing something strange, it's in the signal. I also use Sennheiser HD800S headphones.)
Thanks
Asus Xonar D1 PCI, Windows 8 x64 build 9200. I checked ten times "SpeakerCompensation" tweak (enable/disable) to be convinced of available changes. By disabling it I have noticed significant improvements in sound detail and general depth. Tested some FLACs which I know how it sounds and again can tell that quality is better when "SpeakerCompensation" disabled. Tested with MPC-BE + WASAPI Exclusive. By the way, DPC Latency Checker works correctly only in Win7, in versions above it shows incorrect 500 or 1000us. Users with Windows 8/10 should use LatencyMon to get correct values. My DPC is near 96us. Big thanks for tweak.
Lol, thats pretty noticable difference.
I did clean install of 1.75 (W10) with low DCP and then applied patch. Running thru F2K via WASAPI, well.. difference is really striking. Xonar Essence STX Beyerdynamic DT880 (250 Ohm).
This tweak is pure gold, thats how I hoped this soundcard will sound from start.
"This tweak is pure gold, thats how I hoped this soundcard will sound from start."
exactly
good job perfect pure sound
STX II , Uni Xonar 11.5 v1.80b r3 , Audio-Technica R70x (470Ω)
Run Disable SpeakerCompensation
Get a better and more natural sound !
its worked on 1.81 version driver(1825)? and whether it works on a different driver, not unixonar?
For some reason this tweak doesn't work for me. I tried on the drivers: Unixonar 1.80, 1.81. also tried on the driver from danielkawakami and there’s why I can not hear anything either. I install it like this: turn off all the programs, run .exe, I see an icon in the tray for half a second and then it disappears. I reboot the PC and can not hear the difference
It does work with other drivers. I think the tweak applied correctly. You can check this by going to Xonar driver registry settings by following the guide from FAQ Q&A 5.1. There look if there is a SpeakerCompensation entry. If it’s there then the "disable SpeakerCompensation" tweak is applied.
Not everyone can hear the differences, it could be either because of one's hearing capabilities, speakers/headphones used , music format or music genre.
Speaking of the Xonar drivers from danielkawakami, those do not offer anything over UNi Xonar drivers, they are original drivers with the 5.1 delay fix from UNi Xonar drivers incorporated.
Strange. UNi Xonar 1.80a r3 tweak works well.
Thank you for this very interesting tweak. It does seem to improve the sound a bit for me, although I can't be 100% sure honestly. Too bad we can't know what it does exactly - sort of a mystery!
Xonar DG, W10 x64, 8.1.8.1823 driver (must be the 1823/1.80a r3 setup?), stereo analog, HK SoundSticks, quick FLAC/WASAPI audio tests. About the frequencies, I can hear 8-15 with both settings.
Hello! I found a serious problem when using "disable SpeakerCompensation tweak". I found this not immediately, and for a long time I could not understand what was happening. Here's the point. In all computer games, such as The Division, Quake Champions, GTA V and others, I found a serious drop in performance. Friezes, stammering, jumps fps from 130 to 2. For a long time I could not understand what was happening, why my games started to slow down when I had never seen such a thing before. I have a fairly powerful system i7-4790 / 128GB SSD / 2TB HDD7200 / 16GB DDR3 1600 / GTX1070 8GB / ASUS XONAR DG / Win10 64. My games have always worked stably. But from recent time they started to slow down. I first began to sin on the processor, then on the graphics card, then on the hard drive and the system. But nothing helped. And then I remembered that I used "disable SpeakerCompensation tweak". I turned off this setting and cheers! All the games stopped braking and frizz !. I tested this several times for myself and really. When I disable the SpeakerCompensation tweak, then my games drop in performance and start to slow down. Begin friezes and fall FPS. But when I return back. That all becomes good. Here and so.
Thank you for reporting this. Since applying this tweak I only played 1 or 2 games and I can't say I've noticed any drop in performance. I mostly played Dota2. My PC is Core2Duo E8400,Radeon HD7790, Xonar DX, Win7. I will have to do some detailed tests in more games to see if my system is also affected when SpeakerCompensation is disabled.
Please tell me which driver version you have installed and selected panel configuration, number of speakers/channels, analog or digital(S/PDIF) output, sample rate, the interval at which the framerate drops occurred, and if have noticed higher CPU usage during the those frame drops?
If it's possible for you to do some DPC latency tests with LatencyMon during game playing with and without the tweak. You'll have to play the same game and record the DPC metrics for a couple of minutes after which stop the monitoring, save the report and make a screenshot of Drivers tab(sorted by DPC count). Then you send those files to my e-mail.
No problems so far. Bioshock Infinite, Unreal Engine 4 Demo (Loft in London, Lucid Arch Dreams, Unreal Paris).
Thank you for the feedback. I'm thinking the performance problems may have something to do with audio settings used, OS or in conjunction with specific hardware.
If you can tell me which audio driver version you have installed and panel configuration, number of speakers/channels, analog or digital(S/PDIF) output, card sample rate, video card, CPU.
Core i5-2500k 4.4ghz, gtx770 oc. Driver 1.80a r3, stereo speakers via analog, card sample rate 44.1khz, tuned eq, all other effects disabled.
The GTX 1070 has been reported to cause DPC latency issues in the setups of multiple users on the official Nvidia forums. It seems likely altering sound driver factory settings might aggravate the problem.
I suggest disabling all sound on GPU. It causes major conflicts for me, if I let it be enabled (also have nVidia and Xonar soundcard).
You might be onto something. What video card do you have?
As Mescalamba suggests, you should disable any Nvidia audio device from Device Manager->"Sound, video and game controllers" and apply this tweak again. See if this helps and let us know.
I wanted to add that I had disabled the Nvidia audio device before and after the tweak. The nvidia device has no influence on this problem. (GTX 1080)
I second this problem. Solved by disabling the tweak. PUBG, BF1, Quakelive, and Quake Champions.
I had disabled the Nvidia audio device before and after the tweak. The nvidia device has no influence on this problem. (GTX 1080)
Please tell me which audio driver version you have installed, panel configuration, number of speakers/channels, analog or digital(S/PDIF) output, card sample rate, CPU, OS.
Ok, I've ran multiple game tests on my PC and there where no FPS difference, no FPS drops or freezes between having SpeakerCompensation disabled and enabled. On Windows 7 with UNi Xonar v1.80a I've tested Dota 2 and Mafia 2, on Windows 10 with UNi Xonar v1.80b I've tested Hitman 2016 (DX11 and DX12) and Prey 2017.
This means that the issue is caused either by some specific audio settings or in conjunction with specific hardware. Anyone having performance problems when SpeakerCompensation is disabled please tell me the following information which audio driver version you have installed, panel configuration, number of speakers/channels, analog or digital(S/PDIF) output, card sample rate, CPU, videocard, OS.
Knew my EQ settings were perfect, thanks for this! Whats the sound difference like? almost sounds more acoustic or clearer than without this tweak. Also noticed slightly more crackling at higher frequencies, maybe from sound being more clear or I need an upgrade from my Audio-Technica m50x.
Way more definition, its like there was some slight muddy veil before, now its gone. IMHO, clarity got most buff I would say. It changes EQ tiny bit, but for the better. I have specific EQ in F2K for my headphones, before it was like "alright" and now its "perfectly neutral" just like I wanted it (I know, its boring but I want sound like its mastered).
I can compare to few other soundcards and also heard some recordings with much better equip than I have, so its not hard to judge.
Which are the other cards that you've compared the sound to? I'm interested in knowing if other non-Xonar cards have this kind of frequencies alteration enabled by default.
Since the sound is clearer, it makes sense that crackling and other defects in the tracks are more noticeable. Thank you for the feedback.
After applying the tweak, as in another post i have to reapply each time after a boot with 1.72 *shrug*
Which post? How do you know that you need to reapply each time?