I mistyped it is actually a dual AK4495 DAC and for the price I paid for all the components it sounds great. The USB input is why I was happy about the USB2 driver because before I had to turn off driver signing to install it. Here are some pictures basically the same chassis as the LKS DAC.
I plan to get one of those remote volume boards I mentioned before from Taobao that use the New Japan resistive ladder volume chips. If that doesn't work well I have a relay controlled resistor board I can use instead.
It does different to the LKS ES9018 that I built, but that cost quite a lot more and I'm not sure I could actually pick which was which in a blind test.
I'll try out some of the OS tweaks you mentioned in wesayso's thread to see what happens, I've never heard a noticeable difference in the past. Are you able to measure any difference in the impulse response?
Attachments
I tried to use the USB2 native driver with the 18i20 but I couldn't find an easy way to get it to work. The Focusrite USB drivers are tied into the MixControl software so I would probably have to uninstall that as well to get the native driver to kick in. Uninstalling the driver did not work as it just installed itself right back when I restarted
Not going to spend anymore time with that for now as I need to get the tonal balance right which is proving to be tricky. I have implemented Mid Side processing in FabFilter as per wesyaso's settings in PLPar EQ. Hard to say exactly what the effect is but it is better on everything I have listened to so I think that will be part of the long term setup.
Attachments
Got same experience as you that UMC204HD wont run without its own software/controlpanel and there is a unknown firmware thing in device manager that is not covered by the new MS UAC2 driver so it cant configure it right, so maybe we can conclude that those pro UAC2 devices is not as standard as it was coincidence with ASUS U7 that was covered by the new inbox MS UAC2 driver, of course its okay behavour but wonder why about standards why those two pro UAC2 devices then run without any software installation on a Mac OS.
About get a dedicated box with JRiver to perform sound to die for can share from lately experience that with test platform I7-6700K/Z170/16Gb as soon as hyper threading is disabled in bios and stay away from installing graphics drivers package but only use MS build in VGA driver i get stunning sound performance, so if your box is not used for any graphics presentation other than a GUI to manage box settings you could try the two tweaks, about disable hyper threading it should not mean anything for JRiver because it looks via MS task manager that JRiver always use only one core and for your case with I3 CPU there should be 2 real cores back.
About get a dedicated box with JRiver to perform sound to die for can share from lately experience that with test platform I7-6700K/Z170/16Gb as soon as hyper threading is disabled in bios and stay away from installing graphics drivers package but only use MS build in VGA driver i get stunning sound performance, so if your box is not used for any graphics presentation other than a GUI to manage box settings you could try the two tweaks, about disable hyper threading it should not mean anything for JRiver because it looks via MS task manager that JRiver always use only one core and for your case with I3 CPU there should be 2 real cores back.
Many pro-interface companies tie their mix software to the Asio drivers. Focusrite does it, Presonus does it (although they had so much problems with the 1818vsl that they finally offered a version with Asio only, nothing else added.... but I wouldn't recommend Presonus as their customer service is... now where is that poo emoji?)
Like Byrtt mentioned, another way is to go with Mac OSX. The native Core Audio drivers work great and are very clean. No add-on Asio drivers, just plug and play. Most companies just use Apple's own drivers.
You may be able to install Mac OSX onto your little box, if you feel like it.
Also, consider that Focusrite is pretty much a general consumer company, with lower quality parts than more expensive units from other companies. Same was with my Presonus, which was very sibilant, and a $100 dedicated DAC produced a much better sound that that big audio-interface.
Like Byrtt mentioned, another way is to go with Mac OSX. The native Core Audio drivers work great and are very clean. No add-on Asio drivers, just plug and play. Most companies just use Apple's own drivers.
You may be able to install Mac OSX onto your little box, if you feel like it.
Also, consider that Focusrite is pretty much a general consumer company, with lower quality parts than more expensive units from other companies. Same was with my Presonus, which was very sibilant, and a $100 dedicated DAC produced a much better sound that that big audio-interface.
I will test disabling Hyper Threading in the BIOS I didn't bother to install the Intel graphics drivers this time anyway. I spoke too soon about the wifi the connection has become unstable so I bought a power line adapter to use a wired connection instead.
I don't think it is luck with your U7 just that ASUS made it USB2 compliant. Most cards will work but some don't as the manufacturer didn't make the hardware compliant, usually because they have some extra feature. Why stick to a standard that works for every system when you can roll your own
The mix software is tied into the ASIO drivers on both Windows and Mac and it doesn't work that well without the software as there isn't a hardware switch for everything. Weirdly it can be used as hardware only without a computer attached but the software is still needed to set it to that mode
Perceval, I have no experience of Presonus interfaces but the Focusrite 18i20 is actually very good. The headphone amp is the only part that is sub par everything else is good. It won't remain as my listening DAC for the long term but it is convenient for now to have the input and output on the same device. I would have to spend a lot more money to get a better device, Focusrite did a pretty decent job with this one. Focusrite produce/d some of the best high end analogue processing long before they made audio interfaces so they should know what they are doing.
I have had Hackintosh systems before, I could have done the same with the NUC but Windows 10 is working well, I don't see any benefit in the MacOs for this application.
I don't think it is luck with your U7 just that ASUS made it USB2 compliant. Most cards will work but some don't as the manufacturer didn't make the hardware compliant, usually because they have some extra feature. Why stick to a standard that works for every system when you can roll your own
The mix software is tied into the ASIO drivers on both Windows and Mac and it doesn't work that well without the software as there isn't a hardware switch for everything. Weirdly it can be used as hardware only without a computer attached but the software is still needed to set it to that mode
Perceval, I have no experience of Presonus interfaces but the Focusrite 18i20 is actually very good. The headphone amp is the only part that is sub par everything else is good. It won't remain as my listening DAC for the long term but it is convenient for now to have the input and output on the same device. I would have to spend a lot more money to get a better device, Focusrite did a pretty decent job with this one. Focusrite produce/d some of the best high end analogue processing long before they made audio interfaces so they should know what they are doing.
I have had Hackintosh systems before, I could have done the same with the NUC but Windows 10 is working well, I don't see any benefit in the MacOs for this application.
I thought I would post an update on where I have got to with the measuring and DRC tweaking and what that is sounding like to me.
I do prefer the sound with DRC correction. It has taken quite a while to find a good tonal balance and the right amount of phase correction. It is not easy even with the head start I got from wesayso's experience. Sometimes the changes are subtle but when you start to go back after listening to something good for a while it becomes more obvious that it is an improvement.
I am now convinced that these arrays are sounding better as an overall package than anything else I have heard. There are still some areas and some tracks where they are not as good as I have heard but the differences are quite small. On the rock tracks that used to sound bad on LX521 there is no comparison the arrays are significantly better. Whether that is due to changing the tonal balance or something innately in the speaker is hard to say as I haven't tried the same treatment on the LX521. I might do in the future but with the sound I am getting now I'm not in a hurry to try.
Mid /Side EQ has worked well for me I don't hear any tracks where it doesn't make things sound better, it does need to be linear phase EQ though to have the desired effect.
I tried a correction based on single position measurements and an average of ear positions. That did not work well and sounded very "processed" I put that down to the fact that my room is untreated and leaving more of the room in the measurement forces DRC to try and fix it which does not work out well. If I was to use a single position measurement I would have to dial back the correction and make it much weaker.
Two things made me realise that I was close to where I want to be. I have always liked the tonal balance of my headphones and enjoy listening to almost any music on them without it sounding bad. My headphones started to sound not quite so good which I put down to having heard something better from the arrays.
The second was when listening to a drum recording from a Chesky test CD. The kit has been recorded by a stereo mic in a studio room at 6 feet. I truly could believe that I was in that studio listening to that kit, the acoustic space and sound was totally convincing.
This is the Left and Right channels averaged after correction. This is the average measurement across the listening position used as a base so not an exact listening position measurement. I think it is a good representation of what I am aiming for though so take it for what it is.
I do prefer the sound with DRC correction. It has taken quite a while to find a good tonal balance and the right amount of phase correction. It is not easy even with the head start I got from wesayso's experience. Sometimes the changes are subtle but when you start to go back after listening to something good for a while it becomes more obvious that it is an improvement.
I am now convinced that these arrays are sounding better as an overall package than anything else I have heard. There are still some areas and some tracks where they are not as good as I have heard but the differences are quite small. On the rock tracks that used to sound bad on LX521 there is no comparison the arrays are significantly better. Whether that is due to changing the tonal balance or something innately in the speaker is hard to say as I haven't tried the same treatment on the LX521. I might do in the future but with the sound I am getting now I'm not in a hurry to try.
Mid /Side EQ has worked well for me I don't hear any tracks where it doesn't make things sound better, it does need to be linear phase EQ though to have the desired effect.
I tried a correction based on single position measurements and an average of ear positions. That did not work well and sounded very "processed" I put that down to the fact that my room is untreated and leaving more of the room in the measurement forces DRC to try and fix it which does not work out well. If I was to use a single position measurement I would have to dial back the correction and make it much weaker.
Two things made me realise that I was close to where I want to be. I have always liked the tonal balance of my headphones and enjoy listening to almost any music on them without it sounding bad. My headphones started to sound not quite so good which I put down to having heard something better from the arrays.
The second was when listening to a drum recording from a Chesky test CD. The kit has been recorded by a stereo mic in a studio room at 6 feet. I truly could believe that I was in that studio listening to that kit, the acoustic space and sound was totally convincing.
This is the Left and Right channels averaged after correction. This is the average measurement across the listening position used as a base so not an exact listening position measurement. I think it is a good representation of what I am aiming for though so take it for what it is.
Attachments
Keep up the good work! Listen, listen some more and let that soak in.
If you like what you have: make a good backup!
Then on to little controllable investigations to see what you can improve on.
To get ahead, little steps will get you there. Take the time to live with it to know if it's a move in the right direction.
Above all: enjoy and have fun with it! Try it with a movie sometime in the near future..
If you like what you have: make a good backup!
Then on to little controllable investigations to see what you can improve on.
To get ahead, little steps will get you there. Take the time to live with it to know if it's a move in the right direction.
Above all: enjoy and have fun with it! Try it with a movie sometime in the near future..
Thanks Byrtt, I don't think I can be bothered to mess with APL_TDA at the moment and the next Beta of REW should have a function that will make the graphs look more similar.
For now this is the Spectrogram of the Left and Right Averaged, there are some issues but overall for no treatment I am happy. You can see the peak energy curve is a little wiggly below 200Hz, but it is a whole lot better looking than without correction.
Attachments
Yep, I think I have seen a similar curve before...
Thanks for sharing. fluid.
And enjoy the listening!
There is a reason it is similar. I tested with a Flat target in DRC and then used PEQ by ear in Jriver until I got something that I was happy with. Guess what it looked remarkably like the 1dB/Octave slope.
This is the Left Right Average compared to my preferred target
This is the preferred target against the 1dB fall off slope. I like at bit more in the 200Hz range, a bit less 500 to 800Hz and a slight knee at 2-3K, not very different though and I listened to the 1dB curve for a long time because it was hard to find fault with.
Attachments
Keep up the good work! Listen, listen some more and let that soak in.
If you like what you have: make a good backup!
Then on to little controllable investigations to see what you can improve on.
To get ahead, little steps will get you there. Take the time to live with it to know if it's a move in the right direction.
Above all: enjoy and have fun with it! Try it with a movie sometime in the near future..
Ha Ha, I'll send you an email with how I did it so I will have you as a redundant off site backup for the process
There are a few more things to try for curiosities sake but I am tending to get different or worse results rather than better at this point.
I still have the First One Amp to set up and test and another DAC to try. When I have found my favourite combination of those the other tweaks might come out. I still need to read through all your emails again....
The last phase tweak got me back some of the bass impact without it being wild but I do feel that is an area that I might want to work on as the correction could well be too polite some times. So there probably is no end to the tweaking......
Sorry for al my long winded emails
You seem to like it just a bit warm, just like me...
Any hair raising shiver down your spine yet?
No need to be sorry they are a treasure trove of good information it just seems unfair for me to ask it of you again when I can find it for myself with a bit of effort
Yep I tried to make a curve like the Toole one but I didn't like it as much as the blue one. Don't now about shivers but I have heard some really good stuff, Diana Krall's Live in Paris is sounding very good and Giant Steps by Buddy Rich is fantastic. Get Lucky is really good and the bass drum and low toms in Steve Earl's Copperhead Road is very atmospheric.
Attachments
Thanks Byrtt, I don't think I can be bothered to mess with APL_TDA at the moment and the next Beta of REW should have a function that will make the graphs look more similar.
For now this is the Spectrogram of the Left and Right Averaged, there are some issues but overall for no treatment I am happy. You can see the peak energy curve is a little wiggly below 200Hz, but it is a whole lot better looking than without correction.
Good point to wait for REW get new feature of similarity had almost forgot about that info myself, your current curve looks simply great and in my eyes far over average accounting result is at listening position.
fluid is general warm
Looking at wavelet could maybe tell about the preference a bit less 500 to 800Hz.
Attachments
Last edited:
I haven't been on this thread in a while and looks like Fluid is well on his way to a world class implementation not unlike Wesayso's towers. Nice work Fluid!
I have been using Focusrite DAC both Solo 2G and 2i4 2G and really like the sound. I took the 2i4 to a Head-FI audio meet and many people were unaware that it could be used as a playback DAC. They said it's sound was up there among the best compared to much more expensive boutique DACs at the meeting. We were listening with high end headphone amps and high end headphones so details were revealed. But in general, people used to paying $1200 for a DAC were very pleasantly surprised at the sound quality. Considering it was USB powered - the noise was very low. I measured at -130dB. Better than most power supplies can muster.
One neat thing with 2i4 is it has 4 channels out. So maybe I can use it to drive active 2 way setup like a miniDSP? But with advantage of having a nice DAC capable of 192kHz and 24bit playback. Oh it also records music at that speed.
My only struggle is making Jriver play nice with my Focusrite. I had issues where Jriver appeared to go commando on me and take over stuff in background and I could not get decent FFT measurement of my amps using REW. But probably an issue of USB drivers.
I have been using Focusrite DAC both Solo 2G and 2i4 2G and really like the sound. I took the 2i4 to a Head-FI audio meet and many people were unaware that it could be used as a playback DAC. They said it's sound was up there among the best compared to much more expensive boutique DACs at the meeting. We were listening with high end headphone amps and high end headphones so details were revealed. But in general, people used to paying $1200 for a DAC were very pleasantly surprised at the sound quality. Considering it was USB powered - the noise was very low. I measured at -130dB. Better than most power supplies can muster.
One neat thing with 2i4 is it has 4 channels out. So maybe I can use it to drive active 2 way setup like a miniDSP? But with advantage of having a nice DAC capable of 192kHz and 24bit playback. Oh it also records music at that speed.
My only struggle is making Jriver play nice with my Focusrite. I had issues where Jriver appeared to go commando on me and take over stuff in background and I could not get decent FFT measurement of my amps using REW. But probably an issue of USB drivers.
Thanks Byrttfluid is general warm
120 Hz, 500 to 800 and 2.7K are where the reflections do some damage. But looking at the Spectrogram lowering the 500 to 600Hz region a little more might be worthwhile to try.
Thanks, progress has slowed significantly at this point because it is now difficult to get an improvement.
My Focusrite sounds pretty good too.
Not sure what that could be but the ASIO drivers work best for me. Does your Focusrite have MixControl software or just drivers? If you have Windows 10 Creators Edition you could test out the native USB2 driver to see if that works with your interface.
ASIO drivers and REW work well for me as well as Java drivers but the MixControl Software is hard to get set properly and that has had me scratching my head a few times trying to work it out.
How have you set it up in the past to try and measure your amps?
Last edited: