Topping DX5 II

[Dandelion]

Can someone explain why there are so many different filter types? Isn't the goal the best sound, with the fewest artifacts?
Which is the best for a punchy, natural sound with clear highs?
Which are the differences between those filters in terms of sound?
I received my dx5ii today and I ll set it up in the next 2 days.
Is there anything else I should pay attention to when setting it up?
I ll connect it with usb to my pc and I ll surely use the p-eq, after I take measurements with a specialized microphone I have.
I have an integrated amp which can also operate as a power amp(yamaha dsp a2070).
Do you like dx5ii as a preamp?

There’s a selection because any filter comes with some trade-offs of some kind, may it be phase distortion or pre-ringing or slope or ripple, etc. personally the first thing I’m fine with trading is the steepness near nyquist, past 18kHz personally is irrelevant, so I’m fine having a filter that starts to roll off there slightly and I don’t need something that rolls-off -0.1db at 19.5kHz if it then messes with the phase or transient more. As well ripples in today’s day and age are fairly irrelevant.

 

[Musictech77]

For me when i switched from F1 to F2 it sounded a bit better. A bit clearer.

I flicked through all of them and the linear phase filters sound the best to me (so F2,3,4,5). Don't know what it is, a bit more depth and clarity maybe. It's subtle though.

F1 sounds the least good to me, which is interesting cos it's the default one it ships with.

It's pretty subjective though i think. You just gotta flick through them and see for yourself.

As for other things to pay attention to: Download the driver, the latest firmware (follow the instructions and be careful), and the Topping Tune app for access to PEQ. When you make an EQ profile, right click on it and click "store to device", then you can run the PEQ profile on the unit without having to load the Topping Tune app. So basically, once you set the profiles on the app, you don't have to load the app again, which i appreciate very much. The drawback of that though is that the EQ gain settings only work in 0.5db increments, which is annoying. So things can only be 0.5, 1.0, -1.5, 2.0, etc. I don't know why that restriction has to be there. People with OCD will be pulling their hair out, lol. But hey, having on-board EQ is better than none at all, so i'll deal with it.
Anyway then enable PEQ on the device menu.

Thank you!

**I think that 0.5 db steps are fine for room correction purposes. I find most annoying that you cannot adjust separately the eq for L and R channels...

 

[Musictech77]

There’s a selection because any filter comes with some trade-offs of some kind, may it be phase distortion or pre-ringing or slope or ripple, etc. personally the first thing I’m fine with trading is the steepness near nyquist, past 18kHz personally is irrelevant, so I’m fine having a filter that starts to roll off there slightly and I don’t need something that rolls-off -0.1db at 19.5kHz if it then messes with the phase or transient more. As well ripples in today’s day and age are fairly irrelevant.

So.. Which one is your choice and why?

 

[ziggurcat]

I find most annoying that you cannot adjust separately the eq for L and R channels...

You can.

 

[Musictech77]

You can.

Great! I guess that it was fixed in the latest updates cause I read in this thread few weeks ago that it was not piossible

 

[Bleib]

Hmm, they seem to have removed all DX5 II firmwares

Downloads - TOPPING

www.toppingaudio.com

 

[Raymond lin]

Hmm, they seem to have removed all DX5 II firmwares

Downloads - TOPPING

www.toppingaudio.com

I noticed this too and came here to check it wasn't just me.

 

[ziggurcat]

Great! I guess that it was fixed in the latest updates cause I read in this thread few weeks ago that it was not piossible

It has always been possible. It was not possible with the D50 III.

 

[Dandelion]

So.. Which one is your choice and why?

Mind you technically speaking is music dependent, sine sweeps, noise and what not are useful for the sake of providing standardized measurements but such filters react to transients and stereo information of the underlying material and no song is static or the same. I default at first to a not too steep linear filter (as I said in earlier reply I trade-off a bit of linearity past 18kHz and ripples), I then spot a minimum phase similar alternative and I toggle between the two.
Usually linear phase feels clearer and with more air and spacious, but loses some hit and some meat. If I feel in most of the music I throw under it with the linear one I lose transient hit and the minimum phase alternative sounds often with more bottom end, then I opt for the minimum phase. Today's linear phase though is pretty good. It usually takes me about 30 minutes of testing, I throw at it music with lots of reverb and stereo information, very dry techno, loud music, more dynamic music, from the 60s with a lot of panning, from the 70s, classical music, and so on. I try to put under it a whole bunch of music. Personally I feel the difference with quite some consistency, for example with digital tools I close my eyes and toggle between linear and minimum phase a whole bunch of times until I lose track of what's active, I then slowly switch and recognize which is which. I may have just trained myself into it, but I can spot it for a good 80% of the times and then I take a decision on which filter to use. Your mileage might vary.

 

[Airy]

I have found an album that is causing loads of 'HP DC abnormal' errors. It doesn't require a reboot, but it does cause dropouts in the audio when it happens which is really annoying.

My understanding is that this only affects headphones; is that correct? If so, shouldn't this only be triggered when the output is set to one of the HPA options, and not LO?

For those interested, the album is Ellipse by Imogen Heap. Most tracks do it, but I was getting quite a few on the first track - First Train Home.

Hey i checked this out. and other songs that cause the HP DC ABNORMAL stutters. turns out you can JUST turn down the volume on Spotify. https://files.catbox.moe/mvkfc6.mp4 here's is the vid for proof. With Spotify volume at around 75% you can still turn up the volume WAY loud and not trip the dc protection, but with Spotify volume max, it easily triggers it even at somewhat~ loud volume. Hope this helps you.

 

[ziggurcat]

Hey i checked this out. and other songs that cause the HP DC ABNORMAL stutters. turns out you can JUST turn down the volume on Spotify. https://files.catbox.moe/mvkfc6.mp4 here's is the vid for proof. With Spotify volume at around 75% you can still turn up the volume WAY loud and not trip the dc protection, but with Spotify volume max, it easily triggers it even at somewhat~ loud volume. Hope this helps you.

I think because it's pretty much gain-dependent. I seem to recall that there's a loose "rule of thumb" to set the output gain from the source (if possible) at around 80% instead of 100%, which would probably solve a lot of the DC protection kicking in for people.

 

[jonwb]

I also noticed that Topping took down all the firmware downloads. Even with 1.62 adding SPDIF modes to the menu, I'm still having issues with the Coax connection cutting out.

I'm hoping this means a new firmware update is in the works to address this issue.

 

[Sokel]

I think because it's pretty much gain-dependent. I seem to recall that there's a loose "rule of thumb" to set the output gain from the source (if possible) at around 80% instead of 100%, which would probably solve a lot of the DC protection kicking in for people.

DC has nothing to do with level, it's bad at 10mV and it's also bad at 10V.
It should be blocked no matter what.

If it's baked in (bad recorded) music or gear outputs it on their own (other than ditching both) , it should be addressed.

 

[yorkieaudio]

Still no word on official word on why they pulled the firmware, but does seem it may be bricking units -

https://www.reddit.com/r/BudgetAudiophile/comments/1n23l1u/do_not_use_the_dx5_ii_v162_firmware_it_may_brick

 

[jodrim]

Still no word on official word on why they pulled the firmware, but does seem it may be bricking units -

https://www.reddit.com/r/BudgetAudiophile/comments/1n23l1u/do_not_use_the_dx5_ii_v162_firmware_it_may_brick

Yes the same thing happened for me so guessing they got reports of it and pulled it from their site. Not sure why they wouldn't repost the 1.57 firmware download page though since that is pretty well tested. But it can still be downloaded from them here if same thing happens to someone else:

https://dl.topping.audio/usermanual/DX5_II_FW1.57.zip

 

[ziggurcat]

Yes the same thing happened for me so guessing they got reports of it and pulled it from their site. Not sure why they wouldn't repost the 1.57 firmware download page though since that is pretty well tested. But it can still be downloaded from them here if same thing happens to someone else:

https://dl.topping.audio/usermanual/DX5_II_FW1.57.zip

That seems like the likely reason for pulling. Hopefully they release an updated FW soon for those who had their units brick.

 

[ziggurcat]

DC has nothing to do with level, it's bad at 10mV and it's also bad at 10V.
It should be blocked no matter what.

Oh, sure, but that's not exactly what I was saying. The DC protection kicking in is very clearly gain-dependent, which is why rolling back Spotify's output resulted in that person being able to crank the volume, and not trigger the protection.

 

[Dandelion]

I think because it's pretty much gain-dependent. I seem to recall that there's a loose "rule of thumb" to set the output gain from the source (if possible) at around 80% instead of 100%, which would probably solve a lot of the DC protection kicking in for people.

Yes that's because of positive true peak values due to non-tp limiting and/or codec conversion artifacts, also as a rule of thumb never push anything to 100% for the same reasons, especially passive speakers and such, positive true-peak values become a problem in such scenarios (literally frying them).

DC has nothing to do with level, it's bad at 10mV and it's also bad at 10V.
It should be blocked no matter what.

If it's baked in (bad recorded) music or gear outputs it on their own (other than ditching both) , it should be addressed.

That's because it's not "true" DC offset protection, this thing monitors sub frequencies with a given level and sustain. it's hard to detect true DC on music since you can't know the causes, there's many things that can make it appear as DC even if they actually aren't: aliasing artifacts of digital DSPs used in song production/mixing/mastering, then there's actual DC from outboard gear or coded into digital emulations of such, poor eq and/or limiting decisions, etc. In theory only DC offset should be filtered out or protected from, the rest is simply poor quality bottom-end of the songs/audio material.

Also I still think this so called protection should just activate/deactivate a DC filter at 1hz instead of being too invasive on the listening experience. Again, pretty much totality of consumers audio doesn't have it, nor dj equipment, nor studio equipment.

 

[Patalf le gris]

I received my DX5 II this morning. A delight. I was doing some comparative listening between the Topping's direct headphone output and via my Monolith THX 887, listening on a Meze 105 AER and a Monolith M1070. But suddenly the Topping stopped with the message "HP DC Abnormal" and did not restart... Fragile?

 

[ziggurcat]

I received my DX5 II this morning ... Topping stopped with the message "HP DC Abnormal" and did not restart... Fragile?

That's the DC protection kicking in, usually when the volume level is on the high end.