Quantcast
DX7 DAC: alternatives to DX7 for playing live without loosing the classic MKI sound - Page 2 - Gearspace.com
The No.1 Website for Pro Audio
DX7 DAC: alternatives to DX7 for playing live without loosing the classic MKI sound
Old 1st January 2019 | Show parent
  #31
Lives for gear
 
🎧 5 years
Quote:
Originally Posted by Ossicle ➡️
To me this video demonstrates quite well how thin and 2D the Dexed sounds compared to the DX7.
And to me this video demonstrates the authors ability to actually use the plugin:
Old 1st January 2019 | Show parent
  #32
Lives for gear
 
adydub's Avatar
 
🎧 5 years
Quote:
Originally Posted by Ossicle ➡️
As said I have heard the difference - I just played my DX7 and then downloaded the Dexed. It's the DX7s amp section that makes all the difference (you know, the heavy weighing stuff the digitally generated sound needs to flow through in order to come out of the output at line level).

Similarly, a CD player will sound different through a 100 eur integrated amplifier and through a 4000 eur integrated with same watts and same speakers. At least to a lot of us.

I think this discussion boils down to the fact that two persons tend to be interested in and tend to focus on different aspects of the same sound. I might focus on the bottom end, bite etc. you might focus on how the higher end rings etc. We care less about the other aspects of the sound. Not realising this - that our ears focus on different aspects of the sounds - results in endless disputes on analogue vs digital, hardware vs software, this converter vs that converter etc.

To me this video demonstrates quite well how thin and 2D the Dexed sounds compared to the DX7. I hear this even more clearly when comparing my DX7 to the plugin in my home studio. It's not only that the Dexed is less bassy; it just sounds less musical. I serioulsy doubt an eq could fix that difference.

The Arturia version does sound impressively close to the real thing, though.

Pretty clear differences there. Dexed does not compare well to the DX or the Arturia at all. As you say, all the bass end has gone missing. There's some more subtle differences too, I actually prefer the tone of the Arturia slightly!
Old 1st January 2019 | Show parent
  #33
Lives for gear
 
adydub's Avatar
 
🎧 5 years
Quote:
Originally Posted by Mutantt ➡️
And to me this video demonstrates the authors ability to actually use the plugin:
Yeah, the Dexed is again missing something isn't it? Pretty close though, and very very good for free.
Old 1st January 2019 | Show parent
  #34
Lives for gear
 
zmix's Avatar
 
🎧 15 years
Quote:
Originally Posted by hairbow ➡️
DX7 DAC is not the difference, it’s the lower quality interpolation of sines when computing frequency modulation. Converters are next to negligible.

As such, the only way to get that sound is a mk1 DX7 or a plug-in that computes the sine waves in a specific way.
It's true that the DAC is the least important aspect of the DX7 timbre. It was a clever combination of a "floating" 12 bit DAC running at a near 50kHz sample rate, and in fact it used multiplexing to convert each voice separately (so actually running at ~800kHz.

Good thread here
Old 1st January 2019
  #35
Gear Guru
 
EvilDragon's Avatar
 
🎧 5 years
Arturia did nail it, that I think we can all pretty much agree upon.

Here's another comparison between the three (note that DX7 V wasn't set to proper velocity scaling so it sounds muffled by comparison - but it does show how close Dexed is):



And a corrected comparison between just DX7 V and the hardware:

Old 1st January 2019 | Show parent
  #36
Lives for gear
 
acreil's Avatar
 
🎧 10 years
Quote:
Originally Posted by Ossicle ➡️
As said I have heard the difference - I just played my DX7 and then downloaded the Dexed. It's the DX7s amp section that makes all the difference (you know, the heavy weighing stuff the digitally generated sound needs to flow through in order to come out of the output at line level).
There's really not much going on in the DX7's DAC and output stages. The floating point DAC performs poorly, but in practice the distortion it introduces is next to negligible. The sample rate is high enough that you can't hear any high frequency images, and because of the way it's time multiplexed, you won't hear any high frequency droop from the zero order hold. There are no VCAs or anything to introduce distortion, and I don't think there's any significant source of slew rate distortion either. The effect of the coupling capacitors, reconstruction filter, etc. is quite mild. I've modeled the circuits and thought the effect was negligible, but these can be adequately emulated with an equalizer if you want to be super picky. The biggest aspect of the analog circuits that software won't emulate is the background hiss.

If you think the DX7 sounds bassier than Dexed, that's fine, but it's down to a minor amount of filtering, certainly nothing to do with the DAC or power supply or whatever other nonsense.

Quote:
To me this video demonstrates quite well how thin and 2D the Dexed sounds compared to the DX7. I hear this even more clearly when comparing my DX7 to the plugin in my home studio. It's not only that the Dexed is less bassy; it just sounds less musical. I serioulsy doubt an eq could fix that difference.
It does seem to be less bassy to me, but that's not a good way to compare them since it's not switching directly between the DX7 and Dexed. Dexed seems to be a very accurate emulation of the DX7, but Plogue chipsynth PortaFM or Aly James Lab FMDrive is a better choice for the dirtier sounding 2 and 4 operator models.
Old 1st January 2019 | Show parent
  #37
Lives for gear
 
Ossicle's Avatar
 
🎧 10 years
This is even more subjective than I'd previously thought. No point in further debating what others should be hearing.
Old 1st January 2019
  #38
Lives for gear
 
daviddever's Avatar
 
🎧 5 years
Honestly, the TX7 sounds slightly different to the mkI DX7, if one really wants to split hairs; thing is, nothing will match the experience of the original, especially if one's intention is to match the editing workflow.

The same could also be said for the TXn16, but that was THE road warrior's DX7 equivalent. I'd forgo picky and aim for the practical, to be honest.
Old 1st January 2019
  #39
Gear Addict
 
1 Review written
🎧 10 years
I did a tour 10 years ago playing with an artist who had an album out that was 100% DX7. I used FM7 in Native Instruments Kore (remember that one?! Static electricity could short the whole sound card. Interesting times! ) There were differences, but after proper velocity scaling and as soon as we added fx and hit the PA we were in the right place since live sound (we used wedges) varies so much between venues. With most IEM mixes any difference perceived when listening critically will be nullified. I’d be much more worried about authenticity in a recording session.

Last edited by analogaddict; 2nd January 2019 at 09:36 AM..
Old 1st January 2019
  #40
Lives for gear
 
Moonwhistle's Avatar
 
I've been playing an MK1 DX7 since I was a teenager. There is nothing magical about it compared to dexed unless you think some noise and muffling is magic.
Old 1st January 2019 | Show parent
  #41
Lives for gear
 
🎧 5 years
Quote:
Originally Posted by Ossicle ➡️
This is even more subjective than I'd previously thought. No point in further debating what others should be hearing.
Everyone is hearing the small difference, we are just saying that with proper tools and knowledge, that difference can be minimized to almost nothing.
Old 1st January 2019 | Show parent
  #42
Gear Guru
 
EvilDragon's Avatar
 
🎧 5 years
Quote:
Originally Posted by Ossicle ➡️
This is even more subjective than I'd previously thought. No point in further debating what others should be hearing.
What acreil speaks of is NOT subjective, tho. It's actual, proper facts, which you failed to provide.
Old 1st January 2019 | Show parent
  #43
Gear Nut
 
🎧 5 years
Quote:
Originally Posted by Ossicle ➡️
As said I have heard the difference - I just played my DX7 and then downloaded the Dexed. It's the DX7s amp section that makes all the difference (you know, the heavy weighing stuff the digitally generated sound needs to flow through in order to come out of the output at line level).

Similarly, a CD player will sound different through a 100 eur integrated amplifier and through a 4000 eur integrated with same watts and same speakers. At least to a lot of us.

I think this discussion boils down to the fact that two persons tend to be interested in and tend to focus on different aspects of the same sound. I might focus on the bottom end, bite etc. you might focus on how the higher end rings etc. We care less about the other aspects of the sound. Not realising this - that our ears focus on different aspects of the sounds - results in endless disputes on analogue vs digital, hardware vs software, this converter vs that converter etc.

To me this video demonstrates quite well how thin and 2D the Dexed sounds compared to the DX7. I hear this even more clearly when comparing my DX7 to the plugin in my home studio. It's not only that the Dexed is less bassy; it just sounds less musical. I serioulsy doubt an eq could fix that difference.

The Arturia version does sound impressively close to the real thing, though.


Yes I did the blind test few month ago on a video on youtube ... Arturia vst is the closest sounding to the DX7 MKI .. I couldn't hear the difference ... while the Dexed and all others DX7 sounds different. We can focus more on brilliance or on low end, but the result is the same : they don't sound the same. Period ! and it's kinda bothering me because I hate the war between analog and digital, gear suppose to sound better because it's analog or because it has these filters ... but most of time it's true ... and I want to make music but I want my sound not to be sterile,
Old 2nd January 2019
  #44
Gear Guru
 
EvilDragon's Avatar
 
🎧 5 years
DX7 is not "analog", tho.

Also, implying Dexed for some reason sounds "sterile" is absolutely unfair, and, well, untrue, too. And post #31 shows that it works in the mix as good as an actual DX7. No trace of any sort of "sterility" there...
Old 2nd January 2019
  #45
Lives for gear
 
1 Review written
🎧 10 years
Losing*
Old 2nd January 2019
  #46
Lives for gear
 
🎧 5 years
If your music is so impacted by using Dexed over a mk1 DX7, then the quality of your music is so marginal that a teeny tiny difference between synthesizers was the deciding factor.

I think it was summed up best earlier - for recording, do whatever you want to feel good about it, but no one cares a single bit in a live setting
Old 2nd January 2019 | Show parent
  #47
Lives for gear
 
Ossicle's Avatar
 
🎧 10 years
Quote:
Originally Posted by EvilDragon ➡️
What acreil speaks of is NOT subjective, tho. It's actual, proper facts, which you failed to provide.
I haven't seen any "facts" in this thread which would undisputably show that I shouldn't hear a significat difference between the hw and sw. acreil's post is about how the circuitry affects the sound (those might be facts), and his subjective experience of this. He is of course entitled to that and no one should argue against his experiences, but it is nothing but subjective. This seems to be difficult for some to comprehend.

Some of us hear a clear difference between DX7 and software and some (apparently) don’t, and there is obviously not much that can be said to change this. Many of us trust our own ears more than someones explanations in a short forum post (that includes my own posts).

You seem to have difficulties to accept that. It's rather strange to argue here so strongly against what people hear but never even compared the DX7 to the software first hand yourself .

The OP hears the diffenrence and wants suggestions. I don't have further contributions to this thread.

Last edited by Ossicle; 2nd January 2019 at 11:24 AM..
Old 2nd January 2019 | Show parent
  #48
Lives for gear
 
monomer's Avatar
 
🎧 10 years
In those comparison video's Dexed seems to have 0-time attacks (basically full bandwidth clicks).
Both the DX and the arturia seemed to have a slighly softer attack (not quite 0-time clicking).

Could be a partial explanation for why Dexed sounds brighter and more 'digital'.
Old 2nd January 2019 | Show parent
  #49
Gear Guru
 
EvilDragon's Avatar
 
🎧 5 years
Quote:
Originally Posted by Ossicle ➡️
You seem to have difficulties to accept that. It's rather strange to argue here so strongly against what people hear but never even compared the DX7 to the software first hand yourself .
How do you know I didn't do the comparison myself? I never said anywhere I did or didn't. Bravo for implying things incorrectly.
Old 2nd January 2019 | Show parent
  #50
Lives for gear
 
Ossicle's Avatar
 
🎧 10 years
Quote:
Originally Posted by EvilDragon ➡️
What acreil speaks of is NOT subjective, tho. It's actual, proper facts, which you failed to provide.
I did explain the reasons that I think explain the sound differences in post #29 . Sorry if you failed to understand it. Others did though.

Quote:
Originally Posted by EvilDragon ➡️
How do you know I didn't do the comparison myself? I never said anywhere I did or didn't. Bravo for implying things incorrectly.
Well, did you?
Old 2nd January 2019 | Show parent
  #51
Lives for gear
 
monomer's Avatar
 
🎧 10 years
Quote:
Originally Posted by EvilDragon ➡️
How about you then prove how the PSU affects the sound in a synthesis system that is NOT voltage controlled?
One doesn't exclude the other.
DACs , for instance, are affected by power input.
So are op-amps.

So a 'NOT-voltage-controled' synth still has parts that are affected by the power supply.
Old 2nd January 2019
  #52
Gear Guru
 
EvilDragon's Avatar
 
🎧 5 years
But truthfully, how much would minor fluctuations in supplied voltage affect the sound output? Would it make it totally "un-3D and sterile"? Of course not. Maybe a very minute difference in THD, surely, but that's nothing to write home about.


For the record, I did compare Dexed to a mkI as I have access to one mkI (my old neighbor has it), and while their outputs weren't identical (which is of course impossible, I have never implied that to be the case), to me they sounded about 99.9% the same. Certainly doesn't sound "sterile and un-3D" by comparison. Slap a carefully tweaked EQ on Dexed and they are totally replaceable in a mix. I will say that with Arturia's DX7 V, there's no need to slap that extra EQ. They really nailed it.

As Robotunes said, go make some fcking music.
Old 2nd January 2019 | Show parent
  #53
Lives for gear
 
Ossicle's Avatar
 
🎧 10 years
Quote:
Originally Posted by EvilDragon ➡️
As Robotunes said, go make some fcking music.
Yeah, that's what we should all do.
Old 2nd January 2019 | Show parent
  #54
Gear Guru
 
EvilDragon's Avatar
 
🎧 5 years
Quote:
Originally Posted by Ossicle ➡️
Yeah, that's what we should all do.
Yay, one thing we agree on.
Old 2nd January 2019 | Show parent
  #55
Lives for gear
 
🎧 5 years
Quote:
Originally Posted by Ossicle ➡️
Some of us hear a clear difference between DX7 and software and some (apparently) don’t
Can you list those who can't hear the difference ?

I can only list one name who thinks he can't fix that difference in his DAW:
1. Ossicle.
Old 2nd January 2019 | Show parent
  #56
Lives for gear
 
adydub's Avatar
 
🎧 5 years
Quote:
Originally Posted by Mutantt ➡️
Can you list those who can't hear the difference ?

I can only list one name who thinks he can't fix that difference in his DAW:
1. Ossicle.
You can add me to that list (for Dexed), it sounds like more than just EQ differences.

I wouldn't actually have a problem with this though, for FM I use my TG77, Prophet 12 and sometimes Ableton Operator and am perfectly happy. But I think it's entirely reasonable that if someone wants an emulation, then it should be as close to 100% as possible. And the Artuia sounds much closer than Dexed. I'm not sure why there's so much salt in this discussion.

EDIT: and my OP-1, that's got some great, not entirely clean, FM sounds in it.
Old 2nd January 2019 | Show parent
  #57
Lives for gear
 
acreil's Avatar
 
🎧 10 years
Quote:
Originally Posted by Ossicle ➡️
I haven't seen any "facts" in this thread which would undisputably show that I shouldn't hear a significat difference between the hw and sw. acreil's post is about how the circuitry affects the sound (those might be facts), and his subjective experience of this. He is of course entitled to that and no one should argue against his experiences, but it is nothing but subjective. This seems to be difficult for some to comprehend.
Do an AB comparison yourself then to show us how obviously different they are. You can show the effects of the DAC with a sine that has a very long release time (set it to 10 or so). Using any algorithm with 2 carriers, program identical envelopes for both carriers. Both should be ratio 1.00 with no detuning and maximum amplitude. The SYNC parameter should be on. Hold a single low note (like C2) for a couple seconds, then release it. The distortion will be visible on a spectrogram. You can show the effects of sine table quantization by setting a single modulator to maximum amplitude at the lowest fixed frequency. Ideally you should only hear only vibrato, but on the DX7 (and accurate emulations of the DX7) there will be a high tone around the 4096th harmonic of the modulator frequency. Verify the implementation of the moving average filter in the feedback path by setting the feedback to 7 and increasing the operator amplitude to the point that it produces a high frequency tone, before it breaks into noisy oscillation. The tone should be 1/3 the sample rate. In the DX7 it will be 16365 Hz. Verify the shape of the envelope attack by playing a sine tone with a very long attack time. Verify that the amplitude envelopes don't reset when a note is triggered by using the first envelope stage as a "delay" (long attack time and level under about 50) and the second stage as a percussive attack. Use a long release time (maybe 20-30). The delay time can be measured by comparing it to a second operator with an instantaneous attack. After the 16th note is played, the delay will be shortened by an amount that depends on the current level of the amplitude envelope when a new note is assigned to that voice. Test the control rate of the software-generated modulation (pitch envelopes, LFO, portamento) with a fast pitch envelope. The steps should be visible on a spectrogram. You can show the effects of the output stages by playing sine portamento from maybe 10-100 Hz, and another from maybe 4 kHz until it's audible as aliasing. Record everything at 96 kHz and compare to identical tests in Dexed.

And finally, test your idiotic magical power supply hypothesis by replacing all the bypass capacitors on the voice board with larger capacitors (maybe 10x the specified value) having the lowest ESR and series inductance that you can find. Add 0.1 µF film capacitors in parallel with each of these to further reduce the ESR, and maybe add another film capacitor across the power leads of each IC for good measure. Power the voice board from a good quality bench supply. Measure power supply ripple at critical points in the circuit (the DAC, the analog switch for the floating point resistor ladder, sample and holds, op amps in the analog section) while playing a tone, both before and after the modification. Of course you'll want to undo everything when you're done so that you don't have excessive inrush current when turning it on. Don't break it. Good luck. You might also want to compare the DX7 to the TX7 and TX816 since these have completely different power supply designs.
Old 2nd January 2019
  #58
Gear Guru
 
EvilDragon's Avatar
 
🎧 5 years
THOSE things listed are NOT subjective - they are objective tests.

Very well QED'd, sir.
Old 2nd January 2019 | Show parent
  #59
Lives for gear
 
monomer's Avatar
 
🎧 10 years
Quote:
Originally Posted by EvilDragon ➡️
But truthfully, how much would minor fluctuations in supplied voltage affect the sound output? Would it make it totally "un-3D and sterile"? Of course not. Maybe a very minute difference in THD, surely, but that's nothing to write home about.
Well, i was talking generally, after you made a remark that nothing in a digital synth could be affected by the power supply.

BTW, I had a look at the schematics. The output stage of the DX7 seems pretty complex to me for such a synth. At least, more complex than i imagined.

First of all, the reference voltage for the DAC is taken just from the global +5 and Ground. Not a very good design for a DAC, but on the other hand, the 12 bit DAC has a pretty high noise floor so it would probably at least partially mask influence from power supply.
The weird thing is that this reference voltage is applied in stages. There is a resistor network feeding a range of voltages into a multiplexer that then feeds the reference voltage input of the DAC.
I assume this is to ramp up the output of the DAC at startup to prevent clicks or something like that.

There are several filters in the output path, but they seem pretty high (40 something kHz and up) so that won't have a lot of influence on the highs.

There are several op amps (as buffers and mostly part of the filters) that may or may not have an influence, depending mostly on their specs and incoming signal voltage.

There seems to be some strange stage on the analog output that i think is a digitally controled master volume. Not sure how that influences the signal.

In any case, the signal does go through several stages before it reaches the output jack. I expected less of it.
Old 2nd January 2019 | Show parent
  #60
Lives for gear
 
monomer's Avatar
 
🎧 10 years
Quote:
Originally Posted by acreil ➡️
Do an AB comparison yourself then to show us how obviously different they are.

-snip-
What is your opinion on the clock source? Could jitter be a thing?
📝 Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 112 views: 19750
Avatar for in media res
in media res 10th May 2021
replies: 550 views: 65797
Avatar for syntonica
syntonica 12th November 2020
replies: 2418 views: 162837
Avatar for pmboos
pmboos 2 days ago
Topic:
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump