The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Synths for sale  Search Gearslutz Go Advanced
Analogue vs Digital - Diva and OB8 test Keyboard Synthesizers
View Poll Results: Which synth is First in the 8 bar cycle?
Diva
92 Votes - 51.11%
OB8
88 Votes - 48.89%
Voters: 180. You may not vote on this poll

Old 24th July 2014
  #571
Gear Maniac
 

Holy moly, the tables have turned in the poll.
Old 24th July 2014
  #572
Gear Guru
 
Yoozer's Avatar
Poll is effectively irrelevant since you can now see the outcome and vote like you knew it all along. Hence the screenshot of the people who originally did their guesses.

2
Share
Old 24th July 2014
  #573
227861
Guest
Just noticed there's a member named Juno-106! ha!
Old 24th July 2014
  #574
Quote:
Originally Posted by SWAN808 View Post
yeah I thought unison would draw out some differences - there is a unison bass on this track...but Ill do some open filter and unison sync sounds at some point...
Oops, I wasn't sure if the bass was part of the test so I didn't really focus on it. Kind of hard to do anyways with those sweet Obie brass chords on top. I thought the bass was maybe another synth.
Old 24th July 2014
  #575
Gear Maniac
 

Quote:
Originally Posted by Yoozer View Post
Poll is effectively irrelevant since you can now see the outcome and vote like you knew it all along. Hence the screenshot of the people who originally did their guesses.
No ****?
Old 24th July 2014
  #576
Lives for gear
 
Thunderkyss's Avatar
Quote:
Originally Posted by Don Solaris View Post
Depends who supplies you with synths.

TBH i don't hear much of the muffled sound out of my vintages, and i have some 20-25 or so of them around.
I think that "muffled sound" is just an inacurrate attempt to describe what is missing. I've never been one to say I can here the difference, though I did participate in the voting. I guessed the first one was Diva because it was "better" & I don't mean that as an accurate description. Just the best I can come up with now.

I don't believe the difference is something we can hear over the Internet. I think you need to be in the room.... just you & the instrument. Some people will connect in a way others won't. It's just the way it is.

I can't imagine someone falling in love with a piece of software... I mean "true love" then again, some people can't imagine falling in love with an old box of junk.
2
Share
Old 24th July 2014
  #577
Gear Maniac
 

I'm pretty much calling my battery of plugs my kids. There's feelings involved.
Old 24th July 2014
  #578
I think what we can say though is many of us did hear a difference. But we were also told there was a difference, we were looking for it. With the sound in isolation we were almost 50/50 about which was which, meaning at the very least Diva was subjectively in no way inferior to the OB.

So Diva sounds as good as an analog synth in this situation, and probably in many situations. It doesn't matter if it sounds identical unless you must have that OB sound exact for whatever reason.

I think there are multiple aspects that many people have merged together to get to this hardware is better software paradigm. Sound is not the only part of it. Sometimes we just prefer something that is tactile. Other people love the convenience of software. Personally I hear things differently when I am working in front of a screen vs working on a synth surface.

In the end it's well worth exploring both options for anyone. There is nothing wrong with buying analog synths if that's what inspires you.
1
Share
Old 24th July 2014
  #579
Quote:
Originally Posted by Eigenwert View Post
You are mixing up things. What if the so called "non-experts" just prefered DIVA, outweighing those with the correct choice?

If you re-read this thread, especially the part before the reveal, you might find that your statement dismisses the fact that it was possible to be sure which was which and that there also has been an explanation, why.

And it's pretty bold to name the "non-experts" "idiots". I am pretty sure that nearly everyone, even some of the top producers on GS would be an "idiot" next to a graduate of the McGill University then, because almost no other person will be able to hear the slight EQ curve changes they teach you to hear over there.
The question was not which do you prefer, though. It was, which is first, Diva or OB8?

You are dismissing the fact that there are a lot of people here who, while not an expert in hardware like the OB8, probably use Diva WAY more than the so-called "experts". So likewise, they should have been able to distinguish which was Diva. Urs got it. I imagine he knows what Diva sounds like better than anyone.

I used the word "idiots" tongue in cheek, obviously, and I used it as part of a ficticious scenario where the responses were not 50/50. Anyone around here knows I'm a gentleman.

Look, I didn't twist anything up. It could not be more obvious what the situation is here. This isn't a complicated piece of science here. It was a poll with two options. There isn't much room for discussion and analysis, except by those who will refute anything that doesn't benefit their self perception. Even if I got it right, which I did not, I would still say that based on the numbers, people couldn't tell. Those that are right, and swearning they knew it....well there are exactly as many on the wrong side who thought they knew.

It's day one of Junior College Statistics 101, not McGill U we are talking about here. It's a fourth grade taste test, Coke or Pepsi. The difference is, in that case, people can distinguish, and the results would not be 50/50.

This has been fun, really. You have your opinion, I have mine. But the numbers are the numbers, and you can only argue them away so much. SWAN808 demonstrated that in skilled hands, Diva can be close enough, that even those with better hearing than most, can't tell the difference.

If we repeated this same test with a different song, I think we would find similar reaults. SWAN808 needs to be the one who develops the test again though for accuracy. I find just as many people on the wrong side with me, who I am pretty sure know what they are doing, as I do see people I know have limited experience on the correct side.

Honestly though, if SWAN808 isn't up for it.....it is okay by me, because all this has shown, is that in the face of the obvious, some people will never change their view. That is not the mentality of a scientist. Only in the science community can someone argue til blue in the face.....and then following data that refutes their position, politely congratulate the others, and change his or her own perspective.....setting out to perform new science to advance their new position and understanding. I wish more people in this world were so gracious as nerds that live in labs and make no money, but do it for the advancement of knowledge.

I do not believe I could make the same quality replica as was demonstrated by the OP. I don't think most could. This was a unique opportunity for us all. I thank SWAN808 for such a great job.

Let the bickering continue. You guys are fun and make my daytime a little less painful.
3
Share
Old 24th July 2014
  #580
Registered User
Quote:
Originally Posted by NewsFromTheSky View Post
Diva has slop too. It has separate trimmers for VCO detune, max detune, cutoff slop, envelope slop, PW slop, and glide slop, plus drift and individual voice detuners.
Yes, and as I understand it, this level of detail contributes to the CPU load.

There's really no excuse for people to discuss Diva without being educated about it. Like most software instruments, there's a fully functional demo available at the u-he site.

Download the demo and play with it for a few hours or a few days. Don't get hung up on whether it can emulate a Jupiter-8 with 97 or 99% accuracy. Just see if you can make cool music with it.
Old 24th July 2014
  #581
Gear Addict
 

Quote:
If there are people who can reliably tell the difference, there's no evidence that any of them took this test.
Quote:
...first is Diva. It's cleaner and more focused. It's also louder which is a big unfair advantage (the RMS average may have been matched but I would have used my ears to do a final level check).
excuse me? I couldn't have been more succinct in my reply above. I also find it interesting that some of the people I have the most respect for on the forum, and ones that I know own and have vast experience with analog synths as well as software, also knew the difference.
2
Share
Old 24th July 2014
  #582
Registered User
EDGEK8D makes some great points above.

I find myself thinking that you could post two recordings that were secretly identical, and a brawl would erupt here over which one sounded better.
1
Share
Old 24th July 2014
  #583
Gear Maniac
 

Quote:
Originally Posted by SWAN808 View Post
Thanks man for doing some audio with your JP8...Mmm on listening - something doesnt sound right about this patch on Diva...it can be much brighter than that I think...
Hehe. It can get easily brighter than both of his examples with the analog2 shape. Both Diva and his Jupiter examples were frequency limited to 16kHz!
Old 24th July 2014
  #584
Lives for gear
 
GearAndGuitars's Avatar
 

I hereby declare that this thread is over and I win!
Old 24th July 2014
  #585
Lives for gear
 
grumphh's Avatar
 

Quote:
Originally Posted by keybdwizrd View Post
I find myself thinking that you could post two recordings that were secretly identical, and a brawl would erupt here over which one sounded better.
Nahhh, people here have to good ears for that trick

The interesting thing about the test was not that the sounds are different - because most can probably hear that there are differences - but which one was in fact generated by analogue circuitry and by not ones and zeros.

That was the tricky bit - and in this case tricky enough to fool quite a few people.
2
Share
Old 24th July 2014
  #586
Quote:
Originally Posted by Urs View Post
Well, if we're talking simple circuits such as state variable filters built with two or three OTAs, two capacitors and a bunch of resistors then MNA (Modified Nodal Analysis) can be boiled down to a set of non-linear equations that can indeed be solved in realtime. That's what we do.

Naturally that's not a general approach, but the maths used for MNA is equivalent to the maths we use in our optimised scenario. Just, instead of a matrix we use the equations directly to reduce the number of dependencies that need to be solved iteratively. It's a hell of a lot of work, it still smokes CPUs, but it's worth doing.
Urs, this seems like an opportunity to learn from a master: I can see how a voltage controlled current source (OTA) can be modeled simply, but my above point above was relevant to direct DSP-based real-time COMPONENT LEVEL modeling. Johnson noise in carbon comps, tolerance dependent non-linearities of capacitors at audio rates, and the individual distortions and slew limitations of active devices....not just a thevenin equivalent or generalized equation as is currently possible, but literally an audio path determined with a true component level model. I'd love to see an "aging" knob where each component breaks down randomly per instance. I'm sure with an awful lot of work you could describe each component mathematically in the individual sense, but computing the audio and control paths in real-time can't be possible with current processing power available to anyone outside of university CPU farms! My specialty was analog filter design, so what's possible with modern DSP is not something I'm truly informed on.... So if I'm wrong, please educate me!
Old 24th July 2014
  #587
would be quite cool to age the capacitors and muffle up the sound...
Old 24th July 2014
  #588
Gear Addict
 

Quote:
would be quite cool to age the capacitors and muffle up the sound...
... "I have just the thing for you. For three easy payments"
I have several bins of old & used caps saved because customers do sometimes request old parts or claim you f'd up the sound etc... I know... crazy...

Old 24th July 2014
  #589
Gear Addict
 

...[off topic].. and then there's this beauty I bought recently as a backup.. some moron had removed all the SMD elec caps from a JP-8000 mainboard (bottom one) and did a hack job replacing with a random scrap of through-hole part! Yay, replacing perfectly fine filter caps on the mainboard of a digital synth... really improves the sound. I think some people are reading GS too much!

1
Share
Old 24th July 2014
  #590
Lives for gear
 
Quantum7's Avatar
Diva is still the only softsynth that I have used that truly made me "feel" like I was using Ye Olde Analog. I use and love a software synths, but Diva is from a different planet IMO.
Old 24th July 2014
  #591
Lives for gear
 
Thunderkyss's Avatar
Quote:
Originally Posted by EDGEK8D View Post
It's day one of Junior College Statistics 101, not McGill U we are talking about here. It's a fourth grade taste test, Coke or Pepsi. The difference is, in that case, people can distinguish, and the results would not be 50/50.
I think some of us are getting carried away with the 50/50 results. This is hardly a controlled group. I'm pretty sure it doesn't mean what we're saying it means.
1
Share
Old 24th July 2014
  #592
VST
Lives for gear
 
VST's Avatar
Quote:
Originally Posted by azone2 View Post
...[off topic].. and then there's this beauty I bought recently as a backup.. some moron had removed all the SMD elec caps from a JP-8000 mainboard (bottom one) and did a hack job replacing with a random scrap of through-hole part! Yay, replacing perfectly fine filter caps on the mainboard of a digital synth... really improves the sound. I think some people are reading GS too much!

LMAO
Apologies in advance if it was someone in the thread.
Old 24th July 2014
  #593
Registered User
Quote:
Originally Posted by grumphh View Post
That was the tricky bit - and in this case tricky enough to fool quite a few people.
And perhaps even some of the people who got it correct did so because they guessed correctly.
Old 24th July 2014
  #594
Registered User
Quote:
Originally Posted by jimmyklane View Post
Johnson noise in carbon comps, tolerance dependent non-linearities of capacitors at audio rates, and the individual distortions and slew limitations of active devices....not just a thevenin equivalent or generalized equation as is currently possible, but literally an audio path determined with a true component level model.
I'd prefer to think of it like this.

Old 24th July 2014
  #595
Lives for gear
 
grumphh's Avatar
 

Quote:
Originally Posted by keybdwizrd View Post
And perhaps even some of the people who got it correct did so because they guessed correctly.
The key word being "guess" - 50/50 distribution of answers doesn't lie heh
1
Share
Old 24th July 2014
  #596
Registered User
 

Hindsight Bias - Glossary of Psychology Terms:

Definition:

The term hindsight bias refers to the tendency people have to view events as more predictable than they really are. After an event, people often believe that they knew the outcome of the event before it actually happened. The phenomenon has been demonstrated in a number of different situations, including politics and sporting events. In experiments, people often recall their predictions before the event as much stronger than they actually were.

For example, researchers Martin Bolt and John Brink (1991) asked college students to predict how the U.S. Senate would vote on the confirmation of Supreme Court nominee Clarence Thomas. Prior to the senate vote, 58-percent of the participants predicted that he would be confirmed. When students were polled again after Thomas was confirmed, 78-percent of the participants said that they thought Thomas would be approved.

The hindsight bias is often referred to as the "I-knew-it-all-along phenomenon."

...
1
Share
Old 24th July 2014
  #597
Lives for gear
Quote:
Originally Posted by grumphh View Post
Ever since statistics were invented.
(I know, you are a musician, so you don't need to know this )

But statistically speaking a 50/50 outcome between two choices means that the results rely on pure chance.

I.e. 100% of people were guessing or in other words 0% got it right. heh
This is not the statistical interpretation. In statistics, you have a hypothesis (in this case, people can tell the difference between Diva and an OB8) and a null hypothesis (people cannot tell the difference between Diva and OB8 and are just guessing). When you do not have a significant difference between outcomes, i.e. how many people guess synth a vs synth b, it means you cannot reject the null hypothesis because you are too likely to have seen a result caused by random chance. It does not mean that the null hypothesis is true. So the correct interpretation is that we cannot rule out that everybody was guessing, not that we know that everybody was guessing. Which makes sense, because you can't tell from a 50-50 split whether some people actually could tell the difference but were obscured statistically by people that couldn't.
4
Share
Old 24th July 2014
  #598
Lives for gear
 
maisonvague's Avatar
 

Quote:
Originally Posted by EDGEK8D View Post
Urs got it. I imagine he knows what Diva sounds like better than anyone.
Nope. He was only guessing. Just like everyone else.

Quote:
Originally Posted by prontold View Post
This is not the statistical interpretation.
Nope. It's Grumphh's Law. If grumphh can't hear it, no one can.
Old 24th July 2014
  #599
227861
Guest
Quote:
Originally Posted by SWAN808 View Post
would be quite cool to age the capacitors and muffle up the sound...
Can you be kind enough to share your patches for this test?

Do you have your test in separate tracks that can be posted so we can hear each isolated? Would give more credibility. Did you use any processing on each?


(see reason below)



Quote:
Originally Posted by keybdwizrd View Post
EDGEK8D makes some great points above.

I find myself thinking that you could post two recordings that were secretly identical, and a brawl would erupt here over which one sounded better.
Old 24th July 2014
  #600
Cool test! I saw the answer before did the poll so I'm perhaps a bit biased.

Initially listening thru the speakers on my macbook pro I thought the Diva sounded best and I had no problems to hear a difference between the two, the attack on the brass lead was the biggest difference to me. Diva has more resonance?. Listening carefully on headphones later I think OB8 sounds best and noticed a difference on the PWM lead. I have trouble hearing any difference on the bass sound. and when the sample and drums gets introduced its impossible for me to tell which is which.
Topic:
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump