Vertigo vs Harmor vs Alchemy

[george]

This is a simple audio re-synthesis engine comparison between discoDSP Vertigo, Image Line Harmor and Camel Audio Alchemy from a sample taken from Alchemy Data library (Alchemy/Samples/Factory/Bass/EBass-C1.wav) and imported into every synth without any additional effects.

Uncompressed WAV file can be downloaded at http://bit.ly/n3p05t

Harmor default sound output was too loud and had to be lowered. Alchemy notes had to be lowered two octaves to get the same pitch. Volume between all the synths were tried to be balanced in the most equal way.

Audio is a bit out of sync due the screen recording software but you get the idea.



Cheers,
George.

[ENV1]

1: Vertigo

2: Alchemy

3: Harmor


(IMO of course.)

[Nielzie]

Interesting, The resynthesis engine of Vertigo is probably (by far) the oldest one of those three, but it still seems to be holding up quite well against those other two additive monsters

[Neon Breath]

Very interesting comparison test george. I was quite curious to hear Harmor resynthesis in action, since I can't try the demo yet (mac here). To me, with this single test, it doesn't sound as superior or better than what people said in this thread: http://www.kvraudio.com/forum/viewtopic ... &start=180

By reading people's reviews on Harmor resynthesis engine it was like if Harmor is totally destroying Alchemy's resynthesis all the way, which doesn't sound like it in this test at least. Although it's only a single test with one sound, would be curious to hear more comparisons with various types of sounds.

Similar to ENV1, to my ears its:

1 & 2: Vertigo and Alchemy, very close to each other imo.

3: Harmor

Thanks for the test!

[george]

I will try to make more as soon as I can. Honestly, I have found Harmor far superior to Vertigo and Alchemy in musical loops re-synthesis like drum loops at the moment. But I guess everything is not loops for people who want musical results.

[himalaya]

A test like this is more informative if the source file is also played, preferably as the first example, how else can we compare the resynthesis process?

[Shifrin]

I was under the impression Harmor defaults to a lower quality setting for resynthesis, which can be changed on the 'advanced' tab or some such.. Any details as to which settings were used in this comparison?

Shif.

[ENV1]

I have listened to this via big 4 way speakers on an old 70's Sharp Darlington Precision Amp and Harmor sounded like my Titan Tweeters had turned to mush.

Best overall clarity definitely went to Vertigo, Alchemy sounded kinda 'in-between'.

Would be interested in more comparisons too...

[george]

A test like this is more informative if the source file is also played, preferably as the first example, how else can we compare the resynthesis process?

The sample is from Alchemy Player library and I think I'm not allowed to redistribute it. It is on the Alchemy Demo.

[george]

I was under the impression Harmor defaults to a lower quality setting for resynthesis, which can be changed on the 'advanced' tab or some such.. Any details as to which settings were used in this comparison?

Shif.

Default settings for Harmor and Alchemy. I don't know how to change any advanced settings on Harmor, but AFAIK the re-synthesized wave was using 256 partials. It was displayed on About Box.

[himalaya]

1 & 2: Vertigo and Alchemy, very close to each other imo.

Verigo has strong flutter artifacts, and Alchemy has a constant noise, reminiscent of distortion noise. They are quite different, I'd say.

Best overall clarity definitely went to Vertigo

Listen again...Veritgo's artifacts are the most intrusive as the noise is modulated. Also hear the first bass note, it's very undefined and floppy, whereas Alchemy and Harmor have much stronger bass notes.

[ENV1]

Listen again...

Done so, this time i used the WAV.

Opinion stands, Vertigo sounds subjectively best, Harmor sounds dull and muffled in comparison and Alchemy sounds like something in between.


This is without doing any material analysis, only going by ear.

All i did was checking the spectrum and Harmor does in deed not produce any significant signal over 2k, which confirms my subjective impression. (Absence of the 'sparkly' parts. For lack of a better term, of course i realize that it is a bass sound and not expected to be overly 'sparkly'.)


Sorry.

[Dionysos]

Absence of the 'sparkly' parts.

Harmor has a Denoising feature for resynthesis that's enabled by default & may be causing this, see the Advanced tab.

[ENV1]

Absence of the 'sparkly' parts.

Harmor has a Denoising feature for resynthesis that's enabled by default & may be causing this, see the Advanced tab.

I see.

George?

[himalaya]

Absence of the 'sparkly' parts. For lack of a better term, of course i realize that it is a bass sound and not expected to be overly 'sparkly'

It's not about 'sparkle' as we can't compare to the original so we don't know how bright or not this bass ought to sound. Never the less, Vertigo's lowest bass note has been reproduced without the same body and weight of the other two. Hence, it's not good for bass resynthesis, at least judging by this example, which I wouldn't by the way.


How about transients?

Vertigo's attack is strong but a bit too muffled.

Harmor has killed the attack completely. There is a faint but sharp noise 'click' but the attack is the softest of the three examples. It does not sound like a bass guitar 'snap'.

Alchemy has a strong, snappy attack, similar to Vertigo's. A faint noisy click is also preset but it's softer than in Harmor's example (I m taking into account the overall noise present in all examples).