Reverb Comparisons...?

[Thomas_J]

I was reading the \"Jeremy Soule on Far micing\" topic and noticed many people who wanted comparisons of the various reverb boxes;

Perhaps what we need now is a .wav file of a short dry orchestral snippet (any early Star Wars recording springs to mind as a useful one, even though it would be better with a midi-mockup of course) that people can download and apply their reverb to (at a fixed wet/dry ratio). This could make for a useful comparison of reverb units. (Kinda like that piano sample comparison)

Maybe someone could devote a little webspace to host the clips in mp3?

The snippet wouldn\'t have to be that long. 15-20 seconds or so perhaps?

My guess is that the differences in sound wouldn\'t be that noticable, and to an untrained ear maybe not even distinguishable.

I can apply the tc3000, (maybe lex 300) tc reverb software plugin, waves reverb, the trusty Zoom 1202 and a few others I think I can dig up.

We could make a complete comparison! It would be great in deciding which reverb to go with to best suit your \"style\". It would be a comparison for orchestral stuff of course.

How do you like this idea?

Thomas

[rediesisminore]

Hi Thomas,

I\'ve also been looking for a good reverb unit for a while (I would prefer an external box - I\'m still using a Pentium 100 Mhz!!! and Emu samplers - maybe one day I will upgrade to gigastudio when it will be more flexible - that\'s why I\'m reading through this forum since beginning).

Once or twice I\'ve started a thread about reverb and last time I was pointed to an interesting link on Harmony Central. I think many people would like to give a try to \"Moster Rack\" at www.harmonycentral.com. (\"http://www.harmonycentral.com.\") There you can basically apply any preset of a PCM91 or TC M3000 to a file of your choice.

I\'ve found that for my taste TC unit is better for orchestral and classical acoustic instruments (I\'ve borrowed one for one day and I liked it a lot). Also, here in Geneva, pros are saying that for classical recordings some guys just throw away 300Ls to use the M3000\'s more \"natural reverb\".

I started some years ago with AKAI samplers and Lexicon Alex (the best I could afford). Then I was going to switch to an MPX1, but having listened to it I didn\'t like the way it fitted to my sounds. Then I was looking for a PCM91 and that\'s when they pointed me to the M3000.

Yet, I\'m still curious to listen to a higher end Lex like 300-960, and I wonder if they\'ve all the same reverb algorythm and just different power...

I second your idea and I hope to listen to some comparison between the most powerful units and softwares (even if - due to the amount of system crashes I have to stand daily - I would always opt for a dedicated solution!).

cheers,
Roberto

P.S. I wonder if and what was the reverb used for the Artificial Intelligence score (which I like so much!). That\'s the kind of space I\'m looking for...!

[Dis]

Hehe, I can apply Lexicon MPX-100.

[Chadwick]

The guy I work with got rid of his Lexicon 300 to buy a TC M3000. He loves the \'transparency\' of the TC, and the 300 never really sounded as good as a 480L to him.

Mind you my friend who bought the 300 just raves about it all the time and feels he got a real bargain.

A horse for a course, of course, of course, and no one can talk to a horse of course...

mrE - hmmm - a mystery...

[PolarBear]

That\'ll be intersting...

We need a source file being uploaded. Who can contribute this? Thomas? Have that Star Wars file?

Or anybody else a reliable mockup? King? You\'re needed cause you own (almost) everything concerning reverb IMHO http://www.northernsounds.com/ubb/NonCGI/images/icons/wink.gif

Jump in!

Regards. http://www.northernsounds.com/ubb/NonCGI/images/icons/wink.gif
Hansi

Edit: sorry can\'t help out with webspace http://www.northernsounds.com/ubb/NonCGI/images/icons/frown.gif

[This message has been edited by PolarBear (edited 04-13-2002).]

[Franky]

It aint much but i got a Lexicon MPX 500 i can throw in if needed http://www.northernsounds.com/ubb/NonCGI/images/icons/smile.gif
It\'s great for a monitoring reverb for singers. It has analog and digital outs that you can assign which mean i can send the analog outs with effects (to monitors) and send the digital out without effects to multitrack.

Franky.

[dnortana]

I could contribute results processed by AltiVerb.

Trond

[Thomas_J]

Excellent! I\'m glad many people are willing here! We need more units though! Once I get about 10-15 different units from different guys I will post a short dry clip here on NS and you can mail your processed clip back to me. I will try to make room for them all on my web space.

As for the music clip, any preferences? I\'m thinking film music here of course.


Thomas

[PolarBear]

see below...

[This message has been edited by PolarBear (edited 04-13-2002).]

[PolarBear]

We should also generalize the format... do we want this mp3? Or pure wav? I think it is enough if the results are mp3 192kbit encoded, but the source I think needs to be wav if there is no mockup.

Also we should set a place not to get results that face to be Lincoln Center and others to be in a shower... Am I right?

[zquarles]

I have NFX, acoustic mirror, an old ART effects processor, and onboard effects of a Fostex VM08 mixing board...

[Nick Phoenix]

I think the Quantec Yardstick is the best value for the money. $3000 retail, hard to program, but sounds great. Aes in and out only. I don\'t have one, but maybe you can find someone.
I could process the file with the some of the Sony DRE-S777 sampled algorithms.

[This message has been edited by Nick Phoenix (edited 04-13-2002).]

[ChrisAxia]

Hi Guys,

I don\'t have any of the real high-end stuff but will gladly try with the Realverb in my UAD-1 card. I\'ve also got a TC M-one, and lexicon MPX1 & LXP15.

Chris

[Thomas_J]

Great! now we are getting somewhere. Glad to have you in on this Nick. Actually I do know of a guy with that Quantec Stick and he keeps raving about it.

Now I need to find a really good clip. I\'d love to hear suggestions. I just went through some score tracks and even though the early star wars recordings are fairly dry, the sound quality is so horrible it wouldn\'t do any reverb unit justice.

I think the best thing here would be to go with a simple mockup of any kind. Perhaps a dry GoS test or a VoTA/QLB/Rare Instruments snippet from your library Nick? (just suggesting stuff here)


Thomas

[Parker]

Great idea! I will gladly be able to do Altiverb and the Kurzweil KSP8 if needed. I would suggest that the orignal clip come at 24bit (.wav or .aiff).

Oh also, I hope that the source material has some percussion hits with enough space around them to actually hear the reverb tails! Possibly some timpani shots at the end? http://www.northernsounds.com/ubb/NonCGI/images/icons/smile.gif

Parker

[tob]

ive got a dusty lexicon 480 in my studio so ill process the file with that.

[rediesisminore]

I think the choice of the file(s) to process should fall both on something using samples AND recording of some real instrument being played, thus the reverb itself would be really exposed, as the quality of samples does matter alot.

For example, I own the AKAI version of Advanced Orchestra and I found the reverb (lexicon and TC) didn\'t sound as good as on other samples or on live recordings just because of the poor musicality of part of the samples themselves.

Also, I suggest that the file(s) to be processed contain sections and/or solo instruments but not an entire orchestra. This for the simple reason that you would normally apply a different amount of reverb (different sends) on strings (less) and brass or percussion sections (more), because of distance from listener. Unless the samples are recorded in their placements as Miroslav Vitous. In such case I think the reverb amount should be the same everywhere as distance information is contained in the samples - does it make sense???!!!

Anyway, try using a recording of a good grand piano, or classical nylon guitar, or solo violin or oboe or flute....! http://www.northernsounds.com/ubb/NonCGI/images/icons/smile.gif then we should have a clear image of what the reverb unit can do!

cheers,
Roberto

[PolarBear]

Would be nice to hear some solo vocals processing all the verbs IMHO...

[ChrisAxia]

Hey Parker,

I have heard little about the KSP8 with the exception of some guys on the EQ forum raving about it. I love my K2000 so I expect the reverb unit to be good. In fact, I think one guy said he stopped using his Lexicons once he got the Kurzweil. Is it that good?

Tob,

You\'ve got a Lexicon 480 collecting dust! I\'ll be happy to use it for you if you no longer need it!!

Chris

[tob]

hehe.. the reason to why I dont use it is because Im lazy. Its easyer to pick up old sessions when you dont have to remember the settings on all outboard gear. Thats why I use Reverb One in protools instead of Lexicon 480 and thats why I use Halion instead of Gigastudio ;P

/Lazy by nature, hehe

[Markku Kanerva]

I could add rev.32 from cubase 32 and TC native verb,but are these too common for you guys,since there seem to be quite a few cubase users out there?

[Parker]

Chris, in all honesty I have yet to really push the KSP8 because it is so new. From what I have heard so far (acoustic guitar, nylon guitar and vocals), it is much more to my liking on acoustic material than the pcm91 or Realverb Plug and am glad I bought it! Altiverb is another animal all together, so I cannot compare them without hurting my inner child. In about a month I *will* be on intimate terms with the KSP8.

Parker

PS Concerning the reverb comparison, maybe all reverbs should be set at 50/50 wet/dry, along with similar room and decay settings. Even though the setting may be undesirable, this may be the best way to consistently compare 10+ different reverbs. Or are we simply trying to show off how good the various reverbs *can* sound and compare from there?

[PolarBear]

<BLOCKQUOTE><font size=\"1\" face=\"Verdana, Arial\">quote:</font><HR>Originally posted by Parker:
PS Concerning the reverb comparison, maybe all reverbs should be set at 50/50 wet/dry, along with similar room and decay settings. Even though the setting may be undesirable, this may be the best way to consistently compare 10+ different reverbs. Or are we simply trying to show off how good the various reverbs *can* sound and compare from there?<HR></BLOCKQUOTE>

Perhaps both is needed... the 50/50 wet/dry settings as well as the show off... http://www.northernsounds.com/ubb/NonCGI/images/icons/smile.gif If somebody has the space to post it... Perhaps I can offer something in a few month...

Hansi

[Chadwick]

I can probably offer a couple of TC verbs, a Roland, a Yamaha Rev 5, and some Creamware stuff.

[dnortana]

Regarding how to go about this, it seems to me that there would be a real benefit to printing and returning only the 100% wet reverb track. That way, settings anomalies won\'t become a factor, and each of us can listen to the reverb alone, if we choose. Or, we can mix them together ourselves.

Trond

[B]

Someone was asking earlier for percussion in the track to hear the tails.
Would it be a better idea to use an impulse to hear the tail and the music track for the overall effect?
Although an impulse is by no means an ideal way to evaluate a reverb, it does allow you to hear any ringing or other artifacts in the tail very clearly.

brian

[Thomas_J]

I\'m sorry I\'ve been busy. Trying to find a dry clip that would suit this comparison the best. I suggest a clip of about 40 seconds of length devided into 3 parts:

(all dry clips)
1) full orchestral mix
2) percussion mix
3) solo instrument/vocal mix

I\'ll get back to you when I find 3 suitable clips.

Clips will be encoded as 192kbps mp3s.
I don\'t have enough server space otherwise.

(this would be the first really big \"out-of-the-box\" reverb unit comparison ever! I\'m excited, hope you are too.)

Thomas

[Robert Kral]

One of my huge wish list comparisons for a long time now has been to compare the PCM 90 \"Gothic Hall\" to the Lexicon 300, first preset, and the Lexicon 480, and Altiverb.

Anyone have a Lexi 300??

Another tricky, VERY tricky thing here is going to be to determine amount of mix level/dry level etc. You can easily write off a unit\'s preset as not what you want, but a minor tweek can make all the difference.

[This message has been edited by Robert Kral (edited 04-14-2002).]

[Parker]

Hey Thomas,

Are you saying that the original clips that we all get will be mp3? If so, I think this might compromise the testing. An uncompressed 24bit file has data that even though our ears may not hear, the reverbs will be able to take advantage of during processing. I am particularly refering to the vast dynamic improvement (the outer bits) of 24bit over mp3. By giving the reverbs a high quality source they can produce high quality results, the kind that we as humans will not be able to find error in thus avoiding invalid results due to lo-fi testing elements. I hope this makes sense. Alittle bit of science isnt a bad thing, especially when it comes to testing. http://www.northernsounds.com/ubb/NonCGI/images/icons/smile.gif However, I dont see any problem with the final results being posted as high quality mp3.

To compensate for the size of 24bit audio, I think the source clip can be shorter than 40sec and still be able to get the job done. How big is a 30 second 24/44 clip anyways? And do you know yet what the posting size limit is?

Parker

[Parker]

Hey Thomas,

Are you saying that the original clips that we all get will be mp3? If so, I think this might compromise the testing. An uncompressed 24bit file has data that even though our ears may not hear, the reverbs will be able to take advantage of during processing. I am particularly refering to the vast dynamic improvement (the outer bits) of 24bit over mp3. By giving the reverbs a high quality source they can produce high quality results, the kind that we as humans will not be able to find error in thus avoiding invalid results due to lo-fi testing elements. I hope this makes sense. Alittle bit of science isnt a bad thing, especially when it comes to testing. http://www.northernsounds.com/ubb/NonCGI/images/icons/smile.gif However, I dont see any problem with the final results being posted as high quality mp3.

To compensate for the size of 24bit audio, I think the source clip can be shorter than 40sec and still be able to get the job done. How big is a 30 second 24/44 clip anyways? And do you know yet what the posting size limit is?

Parker

[Parker]

Oops sorry about the double poster. Guess it is safe to say that I am pretty excited about these results as well http://www.northernsounds.com/ubb/NonCGI/images/icons/smile.gif

[PeterRoos]

Hi guys,

If it will be only one 40 second 24 bit file I think I can host it somewhere for a few months.

I just ran a check, such a file seems to be approx. 10 Mb.

Thomas, just let me know.

Cheers,

Peter

[Chadwick]

Isn\'t Thomas talking about the processed versions being mp3s, not the original file?

[PeterRoos]

<BLOCKQUOTE><font size=\"1\" face=\"Verdana, Arial\">quote:</font><HR>Originally posted by Chadwick:
Isn\'t Thomas talking about the processed versions being mp3s, not the original file?<HR></BLOCKQUOTE>


Yeah, but I guess mainly because Thomas is not able to host a big, uncompressed file.

There were some requests for starting with an uncompressed Wave file, which I also would prefer. We can put the original on some web space of mine and put the resulting mp3 files elsewhere.

16 bit may actually be more practical, as not all may be able to open and process 24 bit files. Also part of the test will probably be an excerpt from a 16-bit CD.

I do have (some) bandwidth limits though.
We must therefore take care to post only the page url here, and not the file url. That way, I can change to one another web space if it generates too much trafic.

Cheers,

Peter

[RobertKooijman]

Brian and others,

You can find mpulse responses of various reverb units including the MPX500, M3000 and Quantum Yardstick at:
http://www.echochamber.ch (\"http://www.echochamber.ch\")

Check out \"Echochamber\'s impulse responses\".

Cheers, Robert.

[dnortana]

<BLOCKQUOTE><font size=\"1\" face=\"Verdana, Arial\">quote:</font><HR>Originally posted by Robert Kral:
......................
Another tricky, VERY tricky thing here is going to be to determine amount of mix level/dry level etc. You can easily write off a unit\'s preset as not what you want, but a minor tweek can make all the difference............... <HR></BLOCKQUOTE>

Robert,
This is precisely why I suggest that the processed files be returned as 100% wet, i.e. the reverb alone. Then each interested party can download and do their own mixing and testing. Also, this obviously lets one hear the reverb alone ...
Trond

[Craig_L]

Any more development on the samples for the reverb test?