Challenge to VST reverb developers
-
- KVRAF
- 2070 posts since 2 Apr, 2004
I read somewhere that hardware-reverbs use very complex algorhythms that are extensively researched by those companies who produce them. They've invested a lot of research and money in developings these algo's. That seems to be one of the reasons why these companies are so reluctant to use the algo's in a VST format cause then they would be easy to copy by other developers. So they stick to the hardware or 'semi-hardware' stuff.
-
- Banned
- 6127 posts since 1 Apr, 2004 from Et in Arcadia Ego
Oh, I think something can be done to remedy that..dystonia_ek wrote:UR and Spacemaster are about the best sounding software reverbs I've heard - only thing lacking is low end, to a degree.
-
- KVRAF
- 2818 posts since 30 Aug, 2001 from where dinosaurs are still alive
my gawd, is this true? I thought that everything in software was better than hardware! Moogs, Arps, AmpSims, Guitars, Reverbs. what you're saying is terrible.greendoor wrote:Why - when so many reverb plugins are available - do hardware reverbs still sound better? I think i've found out why, which means some plugin developer will get my respect (or money) when they finally nail it ...
My career is f**ked. Now I don't believe in those vst "toys" anymore.
where's my mayonnaise? I need to squirt some in my oral cavity.
-
- KVRAF
- 2070 posts since 2 Apr, 2004
Let me explain again:Barf wrote:I read somewhere that hardware-reverbs use very complex algorhythms that are extensively researched by those companies who produce them. They've invested a lot of research and money in developings these algo's. That seems to be one of the reasons why these companies are so reluctant to use the algo's in a VST format cause then they would be easy to copy by other developers. So they stick to the hardware or 'semi-hardware' stuff.
-Reverb is very different from most other effects. Think about it: what actually is reverb? Its the 'character' of a room or venue with its own kind of unique reflections etc. In the 'old days' studio's had an extra empty room which was used for reverb. The room was the reverb! They put a speaker at one end and a microphone at the other.
-To get this effect in a 'box' or by software u need to develop formulas which simulate the characteristics of a room. How do you determine those? Due to the complex nature of reverb u need a lot of time, money, effort and testing. Who has this? Not the average synthedit-developer etc. and thats not their fault.
-So what to do if u want to make a good reverb? IMO u either need to put a whole lot of research (=MONEY) in it or buy someone else's formulas(=MONEY)
-Large companies like Lexicon/Roland etc. already have these formulas researched etc and have been using and selling them for some time now in the form of hardware. ONCE they put these formula's in a VST-format it can be easily copied by others. This would mean an enormous loss of investment (=MONEY) for these companies, so why should they make a VST-version? Selling the formula's would mean the same, so they wont sell em or ask a HUGE amount of money, because then no-one would buy their hardware-reverbs anymore.
-
- KVRAF
- 4030 posts since 7 Sep, 2002
Development costs can be high for efficient 'synthetic' reverb algorithm development because it requires a lot of experimentation and research. Multiply several months by a good monthly salary of a DSP professional. However, in the past professionals who could do this already formed companies which are pretty successful today. Probably that's why there are talks reverb algorithm development cost a lot. This should be restated: good reverb algorithms gave a lot of profit to those first DSP professionals. Today it's different I'm sure.
-
- KVRist
- 73 posts since 26 Jul, 2004 from UK It smells funny.
These algorhythims are already in hardware devices. Roland has a room with loads of lexicons, Lexicon as a room full of Rolands products...each company takes apart, tests, steals, nicks, samples, develops each others technology....its' stealable already and has been for a long time! Each company has each others chips....though, i know what your saying - it will be easyer for smaller companys to steal....Large companies like Lexicon/Roland etc. already have these formulas researched etc and have been using and selling them for some time now in the form of hardware. ONCE they put these formula's in a VST-format it can be easily copied by others.
-
- KVRAF
- 4030 posts since 7 Sep, 2002
musicL, to my knowledge it's nearly impossible to 'steal' DSP design analyzing the output of the device alone (without schemes, source codes, etc). You may only experiment using general knowledge base and at some point of time get the same or even better performance.
-
- KVRian
- 1256 posts since 22 Aug, 2003
I don't know - I think "Fruity Reverb", "Ambience", and "GlaceVerb" are all pretty good 
reverb is reverb to most people...
I mainly posted here to comment on the money issue. Don't blame the PEOPLE, for @&#!'s sakes, for not buying something they think is priced too high. At least that's the sense I was getting from page 1 of this thread - that the reason VST reverbs weren't "good" enough was because most people were satisfied with the products they had at the prices they got them for.
Now isn't that ok? "People get upset when a VST is priced over $200.00." Probably because they can't afford it - you don't find many people willing to buy $2000.00 hardware reverbs either. And if that's the reason there aren't reverbs of the quality of the hardware ones - that people wouldn't be willing to pay the high price for it - then if you are one of the people who WOULD be willing to pay for it, why don't you just get a hardware reverb? It's bound to last longer into the future - what if the VST format dies, or something? Would you want that $1000.00 machine dying along with it?
reverb is reverb to most people...I mainly posted here to comment on the money issue. Don't blame the PEOPLE, for @&#!'s sakes, for not buying something they think is priced too high. At least that's the sense I was getting from page 1 of this thread - that the reason VST reverbs weren't "good" enough was because most people were satisfied with the products they had at the prices they got them for.
Now isn't that ok? "People get upset when a VST is priced over $200.00." Probably because they can't afford it - you don't find many people willing to buy $2000.00 hardware reverbs either. And if that's the reason there aren't reverbs of the quality of the hardware ones - that people wouldn't be willing to pay the high price for it - then if you are one of the people who WOULD be willing to pay for it, why don't you just get a hardware reverb? It's bound to last longer into the future - what if the VST format dies, or something? Would you want that $1000.00 machine dying along with it?
-
- KVRian
- 1256 posts since 22 Aug, 2003
Not that I'm unsympathetic - I just don't think there's any sense in blaming anyone for what there is and what there isn't, especially if the people you're blaming are completely unrelated.
-
- KVRist
- 301 posts since 5 Jun, 2004
Even if the VST standard dies you old host and VST plug will be perfectly usable on you old PCRellik wrote:I don't know - I think "Fruity Reverb", "Ambience", and "GlaceVerb" are all pretty good
I mainly posted here to comment on the money issue. Don't blame the PEOPLE, for @&#!'s sakes, for not buying something they think is priced too high. At least that's the sense I was getting from page 1 of this thread - that the reason VST reverbs weren't "good" enough was because most people were satisfied with the products they had at the prices they got them for.
Now isn't that ok? "People get upset when a VST is priced over $200.00." Probably because they can't afford it - you don't find many people willing to buy $2000.00 hardware reverbs either. And if that's the reason there aren't reverbs of the quality of the hardware ones - that people wouldn't be willing to pay the high price for it - then if you are one of the people who WOULD be willing to pay for it, why don't you just get a hardware reverb? It's bound to last longer into the future - what if the VST format dies, or something? Would you want that $1000.00 machine dying along with it?
. The real point IMHO should be that now that these algorithms have more than made up for any extensive initial R&D that they required they should not be priced at premium prices, and with todays CPUs should definetely not be bound to hardware.
The only reason they don't is that where they would charge £500 for a plugin, they can throw in a hardware box which will cost them £10 to produce (exaggeration obviously more like £100) and then the can sell it for £2500 because "hardware" remains more esoteric and glamorous than a cd-rom in neat packaging.
What those who argue in favour of hardware are forgetting one important thing though. Buy ONE hardware box for £2500 and you have ONE instance of the reverb. Even if the plugin costs the same, you can run as many instances as you want, which I am willing to bet is already more than one (or 4 for that matter) with current hardware, but which will rise up significantly with the new families of CPUs.
-
- KVRian
- Topic Starter
- 769 posts since 2 Apr, 2005
Waitingman - underneath your cynicismn I hope you don't really believe that hardware is necessarily better than software simply because it's hardware. There are many, many examples where VST effects are cleary better than previous hardware versions. Especially when that hardware was early digital technology (which is what we are talking about here).
I'm pointing out the current absurd situation, which is basically this: generally we love the sound of digital reverb that was created by alorithmns in software encased in hardware boxes with noisy 16 bit converters. For some reason, VST reverbs (which use 32 bits, have no problems with noisy converters, and have much greater CPU power available than was available in the 1990's) currently don't sound anywhere near as good. So we resort to sampling these hardware/software boxes and uses incredible number crunching power to approximate the sound. If we somehow just get the algorithmn right ... maybe we need more than 8 delay lines. We can afford to run several 8 line reverb plugs on a PC - I would rather run one with 16 or 32 if it sounded much better. Lexicon are clearly holding out on us, understandably. TC Powercore is our best bet at the moment, or a hardware box with spdif. But imagine if a VST developer cracked the formula ... it's got to happen.
I'm pointing out the current absurd situation, which is basically this: generally we love the sound of digital reverb that was created by alorithmns in software encased in hardware boxes with noisy 16 bit converters. For some reason, VST reverbs (which use 32 bits, have no problems with noisy converters, and have much greater CPU power available than was available in the 1990's) currently don't sound anywhere near as good. So we resort to sampling these hardware/software boxes and uses incredible number crunching power to approximate the sound. If we somehow just get the algorithmn right ... maybe we need more than 8 delay lines. We can afford to run several 8 line reverb plugs on a PC - I would rather run one with 16 or 32 if it sounded much better. Lexicon are clearly holding out on us, understandably. TC Powercore is our best bet at the moment, or a hardware box with spdif. But imagine if a VST developer cracked the formula ... it's got to happen.
-
- KVRAF
- 2818 posts since 30 Aug, 2001 from where dinosaurs are still alive
no, I often talk without minding.greendoor wrote:Waitingman - underneath your cynicismn I hope you don't really believe that hardware is necessarily better than software simply because it's hardware.
I use little or no hardware and I love my setup, so no real critique.
greendoor wrote:But imagine if a VST developer cracked the formula ... it's got to happen.
Just figure that less than 1 year ago I was here screaming because there wasn't (and still isn't, you lazy ass developers around
) a decent real time pitchshifter that doesn't turn tracks to mud.one model on all:Digitech's Whammy
The best response of'em all was from Xoxos: do it yourself. (which taken positively, looks almost reasonable
)I believe it's going to happen somewhere soon.
as the direction of the thread showed, it's all up to musicality of algos & the musicality in creating and/or copying them.
btw, I don't think convolution is just a resort for verb pervs deluded by vsts. it's a new frontier in processing through sampling.
for example, I was really skeptic about TapeBus. now, no mix is ready without it. So much that, I could hardly think of something else, when the update with more tapes came.
one curiosity, have you tried Pantheon?