|
A soundcard in its most basic terms is essentially a input/output system with an Analogue to Digital and Digital to Analogue convertor in between. Some soundcards will throw in other things to confuse people - a preamp stage for example is common in soundcards aimed towards home recordists. When you get beyond a certain price point you no longer see all in one 'soundcard' solutions - you see discrete modules - these can be used with a computer in between but arent designed exclusively for computer users. Discrete AD convertors. Discrete DA convertors. Discrete preamp stages. And you essentially have to build your soundcard out of massively overengineered, massively expensive metal boxes. In this case, you would need a separate AD convertor like a Prism AD2 to record sound via an input. You would need something like a Benchmark DAC-1 to convert the digital signal stored temporarily on your computer back into a tiny current so it can be output to a speaker amplifier. And you would probably need a preamp like a Neve 1272 before the ADC to ensure the signal is full scale at the input. The soundcards I listed are basically just convertors and clocks. Thats it. Everything else costs extra. Gulp. 
|
#
¿
Jun 29, 2007 19:20
|
|
|
|
| # ¿ May 13, 2024 22:02 |
|
|
Blackadder posted:Critique my studio... its all "prosumer" type gear. Please tell me that I have enough because I don't want to buy anymore. Well its how you use it that counts...I'm not entirely sure what kind of answer you are expecting.
|
|
#
¿
Jul 6, 2007 13:28
|
|
|
Id go with convolution reverb if your computer is up to it. You can run certain convolution processors with 0ms latency (like Voxengo Pristine Space) but the CPU hit is fairly big. If that doesn't bother you, then for 120 bucks or so you get the basic unit and get to use alot of impulses for free - these include some well recorded acoustic impulses and impulses recorded from famous digital hardware reverb processors (such as the Lexi PCM91 which does some really nice sounding plate emulations). You can't buy a PCM91 for less than about 1.5 grand so this works out to be a good deal. Convolution processors can also be used emulate many hardware reverbs in this way - all of the TC reverbs for instance and Pristine Space. Its also worth noting what type of reverb you want - there are many types - reverberation from real acoustic spaces, plate reverbs and spring reverbs. Spring reverbs would be the type of reverb you typically get on a guitar amplifier. A Spring reverb is basically a transducer and a pickup with a spring between them - the pickup detects the mechanical vibration of the spring caused by the transducer and converts it into an analogue signal. These are fairly compact which is why they are built into guitar amps. Plates are the more expensive versions of spring reverbs and use a similar principle except that this time theres a metal sheet in between the transducer and the pickup. The pickup detects the mechanical vibrations of the metal plate. These are fairly big however. You know what to expect from real acoustic spaces. Plates are fairly expensive, fake reverbs. Plate reverbs do not sound like real acoustic spaces but they are so commonly used in music production and so ubiquitous on studio recorded vocals that they have basically become desirable in and of themselves. If you are shooting for a studio sheen kind of reverb - then you want plates. All digital reverb processing is done on digital signal processors so you can stick these into a 1u rack or whatever and emulate the sound of a plate reverberator. The PCM91 is one such unit and its very well known but expensive. Eventide do alot of digital effects processors so I'm sure they do a reverb unit of some kind. If they do, its going to be expensive. I'm not a big fan of TC's reverbs but they do quite a number of digital reverb processors. Mostly though, I just use SIR 1.008 - which is a free convolution processor like Pristine Space (but not as good). I'd do all of this processing digitally and after the recording is in your DAW - I'd rather not perform the extra AD/DA conversion stage and ship a digital signal outboard and back in. But I'm fairly certain most of these units have digital I/O so it should all be good.
|
|
#
¿
Jul 13, 2007 11:19
|
|
|
RivensBitch posted:Apples to oranges really, I like convolution reverbs but to my ear they are not a "classic" reverb sound so I think they have different applications than i'm used to. Wha? Obviously the results of convolution based reverb will vary depending on how well recorded the impulse response is and the means by which said impulse is generated. But to suggest that the process is somehow not capable of 'classic' reverb sounds is misleading. A convolution reverb can really sound like any reverb - spring or plate or reverberation from real enclosed spaces - it depends on how the impulse is generated and in what acoustic space it is recorded. There are some Lexicon PCM91 reverbs on noisevault.com and those are well recorded impulses. Many of the plates sound almost exactly like their hardware counterparts in practise sans the $2000 pricetag. So go and get SIR 1.008 and these free impulses and see for yourself. But even then you don't get to the real meat of convolution processing until you are working with 4 to 8 channel convolution processors. Voxengo Pristine Space is the one I use since its not that expensive and can operate at 0ms latency. Chaining impulses allows you to decorelate the left and right channel outputs by mixing in varying amounts of several different impulses and you can pan a 100% wet signal and it wont sound weird. I am so consistantly impressed with the results that when I get home, I'd like to throw some soundclips your way to demonstrate just how awesome chain convo plugins are. This post isn't even going to take into account that you can use convolution processing for much more than reverb - you can import loops of any length (though longer = bigger CPU killer) or any sample in wav format. And you can convolve that with the incoming signal and impart certain properties of the sample onto the original sound. This makes convolution absolutely the best value for money out of any signal processor cable of generating reverb. There simply isn't any contest. You can SIR + great impulses from classic hardware like those EMTs, the System 6000 and the PCM91 for gratis. A chain convolution processor like Pristine Space will cost you about 120 bucks. The impulses are still free. Not to mention you can create your own impulses and deconvolve ready made ones. WanderingKid fucked around with this message at 11:18 on Jul 24, 2007 |
|
#
¿
Jul 24, 2007 11:07
|
|
|
Dirk Muscleton posted:I have a recording transferred from god knows how many CD-R generations. Along the way, someone had a scuffed CD or a lovely burner or maybe even both, and the result is a mono recording with constant clicks, snaps, pops, and yes, even crackles in "stereo". Ugghhhh, I have no idea why you are trying to do this and I have no idea what a 'karaoke option' is but assuming you get the exact opposite of what you want then you are on the right track - to whatever horror you are trying to create: 1) Use Karaoke option thingy on the original audiotrack (as you have done). This leaves as you said only the 'bits in stereo'. You said the original recording is mono so I have no idea how you managed to come up with a bone fide stereo mix from a mono mixdown but whatever. 2) Invert the phase of *both* channels in the 'stereo bits' audio track. Copy this to clipboard. 3) Paste this into the original (stereo?) audiotrack. There must be no horizontal offset. 4) Save the result as a stereo wav. If there was some degree of time offset between the original audio track and what you pasted into it then it wont work properly. If you do this right, everything that was left from using the 'karaoke option' will be absent from the final result. Which I'm guessing is exactly what you want.
|
|
#
¿
Aug 21, 2007 17:30
|
|
|
Dirk Muscleton posted:Garbage. I tried this in both GoldWave and Audacity (didn't have the heart to try it in WaveLab), and the clicks sounded even louder than before. However, if "invert" and "reverse" mean two entirely different things (as well they might), I am a complete and total dummy on this point. I inverted, but did not reverse (nor did I see an option to reverse - which is why I still have hope that this is just a case of two words for the same function). Invert is not the same as Reverse. When you Invert a wave, you will essentially turn it upside down vertically. So it is in opposite phase. When you reverse a sample you flip it around horizontally so it plays backwards. quote:The recordings in question are live recordings, and tracking down a fresh source is very, very tricky. This is what doesn't make sense. There is nothing in between left and right channels. A stereo recording has 2 channels - left and right. Unless you hard pan a sound left or right (such that it exists only in 1 channel) I cant see how this option can remove everything thats centred. What this process seems to do is to take a stereo pair recording. Invert either the left or right channel and sum it to mono. Then save the resulting mono wav as a stereo wav. Bear in mind that when you do that invert phase trick, it must be exactly in phase with the original signal. If its off by a degree of samples then it wont work properly. The more 'off' it is, the worse it will sound. And if you are 180 degrees off, it will actually make the stuff you want to remove double in amplitude. quote:Since a sample would probably be worth a billion words (see above), here you go (290KB mp3). The original file, of course, is nice and lossless, but I figured a .FLAC might be bordering on gratuitous. I'll check into this when I get home. But before then, could you also record a clip in the same quality as what you uploaded with the karaoke option thingy used on it? I'm sorry if you already did this but I'm at work at the moment and obviously cannot listen to music :\
|
|
#
¿
Aug 24, 2007 14:09
|
|
|
Slowfuse posted:At this stage, my prospective DAW hardware set up will be: You don't need a MIDI interface. Not unless you are running absolutely tonnes of MIDI hardware. If you are all software, you don't need a MIDI interface. If you want to know why: A single MIDI port has 16 MIDI channels which you can assign to any combination of inputs/outputs you want. So if you never use more than 16 MIDI channels you absolutely dont need a MIDI interface. MIDI Interfaces are basically racks that are ear to ear MIDI ports. MOTU does a MIDI Interface called Timepeice or something and that has 8 MIDI ports, which is 128 MIDI channels. Example. Say you have an Access Virus Ti. That synth is 16 parts multitimbral which means that it can send/receive MIDI via 16 channels simultaneously. So you can (amongst other things) play 16 different arrangements, using different presets at the same time using just this one synth. But if you use all 16 channels it tends to gently caress up and not trigger in time and someone over at the unofficial virus forum suggested it was probably a result of the low bandwidth of a single MIDI port. So the idea with MIDI interfaces is to spread high bandwidth (running tonnes of MIDI channels at the same time) over a whole bunch of ports meaning you can run them all with less chance of mistriggering notes and drop outs and other glitches you normally associate with running tonnes of MIDI gear via MIDI thru out of a single MIDI port. quote:As I understand it, latency is caused by a combination of: They use the fastest sampling rate (in a Pro Tools 192 system: 192,000hz or a DXD convertor which is twice that rate) and every realtime process is crunched by dedicated DSPs. Not CPUs or computers which share resources with other applications and processes. That largely eliminates latency introduced by plugins as well.. WanderingKid fucked around with this message at 11:35 on Aug 30, 2007 |
|
#
¿
Aug 30, 2007 10:46
|
|
|
It should still not be a problem. A MIDI data stream is pretty tiny - a constant 31.25kbps. If you are getting 20ms of latency its almost certainly an ASIO/DMA buffer thing or a sampling rate issue. Possible solutions: 1) Increase the sampling rate. This will stress out your computer more though. Doubling the samplerate from 48khz to 96khz with the same ASIO buffer size will halve your latency. 2) Decrease the size of your DMA/ASIO buffer. Will most likely be an option in Reason or your soundcard's control panel. Although having too much poo poo running on your computer can cause drop outs and other interruptions which result in jitter during playback/recording if your DMA/ASIO buffer size is too small. The only time you get MIDI 'lag' is when you are using MIDI thru to hook up like 5+ synths together. But all sorts of weird things happen when you daisy chain loads of instruments through the same port. Some of them start playing 'drunk' and not hitting notes at the right time. Its strange. WanderingKid fucked around with this message at 12:48 on Aug 30, 2007 |
|
#
¿
Aug 30, 2007 12:45
|
|
|
nrr posted:This is pretty much the sweet spot that I gather almost everyone is trying to hit. I'm also in the market for an audio interface and have figured that if I'm going to pay $2-300 for anything half decent, then why not look into spending another hundred dollars or two and landing maybe a used/floor model of the next tier of interfaces? Something that's going to be a bit beyond my needs currently, but really show it's value a year or two down the track when my recording needs grow, and I don't need to throw down another thousand dollars for an upgrade because it can still cover my needs. Beyond a certain point the difference in A/D conversion between interfaces is so small that nobody on gearslutz could tell the difference in blind tests. They've had users doing it for ages now and I can't tell the difference between an 828MKII A/D stage and a Rosetta 200 A/D stage blind. Some people claimed they can, but at least half of them got it wrong when the results were posted so there you go. The stability and ease of use of the software though is really important to me. I want to be able to take my audio interface with me and work on other people's computers and in other people's DAWs if I have to and I want it to be as simple as 'plug in and go'. I've had some horrible experiences with my mate's ESI interface which has a driver so shoddy, it isn't even detectable in my DAW of choice. Which is pretty rubbish when you want to get something done. I've never used an I88X but if its available for less than half retail price (otherwise as new) you can bet with some certainty theres a reason for it. Swivel Master recommended an RME Fireface a while back and after trying it out I went ahead and bought a Fireface 400. The software is really good and to date havent had a single issue with it. Its fairly pricey but I figure if you want to get something that will go the distance, you should save up some more and get something thats a keeper. Definitely audition one or take advantage of a home trial if you can.
|
|
#
¿
Sep 2, 2007 23:12
|
|
|
I can't tell the difference between the AD stages of the MOTU 828MKII, RME Fireface 800 and Apogee Rosetta 200 in a blind test and neither could nearly all the gearslutz members who decided to participate in those tests. After that I decided that conversion is important up to a point. Then theres so little difference that it basically comes down to whats in your head. Preamps are a different story, as is clocking so if you want to throw tonnes of cash at an audio interface, I suppose you want solid clocking, awesome spooge worthy preamps and a solid driver that gets updated for compatibility and bug squashing. The preamps are probably where most of the cash will go as you can really spend as little or as much on those gizmos as you want. And depending on their construction and method of amplification, they tend to sound noticeably different too. If you are just recording guitar and vocals then you probably don't need as much analogue I/O as the Ultralite provides. I've never used one but considered buying one (I eventually settled on a Fireface 400 and I am a happy bunny). Doing the research though uncovered a few things you might want to investigate further before you throw wads of money at it. The preamps can be pretty noisy apparantly in high humidity. Which sounds weird, almost like an individual defect until I realised several people report the same behaviour including one guy in this thread I think. Rivensbitch has long pointed out that trouble free operation of MOTU firewire interfaces requires you to run the right firewire chipset or things go south pretty fast. Possibly not an option for you if you are using a laptop. I would consider sinking a little extra into an FF400. The driver is rock solid. The preamps are kind of...meh but my expectations were sort of high (I used an Avalon U5 for about a year prior with my Delta 1010 which was fun). I dont notice any difference in AD or DA conversion between my Delta 1010 and my FF400. But then I've never done real controlled A/B tests. Either way its neglible. This might sound negative at first but the software just 'works' with everything which is great. Its well built, small and its never skipped a bit. It looks rear end ugly in publicity shots but its alot better looking when you get it. Its doesn't have a purple rack face for a start and those stupid looking plastic ears are removable. You also have a fair bit of connectivity in case you want to add a bass or stereo mic vocals and guitar at some point.
|
|
#
¿
Sep 10, 2007 19:04
|
|
|
I have 3 soundcards and I honestly cant tell the difference between any of them when mixing down recordings to the AD stage of all 3. Even when slaving the other convertors to the Fireface's clock. Whilst you do have a point, you also must not fall into the trap of believing other people's bullshit when you clearly can't hear any difference yourself. I mention the gearslutz tests (which all have disclaimers) because I couldn't tell the difference between them either. Out of interest, with a touch of post processing I also cannot tell the difference between a Moog Voyager saw wave oscillator and a Virus B saw wave oscillator. There is of course a difference in the sense that the signal from a Virus oscillator is reconstituted out of a finite number of samples whereas the signal from the Voyager oscillator is the AC signal from a variable voltage switch. But after EQ, normalisation and a touch of additive synthesis I cant tell the difference between them either. WanderingKid fucked around with this message at 19:09 on Sep 11, 2007 |
|
#
¿
Sep 11, 2007 18:58
|
|
|
Yojimb0 posted:I apologize in advance for the very long post... If you read this, you're a saint: Well you have a tonne of options. I would first recommend keeping your existing keyboard as a controller. If you need help setting up MIDI cc values you can post here and I'll see if I can work you through it. Use it to control a VST softsynth. If you are fairly new to synth programming you should start with a subtractive virtual analogue synthesizer (VA) as these are by far and away the easiest, most intuitive synths to program when you first start out. The programmability of VAs depends on alot of things - how complex it is (how many modules there are and thus how many sound shaping tools you have at your disposal), and how the UI is designed. Some have really poor UIs and its difficult to visual what you are doing to a sound. The easiest soft synth I found to program was reFX Vanguard so I recommend you download the demo and have a fiddle. I don't much like this instrument but its a really good place to start learning. The main thing to remember is that with VAs, you start with a set of carrier oscillators which generate a basic waveform - a sort of 'block' of sound with harmonic content. The ones to start with are: Sine - Carries fundamental frequency (lowest pitch reference) and nothing else. No harmonics. This waveform is commonly used in the synthesis of non harmonic instruments or atonal instruments - membranophones like kick drums. Saw/Ramp - Carries fundamental frequency and all even and odd harmonics (2nd, 3rd, 4th, 5th, 6th, 7th etc etc) thereafter in linearly descending intensity. Commonly used for harmonically rich sounds - particularly synthesized string instruments. Square - Carries fundamental frequency and all odd harmonics (3rd, 5th, 7th, 9th etc) thereafter in linearly descending intensity. Commonly used in the synthesis of wind/reed instruments. Pulse Width Modulated (PWM) Square - same as a square wave but the harmonic content varies intermittently. You could call this a rectangle wave where the width of each rectangle is variable. You can stack oscillators together to create more complicated timbres. Think of this as the means by which you scuplt the tone of your sounds, by tuning different waveforms or similar waveforms against each other. The results you get from just an oscillator module will sound very basic - you are just looking at texture and pitch, nothing else. The output of this module is then typically routed to a filter input stage. A filter has two basic controls - cutoff and resonance. A low pass filter for instance is used to reject that part of a signal which exceeds its cutoff. Basically, if you imagine the entire scope of human hearing to be a range between 20hz and 20,000hz (thus expressing the lowest sub bass all the way to inaudibly high frequency treble) then setting the cutoff to 1,000 hz will make the filter start attenuating everything above 1,000hz. The steepness of this slope is determined by the number of filter poles (typically 1 to 6, 1 being a single order filter that attenuates at a rate of 6dB/Octave 6 being a very high order filter that attenuates at a rate of 36dB/Octave). Resonance increases the amplitude of the signal at the cutoff point and can pronounce the effect of sweeping the filter or used to change the pitch reference of a sound. Experiment with these basic controls to get a feel of what it is capable of. Most VAs have multimode filters which can typically act as one of the following: 1) Low Pass filter - attenuates frequency above cutoff. i.e. it allows low frequency (bass) sound to get through but blocks (treble) high frequency sound. The threshold at which this starts to occur is the cutoff. 2) High Pass filter - attenuates frequency below cutoff. i.e. it allows high frequency sound through but blocks low frequency sound. 3) Band Pass filter - attenuates frequency either side of the cutoff. i.e. it allows a frequency bad through whilst blocking extreme high and low frequency sound. The width of the band is controlled by resonance. 4) Band Stop filter - attenuates frequency at the cutoff. i.e. works in the opposite manner to a band pass filter and blocks frequency where you set the cutoff. It lets extreme low and high frequency sound through. The amount that is let through is controlled by resonance. The filter is the primary sound sculpting tool on a synth. The term subtractive synthesis comes from the manner in which you generate a block of harmonics using basic waveforms and then 'subtract' the frequency ranges which you don't want. Think of it like carving a more interest shape out of a solid block of concrete - you chip away the material you don't want. The filter then outputs this modified signal to an LFO which stands for Low Frequency Oscillator. This can be used to 'modulate' a sound. Modulate is a fancy sound for 'vary over time'. An LFO for example can vary the pitch of a sound over time (using basic waveforms like sine waves, ramp waves, square waves etc) to simulate vibrato. Or it can vary the filter cutoff over time. I'll go into more detail with this and modulation sources and destinations after you get basic sound design. Its not necessary at this stage but becomes very important later on. The LFO output is typically routed to an amplifier stage. This allows you to vary the physical sustain/decay of a sound. Whether it has a hard attack or soft attack when you strike a key. Whether it continues to 'echo' out after you release a key. An amplifier envelope works like an LFO modulating volume except that that it does not cycle infinitely. I.e. An LFO modulating volume set to saw wave would cause the sound to fade in and out periodically. An envelope doing the same thing would do so only for a single cycle. I.e. you can make a sound fade out. An amplifier envelope lets you control how long it takes for a sound fades in (attack) or fade outs when holding depressing a key (Decay/Sustain) and how long it takes to fade out after you release a key. If a patch is difficult for you to play or feels unresponsive - consider altering its envelope attack. A filter envelope works in exactly the same except that it varies (you guessed it) the filter cutoff over time, instead of volume. This becomes very important at very short time values where it is used to synthesize 'pluck' type sounds or swelling crescendo like pads. Fiddle around with these basic controls and get a feel for how these modules interract and how they can be used to sculpt sound. They will the same on every analogue/virtual analogue synthesizer so once you become proficient at programming one of them, you can technically program all of them. You just need to get used to their user interface. More complex virtual analogues have more modules. reFX vanguard is fairly basic - 3 oscillators, 1 filter, 3 LFOs (non routable), amp/filter envelope stage and some effects. An Access Virus is fairly complicated - 4 oscillators (mix of basic waveforms and hundreds of wavetables), 2 routable filters, 3 routable LFOs, amp/filter envelope stage, modulation matrix (where pretty much every variable on the synth is a possible modulation destination) and effects. To synthesize ambient type sounds or harmonically complex sounds you will eventually need a synth with alot of routability, flexible LFOs and a modulation matrix. Its best not to tackle all that when you first start out because it will gently caress with your mind but later on it will probably be something you actively look for in a synth. Hope this helped.
|
|
#
¿
Sep 27, 2007 01:33
|
|
|
hmmxkrazee posted:I have a XLR -> 1/8" adapter thingy that I use to plug my karaoke mic into my mic slot on my computer. Will this work with a guitar as well? And I'm still open to suggestions for guitar recording software that comes with effects. A guitar takes neither an XLR or 1/8" TS connection. It takes a 1/4" TS jack. So the answer would be no? What exactly are you trying to do?
|
|
#
¿
Oct 10, 2007 01:16
|
|
|
If you are pushing a load of system resources, freeze as many tracks you can and drive the samplerate as high as it will go. Doubling your sampling rate will halve your latency. Then lower your ASIO buffer size as far as it will go before you start to get pops/clicks and stuff.
|
|
#
¿
Oct 11, 2007 00:52
|
|
|
You are probably better off rewiring FL Studio to Reason and using a program like Native Instruments GuitarRig from within FL Studio. You do your live monitoring on guitar through FL Studio. You record it in FL Studio. You do your tracking, arranging, mixing, whatever in Reason. Try out the demo - you should be able to do audio recording in it thus saving you the cost of the software. You probably won't be able to skimp on GuitarRig. There may be a demo available from NI's website that you can toya around with and see how it goes. For tube sounds. Go get the freeware plugin TLs Saturated Driver. I use it all the time and it rocks. For something a little more interesting, check out Tri Dirt which is also freeware and is a 3 band EQ with saturation stages. For cab simulations go get Voxengo Boogex (also freeware).
|
|
#
¿
Oct 11, 2007 11:15
|
|
|
Are they even working properly in OSX and Windows XP yet? 
|
|
#
¿
Oct 31, 2007 22:55
|
|
|
almightyjimbob posted:Why is this? I'm trying to wrap my head around this one... Step 1 - .wav file is loaded into RAM. Step 2 - Digital to Analogue Convertor (DAC) retrieves .wav file from memory via Direct Memory Access (DMA). Step 3 - DMA transfer streams regular blocks of memory to the DAC. The bigger the block of memory, the more lag there is. This is your DMA buffer and the size of the block can be changed. Bigger blocks = more lag. Smaller blocks = less lag. Step 4 - Convertor takes first block of memory and converts it into an electrical AC signal. It does so at its sampling rate, for the sake of arguement lets say 48khz. That is, it is converting 48,000 digital samples per second. When it finishes the first block it moves onto the next block of memory and does the same thing. There must not be any interruption in this process otherwise you will get stutters/pops/clicks and other artifacts. - If you double the samplerate of the convertor it will finish converting the same sized block of memory in half the time - it is working twice as fast and can move onto the next block sooner. A convertor working at 96khz is converting twice as many digital samples per second as a convertor operating at 48khz. Even though 96khz wavs are twice as large as 48khz wavs (thus taking up double the memory) the DMA buffer is purged and refilled in half the time anyway so the latency will still be half provided the DMA buffer size is the same. If you add analogue outboard and/or plugins with fixed latency, then doubling the samplerate won't necessarily get you half the latency but most of the time it will almost certainly be less. If you are all digital with no need for delay compensation then you should get half the latency by doubling any given samplerate. Beyond a certain point (above 96khz), the only real point to using higher samplerates is to get ridiculously low latencies. Pro Tools 192 systems and DXD convertors (like the prism ADA-8XR) have stupid low latencies thanks in large part to their stupid high samplerates. By the way, if you are having trouble visualising this process - a CD burner works in essentially the same way when writing memory directly to a blank disc in realtime. WanderingKid fucked around with this message at 23:27 on Nov 3, 2007 |
|
#
¿
Nov 3, 2007 23:17
|
|
|
almightyjimbob posted:I think I sort of understood that. What impact does 16-bit vs. 24-bit have on latency, if any? It doesn't have any. The only thing that really affects buffer latency is the size of the DMA/ASIO buffer and the samplerate of the convertor. Plugins that add latency will also have an effect, and long chains of outboard will add to it, but not that much. working at higher bitdepths and samplerates is more taxing on your computer's CPU and memory resources though. Applying realtime processing on 96khz recordings means of course that post processing has to be performed on twice as many samples as a 48khz recording so that will bitch your CPU. Working at higher samplerates consumes alot more physical memory because the physical file sizes you are working with are much larger. This sort of means that you may have to buffer more audio on its way to the convertor to avoid getting pops and clicks caused by your PC stressing out - that or upgrade your PC. On pro tools HD systems where you have dedicated resources to handle all of that (DSP farms) its not an issue and you can run at crazy low latencies without any problems.
|
|
#
¿
Nov 6, 2007 06:10
|
|
|
RivensBitch posted:Vista is actually the "best windows yet" and I say this as someone who has always waited to upgrade my OS (I think I was running 98 up to 2 years after XP was released). Once the hardware and software catches up I think everyone who has held back will be pleasantly surprised. However some people might not be happy with how long that catching up takes, I spoke with a contact at Mackie that I knew through my old job and his estimate for vista drivers was "weeks, maybe months... I know they're writing one, at least I'm pretty sure they are". I went home over christmas and my parent's computer has Vista on it. Without my rig I basically just used software that I burned onto CD when I left for home and everything worked great. Which I didn't expect. Pretty much everything in Vista except for UAC is better than XP and after initial reservations I want to upgrade. When I went back to XP a couple of days ago it just wasn't the same. I want my weather monitor, sidebar, aero glass etc. back again.
|
|
#
¿
Jan 8, 2008 14:09
|
|
|
Aenovae posted:This is a great thread. 1/8" TS unbalanced is the same as 1/4" TS balanced with the only difference being the shape of the plugs. You can run an 1/8" to 1/4" TS jack to jack cable and it would be fine. You would use that if you wanted to connect 2 devices that have 1/4" I/Os and 1/8" I/Os. TS jack to jack cables have 2 conductors (signal carrying line and ground). 1/4" TRS balanced is different because there are 3 conductors instead of 2 - signal carrying line, signal carrying line (antiphase) and ground. This type of cable is used by systems that electronically balance for noise rejection. All these different shaped plugs are just that - different shaped plugs. Square box. Square hole. I maintain that it doesn't matter what you use so long as you are comfortable using it and you enjoy working with the tools you have. At the same time you don't want to have an attitude like 'gently caress it, it'll do'. I think its a question of attitude more than anything else. If you are motivated and you think everything through and learn to improvise around the limitations of your gear you can make great music using cheap hardware and software. On the flipside you can have the most expensive spooge worthy gear in the world and if its obvious you don't really care about the end product then its going to show in the music you make. He is right about one thing though - if you do get a song released, most people will listen to it on crappy speakers and/or headphones. Even if you have expensive monitors you will at some point have to give your songs a listen on cheap as poo poo speakers to see how the mix translates. Also. NS10. I rest my case. 
WanderingKid fucked around with this message at 14:26 on Jan 15, 2008 |
|
#
¿
Jan 15, 2008 14:21
|
|
|
One of my housemates has a Toneport. I think its a complete peice of junk and it sounds like complete rear end. I have never heard his Les Paul sound so bad. Take a look at the hardware and its basically a DI box and a DSP that crunches some very bad sounding distortions/amp models/phase modulation effects. I cannot for the life of me understand why you would ever want to buy a Toneport when you could buy a mono DI for like a tenner and then use freebie software effects that sound better (like Voxengo Boogex for amp modelling, TLs Saturated Driver for overdrive, SIR for reverb etc). drat if all you want is cab simulations you can use any convolver (this includes SIR 1.008, SIR 2.0, Voxengo Pristine Space and Waves IR which are all most commonly used for reverb). The results would depend absolutely on how well the impulses are generated and recorded but you have a ridiculous amount of choice and you can make your own. Boogex just happens to be a freebie with a bunch of impulses played through various popular guitar amps and recorded using various popular mics. GrAviTy84 posted:^^^seconding the post two posts above mine, find a good rep and they'll treat you well. Better in what way? The important thing is you listen to them yourself first before you buy them. Other than that all monitors will sound different and more expensive doesn't necessarily correlate to better. It doesnt really matter what you use so long as you enjoy using them and get used to the way they sound. I use a pair of Sennheiser CX300 earbuds for monitoring and they cost me a fat 40 bucks. I also have a pair of Dynaudio BM5as either side of my flat screen which I barely ever use and only to secondary ref. One of the things that turned me off them was because I move house pretty much every year and these little crates sound dramatically different in every room I play them in. I'm tired of it and its great I can carry around a pair of buds and they sound the same where ever I move. The important thing is you like using your monitors day to day and actually use them day to day. Eventually you will just get used to the way they sound, begin to recognise where they are deficient and compensate accordingly in the mix. quote:Don't neglect cable pricing, they are just as expensive as everything else and definitely don't fall under the "little poo poo" category, even with bargain bin stuff, there are a LOT of cables to buy. I would encourage people to buy the cheapest cable they can get their penny pinching hands on. The only consideration you ever need to worry about are mechanical so if you are likely to flex your cables alot around the termination points you probably want one that is sturdy enough to handle that sort of flexing repeatedly. All electrical considerations in speaker cable are in a round about way meaningless when considering the length and thickness of the conductors used. You could use a coathanger and it would work just as well as any other cable on the market. WanderingKid fucked around with this message at 05:54 on Mar 27, 2008 |
|
#
¿
Mar 27, 2008 05:35
|
|
|
iw posted:I find that having longer releases makes the compression sound less "compressed" -- it bothers my ears more if all the releases are that quick. I generally try to reach a compromise: set the release high enough so that it doesn't sound harsh, and set it low enough to avoid pumping and/or breathing. I find that vocals are the most forgiving of long release times. Release 'holds' the gain reduction for longer after the input signal dives below threshold. I.e. it just takes longer for the compressor to realise when it should stop compressing. What longer release times can do (when coupled with short attack envelope times) is to reduce very sudden changes in volume caused by the compressor kicking in suddenly and applying massive gain reduction and then suddenly releasing. If you want less compression you set the threshold higher and turn the ratio down. If you set the release time long enough and the attack time short enough (I've seen some compressors that can hold release for up to 4 seconds) then you can get to a stage where it is taking so long for the compressor to release that it detects another peak over threshold and starts compressing again. i.e. you can get sustained, permanent gain reduction. So in a roundabout way you can get to the point where the simplest solution is to just turn down the channel volume instead. Its hard to tell most of the time (except with really extreme release times like 4 seconds) because how you set it up depends totally on the vocal and you will need to work out a different threshold, ratio and release time for every singer. Fortunately its not hard if you have a wave editor and can see exactly how much gain reduction you need by looking at the waveform and reading off the amplitude on the y axis. You only need to use a de-esser if you have a problem frequency range that is too pronounced compared to everything else. The classic method of de-essing is to route your input signal (vocal track) to a bandpass filter and sweep the cutoff until you can you have isolated the most sshhhhhh or ssssss but attenuated everything else. The filter will attenuate everything either side of the cutoff. It is usually harsh consonant ess sounds with the lowest part of it being anywhere from 4khz up to 16khz and beyond). Once you have isolated as much sibilance as you can, unplug the filter, route the input to a compressor and plug the filter into the compressor's sidechain with the same cutoff and Q. Voila - you now have a frequency dependant compressor designed to compress only essing sounds. Set the compressor up as desired with lower thresholds and higher ratios roughly corresponding to less esss. In my humble experience the most important things about getting a vocal right is how good the singer's performance is, the sound of the room/booth in which the performance was recorded and the quality of the recording (as little ambient noise as possible, properly preamped mic, recording as hot as you can without necessarily resorting to soft clipping etc.). If you can nail those three to your satisfaction you don't have to do an awful lot of work with post processing. At the very least its alot easier to achieve what you want. I would say thats a good 80% of the work right there so if you've got these three steps in the can then you are already most of the way there and the rest is all technical faff trying to make it homogenise well with the guitars and drums and whatever else you want to play at the same time. WanderingKid fucked around with this message at 19:42 on May 6, 2008 |
|
#
¿
May 6, 2008 19:04
|
|
|
One of my housemates thought he broke his SM58. He started taking it apart on my desk just now. He took the capsule off (theres a dent in it and the dented area is a little bit rusty). Someone has been eating it no doubt). Smacking the diaghragm didn't make any noise. Little bit of static. We unscrewed the body and the two wires were intact. He gives it a shake and mutters something into it (wires dangling and the grille flailing about. I would have been mildly alarmed if it was any other mic but this one is the popeye of microphones). Theres a bit of static but thats all. I yank out the XLR cable and try to push it in hard again and theres a wee bit of life in it. We do this a couple of times and he smacks it on my desk which annoys me because it hurts my desk more than the mic. It didn't really hurt the mic at all. He eventually isolates the problem - the male XLR pins at the base of the mic were pushed in (I have no idea how this happened) and there wasn't a good enough contact to pass a signal down the cable. He pulls them out a bit with a pair of tweezers and toilet roll from the bathroom and I stick a small wad of blutac in it. I don't know why I'm surprised because I've seen youtube videos of people running over an SM58 in a 4x4. But it works.
|
|
#
¿
May 28, 2008 22:52
|
|
|
TC Konnekt uses that DICE-II chip that has been a huge cause of grief for loads of Konnekt users. I was about to get a Konnekt 48 but was discouraged in the end because of the unresolved driver and DICE II problems that Konnekt 24D users report to this day. I heard their PC drivers are in better shape but there are folks with Macbooks that still get hell from these cards. I wasn't keen on the Saffire (the white one not the black one) because when I tried it out on launch day at the Temple Bar music centre it had a number of annoying workflow routines (possibly bugs?) which did things like default all the mixer faders to 0dB when you load a saved mixer state. Made worse by the fact that the outputs are really amazingly loud. Also loading the onboard DSP effects are a pain in the rear end to load if you want them to run off the DSP. The hardware was great but the software annoyed me to no end. I couldn't live with it day to day. I don't know about the other two but if you are going to get a soundcard one of the big things is the software and the drivers. I've just about had enough of having to cart my FF400 around with me whenever I want to mix someplace other than my own room because everyone I know seems to have soundcards with drivers that suck or sometimes don't even get detected by my DAW. I hear great things about Echo and over at Gearslutz a recent poll showed that 0% of all the Echo respondents had technical problems with their firewire cards. RME came in second and I can vouch for that because they seem to get driver updates on a stupidly regular basis and the thing has never crashed on me and has worked in every rig I have cared to throw it into. The software is a joy to use. I have good things to say about M-Audio's Delta PCI drivers (the mixer routing is not nearly as flexible as RME's software mixer but the card works with everything and has never skipped a beat in 3 years). Stay the gently caress away from their Firewire cards though which still have persistent driver problems. I believe some of those cards also use DICE II but I'd have to check that out before I say that for certain. WanderingKid fucked around with this message at 09:58 on Jun 19, 2008 |
|
#
¿
Jun 19, 2008 09:47
|
|
|
RivensBitch posted:I think WanderingKid has one, but I don't. I love my TC Powercore, even if the virus plugin can be a hassle at times. I'd say the TC reverbs are unparalleled, the VSS3 has your classic digital sounds, and the DVR2 has your glossy super sheen reverbs. It's the exact same stuff that runs on the system $6000, but it's 1/6th the price. I don't have one but I do lust after one. Unfortunately its not at the top of my lust list (which grows ever greater the less money I have). That dubious honour has to go to the Sunsyn (and yes if you have one I will pay you an obscene amount of money to take it off your hands.) Edit: speaking of reverb, number 2 on my lust list is an Eventide Orville which is quite simply the most spectacular effects unit I have ever heard. I maybe should have bid on the unit that eventually sold for 1.8 grand a few months ago on eBay but I cant afford a Sunsyn and an Orville. Oh if I had both you would never see me again. My metamorphosis into ultra hermit nerd would be complete. WanderingKid fucked around with this message at 21:12 on Jun 30, 2008 |
|
#
¿
Jun 30, 2008 21:06
|
|
|
wixard posted:Have you ever used their Eclipse? I am seriously starting to think about dropping way too much money on a personal rack of dynamics and effects. What amounts to a 1U H3000 would be nice, but I'm kinda worried about it being tough to navigate. H3000 is ancient but alot of folks over at gearslutz swear by one thing: If you want to get the Flood/Eno sound then thats the one you want. The H7600, H8000, Orville etc apparently sound very different and are capable of very different things but I couldn't say - never met anyone with a H3000. I've only heard the Orville and I am frankly just gobsmacked. The most unbelievable effects processor I have ever heard. I don't know about the H3000 but it looks similar to the Orville (at least superficially). The Orville is kind of fiddly to program. Its not a comfortable thing but then there are so many effects and they are so fully featured you really need to read the manual and toy around with every variable to get an idea of what it is capable of. Its definitely got a learning curve but the presets sound freaking incredible anyway. There I said it. For some reason I expected the auction to go way past 2.5 grand but it didn't and someone got it for 1.8. I think maybe I should have bought it but I would have maxed out my credit card and I would die a little bit inside if a Sunsyn went up for auction and I couldn't pony up the dough in time. An Orville or a H7600 or H8000 would do you for reverb, chorus, flanging, delay and that real time pitchshifting stuff (which is *insane*). You need never use anything else really. Seriously - go all the way. 5 grand. H8000. now. We can put up our cardboard house later. Oh yeah, check out this thread for some audio demos. It will make you cream your pants I'm not joking. And I don't even like the Andy. WanderingKid fucked around with this message at 00:57 on Jul 1, 2008 |
|
#
¿
Jul 1, 2008 00:51
|
|
|
Yeah the 7600 is sort of what the Orville has become now. The Orville is no longer in production. Recently I've started to get a real boner for Eventide gear. This is coming from someone who produces nearlly all digital and in software. Theres just nothing in software that can do what an Ultra Harmoniser does. Its feisible. You would just need skynet to process the 20 second long reverbs and real time pitch shifting without your entire rig crashing.
|
|
#
¿
Jul 1, 2008 02:09
|
|
|
iamlark posted:Alright, I have to ask now, what is a sunsys? Jomox Sunsyn. Best analogue hybrid polysynth ever. Sadly discontinued in 2007 after only something like 200 units were made. Jomox is only one guy and he basically said that building Sunsyns was a complete nightmare, the VCAs are no longer in production and his original board layout was a complete mess. IHe said that continuing to make them would necessitate an entire redesign and there are the outstanding bugs in the OS which prevent multi timbral mode from working properly in existing Sunsyns. He pretty much said that he longer wanted to ruin himself financially and waste his life making more of them. He said that he nearly went bankrupt. To be honest I don't blame him since under similar circumstances I wouldn't do it either. But what I wouldn't do to own one. I will already pay well over its original as new price (of about 3,000 euros) for a second hand unit.
|
|
#
¿
Jul 1, 2008 10:18
|
|
|
I use Cubase from time to time but I don't own a copy myself so I'm trying to go on memory and just well, logic. If what I am saying doesn't match up with whats on your screen I apologise but hey, someone will come along who does have Cubase and set the record straight. Your Ozone is a midi controller. It doesn't produce any sound itself. Rather you use it to control a synthesizer module. Therefore the first thing you need to do is connect the MIDI input port on the Ozone to the MIDI output port on your soundcard. You need to connect the MIDI output port on the Ozone to the MIDI input port on your soundcard. This lets you send MIDI messages between the Ozone and Cubase. Now in Cubase there should be an options menu which lets you choose your soundcard and map your MIDI input/output ports. Choose your soundcard. Also theres an option in preferences thats something like MIDI control change to automation track. Tick that box so that Cubase records all your button pushing and knob twisting.  If you get this bit right and set it up properly you should be able to test this. So open a new project and open that VST docking window thing and add Steinberg VB-1 to the list. Now go back to the main window. Add a MIDI track to the playlist and you should see to the far left of the screen theres a box with 'in:' and 'out:' Set in the input to your soundcard. Set the output to VB-1. Now double click in the playlist where the time line is near the top of the screen. Do it next to your MIDI track. This will plop down a little boxout called MIDI01 or something. Right click this box and scroll down to some line called something like 'MIDI' and click 'open key editor' If all goes well you should be able to plink some keys on the Ozone and you should hear VB-1 playing and see the corresponding piano keys lighting up in Cubase's piano roll. WanderingKid fucked around with this message at 20:38 on Jul 3, 2008 |
|
#
¿
Jul 3, 2008 20:36
|
|
|
Haha I just switched on my Xpander when I got home from work and over night its like its decided to gently caress with me. 3rd voice was nearly a semi tone off. 4th voice was quiet. 5th voice was quieter. 6th voice was almost inaudible. All of the last 3 voices were to varying degrees out of tune but not to the extent of voice 3. Voice 1, 3 and 4 were sounding twice as loud in the left channel compared to the right. Voice 2 and 5 were sounding twice as loud in the right channel compared to the left. At first I thought I killed the VCF but the filter was working just fine on voice 1 and 2. 30 minutes warming up and voice 5 and 6 are coming back to life. Still an alarming disparity in left/right channel volume on all voices but 6. I'll give it an hour and hope that problem just unfucks itself and then I'll start retuning this sucker. I had almost forgotten what a complete pain in the arse analogue synths are. The tuning stability of the DCO on the Juno-6 spoiled me rotten. Edit: Heh. I found a magic menu option that auto tunes everything. Wowza. WanderingKid fucked around with this message at 02:20 on Sep 16, 2008 |
|
#
¿
Sep 15, 2008 19:47
|
|
|
Most 5"/6" speakers don't have very good bass reproduction. Thats just physics at work right there. For reference I have a pair of Dynaudio BM5as (6.9 inch woofers) and the bass reproduction on them is really not very good. They sort of sound like rear end actually but you get used to it (and getting used to your monitors is really the most important thing anyway).
|
|
#
¿
Sep 19, 2008 20:35
|
|
|
The last time I looked into the Konnekt the word on the street was this: Great hardware for the money. Horrendous drivers from the blackest fetid pits of hell. Stay away if you have an ibook. It got so bad over at the TC support forums that I held off on buying a Studio Konnekt 48. This would be about a year ago and nothing really changed for about 6 months after that. I haven't checked since then but figure I don't need it anyway since my FF400 basically works with everything, all the time and has great driver support. I tried out the Saffire at temple bar music centre when it first came out and was really disappointed with the software which had some terrible bugs like defaulting all the channel gains to 0dB whenever you save a mixer state. The main outputs are really really loud too and this is the killer - without gain controls on my monitors I just couldn't be doing with poo poo like that. I dont know if they ever fixed that but if you have powerful speakers with no gain controls on the unit itself then get a second opinion because holy poo poo.
|
|
#
¿
Sep 22, 2008 10:21
|
|
|
As long as their drivers suck I wouldn't recommend any TC card, no matter how good the hardware is. If you can't use it day to day without it pissing you off then its no good. Especially since you have cards from the likes of RME and Echo that have good hardware for the dough and great software and support. In fairness to TC, their customer support was awesome when the tweeter in one of my monitors blew up but I've heard too many bad things about their Konnekt drivers and DICE-II to ever want to put myself through the hassle of it should I experience problems. I dunno.
|
|
#
¿
Sep 23, 2008 00:40
|
|
|
RivensBitch posted:Actually I get the exact same thing, cpu spikes here and there at the low buffer settings are definately audible. It's my biggest headache because as you said, anything but the lowest buffer setting is totally noticeable when you're playing 16th notes at standard tempos. If you mean proper softsynths (not hardware VAs) then you need to buy a Muse Receptor.
|
|
#
¿
Oct 2, 2008 21:37
|
|
|
Floops meters make no god drat sense either and theres no way to change them.
|
|
#
¿
Oct 8, 2008 13:41
|
|
|
Fruity Loops is now the benchmark for competancy in music production. If you can't make something sound good in Fruity Loops, you should save the money you've been planning to drop on Logic Pro, buy a hand gun and shoot yourself instead.
|
|
#
¿
Oct 8, 2008 19:48
|
|
|
I think you can get it without the remote for cheaper. Like 200 euro cheaper. For the money its amazing because it just gives you so much stuff but I've never managed to get Fabrik working properly. I get massive intermittent CPU spikes when adding/removing any plugins which instantly causes thousands of ASIO buffer underruns.
|
|
#
¿
Jun 10, 2010 07:18
|
|
|
Bah. Core Audio is some Mac only thing. BAH. A shame because Fabrik-R sounds really nice and doesn't use any CPU load...until you load/unload a plugin and then 99% CPU load, audio turns to loud, terrifying glitchy noise, DAW underrun counter goes nuts. Sometimes it just does this randomly, even when I open a project with nothing in it but Fabrik-R on a send bus. WanderingKid fucked around with this message at 10:48 on Jun 10, 2010 |
|
#
¿
Jun 10, 2010 10:44
|
|
|
Fuckin' Macs... or rather, fuckin' PCs...
|
|
#
¿
Jun 10, 2010 16:35
|
|
|
|
| # ¿ May 13, 2024 22:02 |
|
|
Huh, apparently FL Studio doesn't have a Mac version and it runs like poo poo in a windows virtual machine. No Apple for me I guess...
|
|
#
¿
Jun 11, 2010 11:25
|
|