Forums | Search | Archives

All Forums  Hardware & Soundcards  Presonus vs. Focusrite 8 ch pre/comp litepipe out

Author	Topic
DeluXMan Location: Canada Posts: 330	Posted - Tue Jun 17, 2003 11:44 pm  Does anybody have an opinion or know which company has the better reputation when it comes to A/Ds and preamps in general?
VoodooRadio Location: USA Posts: 3971	Posted - Wed Jun 18, 2003 5:37 am  Sure! I've used both Presonus's (Digimax and M80) and Focusrite's (Octopre and Voicemaster). IMO, (you did ask for an opinion) I think the quality is comparable. For another product in a similiar quality/price range, you might check out the Joe Meek stuff (I have a VC1 Studio Channel). I recently "acquired" (a recording project for equipment trade) an Avalon VT737SP and although it's not an 8 channel pre, the quality is outstanding! My advise is.... take a trip to your local dealer and spend some time "auditioning" various models. _________________ I said Good Day! Voodoo
DeluXMan Location: Canada Posts: 330	Posted - Wed Jun 18, 2003 3:31 pm  Thanks for the advice Voodoo! I need at least 8 channels to record drums, but then if the Avalon would make the drums sound better i guess i could rent 4 units for a month for the price of an octal unit. Oh no more decisions. Hmm...
DeluXMan Location: Canada Posts: 330	Posted - Wed Jun 18, 2003 10:45 pm  The Joe Meek stuff looks good, I'll definately have a look at that. But what do you make of this comment from Joe Meek about their compresser vs. plug-ins: I assume plug-ins refers to digital compression, so is there any truth to this stuff about TTT? " TTT - Transparent to Transients. The classic JOEMEEK sound, impossible to replicate in a plug-in. Unlike other compressors, (VCA's and plug-ins) the JOEMEEK optical circuit allows the transience of a signal through unharmed. This means more crunching compression, without any 'dulling' side effects. "
SteveG Location: United Kingdom Posts: 6695	Posted - Thu Jun 19, 2003 2:59 am  'TTT - Transparent to Transients. The classic JOEMEEK sound, impossible to replicate in a plug-in. Unlike other compressors, (VCA's and plug-ins) the JOEMEEK optical circuit allows the transience of a signal through unharmed. This means more crunching compression, without any 'dulling' side effects.' [/list:08ce5cfacd] Yes, there is some truth in it. But it's not by any means impossible to replicate. What they are doing is making a virtue out of an absolute neccessity. The optical circut they use has a selenium cell driven by a LED, and it is the cell's rate of resistance change that limits the attack time of the compressor. They use the cell because it just exhibits resistance - it has no frequency or polarity constraints that an optocoupler would have. It has no frequency constraints because it has virtually no frequency response to speak of at all! But it does make a good variable resistor, so they use this as a gain control within the compressor. But it can't respond to transients, so they always get through. And yes, you can do this in the CEP Dynamic Processor, although the rate of change isn't quite as smooth. It does give you quite a 'thwack' on drums, though - the initial transient is unsuppressed, and the level of the rest of the hit is all that is altered. There is a thread about this 'Understanding the Level Detector'. The files referred to are still in place. _________________
DeluXMan Location: Canada Posts: 330	Posted - Thu Jun 19, 2003 6:09 pm  It sounds like maybe another ledgend is living in the past? 8) As far as i can tell the CEP dynamics processer can deliver a mighty whollap without resolution problems at 48kHz. How smooth isn't the rate of change in the CEP processor? Can you quantify/qualify? Wouldn't running at 96kHz. 24bit or higher overcome such unsmoothness? :P btw - i once made a super-wawa pedal using selenium cells driven by LEDs to control the center frequency. and yes, they are slow! Oh! i'm also wondering and have been assuming that the recovery time is similar to the response time for a selenium cell, but there may be a lucky asymetry that just sounds good prehaps, that relates nicely to attack/release settings on a vca compresser.
SteveG Location: United Kingdom Posts: 6695	Posted - Fri Jun 20, 2003 1:39 am  DeluXMan wrote: How smooth isn't the rate of change in the CEP processor? Can you quantify/qualify? Wouldn't running at 96kHz. 24bit or higher overcome such unsmoothness? :P btw - i once made a super-wawa pedal using selenium cells driven by LEDs to control the center frequency. and yes, they are slow! Oh! i'm also wondering and have been assuming that the recovery time is similar to the response time for a selenium cell, but there may be a lucky asymetry that just sounds good prehaps, that relates nicely to attack/release settings on a vca compresser. The sample/bit rate doesn't make a scrap of difference to this at all - it only relates to the the time constants in the level detector. I haven't timed it, but you can do this for yourself by recording a constant frequency and amplitude sine wave and processing it with the dynamics processor - you should be able to determine the attack and decay rates quite easily. But unlike the selenium cell, at least the rates in it could be modified. I suppose that we could request a rate-change parameter - let's face it, the processor is already complicated, and one more control would hardly seem to matter. And it would also go along quite nicely with my request for a method of side-chain access - which has to be applied to the same module. _________________
DeluXMan Location: Canada Posts: 330	Posted - Fri Jun 20, 2003 3:12 pm  Thanks SteveG. I'll try the sinewave tests. :P I have been assuming, correct me if i'm wrong, that for both the gain processor and level detector you can set any attack time smoothly, ie. floating point - .01ms. for attack time, for very sharp slopes. Also, until the details of the dynamics processor are documented, I assume that if you enter the same value for detector and processor the result is as you might expect for a single attack/release parameter, but with a sharper [second order or higher] response, or if not sharper then steeper. If true, I don't see where the limit is to the smoothness you can achieve in a digital implimentation, as long as the clock rate is fast enough, which must be fast enough to provide a decent number of samples to describe the attack/release curves. As the attack times approach the sample rate you get fewer and bigger steps, but if you keep the sample rate say about ten times the attack time you get a decent number of points to describe the attack curve. This is assuming also that attack/release values are floating point. For .01 attack time this slope is expressed over 40 or so samples at 44.1kHz. Or does the roughness of 'plug-in' compressers come from some other limitation not yet considered here? The reason i find this important to examine is that many good people are accepting hands down claims like that made by the joemeek co., that analog compressers are still better than digital implimentations.
SteveG Location: United Kingdom Posts: 6695	Posted - Sat Jun 21, 2003 8:36 am  DeluXMan wrote: Thanks SteveG. I'll try the sinewave tests. :P I have been assuming, correct me if i'm wrong, that for both the gain processor and level detector you can set any attack time smoothly, ie. floating point - .01ms. for attack time, for very sharp slopes. Also, until the details of the dynamics processor are documented, I assume that if you enter the same value for detector and processor the result is as you might expect for a single attack/release parameter, but with a sharper [second order or higher] response, or if not sharper then steeper. The responses are not simple to determine - as you will find if you do some experiments! The level detector seems to have the ability to 'hold off' doing anything for some time, but trying to correlate the settings with what you can measure is a somewhat frustrating experience, and yes, I would like to see a more detailed explanation of exactly what is happening myself. The attacks are smooth, certainly, but the two sections of the processor interact to a significant degree, which really does make setting it up a process of exploration! Quote: The reason i find this important to examine is that many good people are accepting hands down claims like that made by the joemeek co., that analog compressers are still better than digital implimentations. Analog compressors that have a reputation usually have a 'sound' of their own - and some people really like this. The problem is that these people tend to be the 'older' guys, and for some reason, people seem to venerate every word they utter. And since some of it is an utter load of mis- or ill-informed twaddle, it causes a lot of confusion. What may have been a good sounding compromise back then now sounds like exactly what it really is - ie, a compromise. Yes, these old devices will work fine on some sources, but on others they will sound awful. They are not a universal 'fixall'. Will they help you to recreate that 'valve' (tube) sound? They probably will if they have valves (tubes) in, but a surprising number of them didn't in some quite critical areas - so the 'sound' question is not as simple as you might hope for, either. But Ted Fletcher's (Alice, JoeMeek, TFPro, etc) designs for compressors are probably as good as you can get from a selenium cell architecture. And he's a nice guy. Yes, he's been around forever, but unlike a lot of the producer 'pundits', he knows what he's talking about. And his lab is tidier than mine! _________________
	Topic

Page: