Hi Peter,

Your proberbly aware but PCIE latency may be also be good place to focus in code as well. I'm finding a pcie latency of about 128 on the Jul@ card ( or usb interface for a hi face ) is worthwhile. Using a utility you can set the audio device to a different latency to other devices which run at BIOS default (often 32).

Nick.

Mani,

Glad to hear things have gone well. 260lbs of iron and copper now that's some investment in sound  ;
Best
Nick

I'm watching the multi core discussions with interest. I think Peter's demand for processors is out stripping Moore’s law , we were fine with 2 cores 18 months ago now we might need 24, zikes ! 

Still if that’s where the sound is going I'll be signing up too, I'm glad I put off upgrading for now.

Nick.

Mani hi,

That’s some power setup, with a 13kVA isolation transformer I hope your floor is reinforced  .  I have heard good things about dedicated spurs and ionic grounds. Looking forward to hearing how things go.

Nick.

Just looked the max memory values up, there's quite a difference. It looks like Home Basic 8gb, Home Premium 16gb, Ultimate 192gb on 64bit.

I'm thinking home premium with upto 16gb should be ok.

Nick

Peter hi,

Are your high sfs value with z-04 ? Or are you trying new code to get to these values ?

Nick.

Ps any improvement in sound with these high values ?

Telstar,

Thanks I think I am likely to go for Premium for exactly the reasons you have given. Thanks for the help, I can purchase with confidence.

Nick

Sting "The Soul Cages".

I last listened to this when my system was vinyl based, it went way beyond the speakers. Hmmm might get it again on CD just to give it a try.

Nick.

Thanks Telstar, I'v been taking a look at some of the filter theory on Wikipedia, fascinating stuff but I haven’t seen math theory like this stuff in years  .

Peter some serious maths when the time comes to look at the crossover algorithms..... respect  !

Nick.

I have been trying to fathom how a software cross-over filter might work. The logic I am stuck on is this;

In order to construct a digital pass filter it must surely be necessary to analyse a “N” samples at a time in order to determine frequency components of the data. I guess the size of N would be related to the lowest frequency the filter is to act on ??. What has me puzzled is that if the filter algorithm does need to analyse N samples would this lead to the filter ringing in a similar fashion to an oversampling filter ? Might this undo benefits of the faithful treatment of data that Arc Prediction and NOS achieve ?

DSPs were not really around when I did this sort of thing so I haven’t much of a clue, so sorry of this is bit of a stupid question.

Nick.

Peter,

Thanks for the explanations, interesting stuff. I had not read about anyone using GainClones in this way. Thinking about it  why not ?

All of this has made me start to think about the possible benefits of different cross-over approaches. No plans to act too quickly but defiantly a topic to think about in the background 

Regards,

Nick.

Roy,

I missed this one...


I think you are describing what I am doing. The DAC feeds via a pair of normal (left and right channel) interconnects from the DAC or preamp into the Amp. Inside the amp (for each channel) the inputs either go via the TVC to the inputs of Gainclone 1 and 2 when I want a volume control or in the case of the direct inputs the input RCAs are connected directly to the inputs of Gainclone 1 and 2.

For the Gainclones I have this gives and input impedance (of both Gainclones 1 and 2) of about 12k ohms so you need a preamp / DAC that can drive this input. Its fine for most solid state output stages but I added the TVC so that I could drive the amps inputs with a Valve stage if I wanted to.

Regards,

Nick.

Hi Roy,

My first question would be how efficient are your speakers ? People do use Gainlones with moderately efficient speakers but they give about 35 ish watts which only goes so far I guess. I think Peter St and I have both found you can get a long way using them with efficient speakers and easy load characteristics.

For my amp I bought a couple of Peter Danial’s premium kits giving me 4 Gainclone modules. I replaced Peter D’s supplied main caps with some BlackGate Std 1000uf caps (no BG N Types left anywhere  ). You need + and – 25 ish volt rails to power them, in my amp this is created using couple of 300Va R-Core transformers (people also recommend good quality torroids). Peter D’s kit has all the rectifiers and PCBs so it very easy to assemble. I am very impressed with the kit, its very nicely designed and gives an extremely pleasing sound when built.

http://www.audiosector.com/lm3875.shtml

I have direct and volume controlled inputs. The volume control is via TVC’s wired before the Gain Clones. I agree with Peter St that direct input from the DAC is better using digital volume control. I spent about £1500 (inc £500 for the TVCs) which is way, way over the top because I went for high spec components, but the amp is still incredible value for money. I retired 5Ks worth of parallel 300B mono block amps when I built the new amp and have not looked back once.

Thinking about it using an active cross between the DAC and Gainclones would be interesting but the experience of using the NOS1 DAC direct connected with digital volume makes me feel that if Peter St cracks the software cross over using the NOS1 that would be the way to go.

I think Peter St would have to provide help with the Juli@ config questions, I am not sure what he would have in mind there.

Nick

Peter,

This is interesting, I would be tempted to bay pass the SubWFilter, but you have me wondering what the benefit is of the stacked Gainclones ?


Gainclone 2 only sees the bass spectrum passed by the Line Corssover filter which prevents it having to process higher frequencies ?

Gainclone 2 maybe has a PSU feeding it that is optimised for good low frequency response (eg more capacitance or something) ?

The DAC driving the input sees ~20k Ohm load rather than two gainclones in parallel giving say a ~10k Ohm load (but the output impedance of the NOS1 is easily low enought to drive two or more Gainclones).

Some quality of the high gain from having three amp stages ? but I am not sure this in itself would be doing anything. Perhaps you need this to keep up with the 115db response of your Mid High driver ?
Any insights ?

Regards,

Nick.

Peter hi,

If I understand your post I think your speakers may have two driver channels a combined High/Mid and separate Bass (?), whereas I have three driver channels High, Mid and Bass.

In my setup, working forward from the each Gainclone module’s output in my amplifier towards the speaker driver, each speaker channel looks like this;

Horn loaded Tweeter

Gain Clone Module 1 > Bi Wire speaker cable 1 > Tweeter Cross Over > Tweeter Driver.

Bi Polar reflex Bass

Gain Clone Module 1 > Bi Wire speaker cable 1 > “High Level” feed to Active Bass amp within the speaker via its high level input, the speaker’s internal base amp has volume control and roll off pots to trim its Bass integration with the other drivers > Bass Drivers.

So yes as you say this Gain Clone Module 1 does two jobs, first it drives Tweeter and second it acts as a "pre amp" for the active base amp within the speaker. The good thing here is that speakers internal bass amp only present a load on the K Ohm range so Gainclone module 1 only really has to “drive” the tweeter.

Horn loaded Mid Range

Gain Clone Module 2 > Bi Wire Speaker cable 2 > Direct connection to mid-rage horn driver without any cross over.


I build the Gainclone amp with two modules per channel as left / right mono blocks electrically going back as far as to the amp’s power cable but everything is in one case. There are two Gainclones modules per channel fed by a separate 300Va R-Core transformer and rectifier. (As an upgrade I guess I might gain an improvement by adding a couple more transformers and rectifiers to get mono blocks on all 4 Gainclones modules back to the power cord ?)

As you say my setup is not quite true Tri-amping because the speakers active bass amp is fed by Gainclone module 1 (above). If I were to go down the active software cross-over route I would tap in to the existing active base amp within the speaker cabinets by passing the internal bass volume control and cross over to drive the amp it at line level from a DAC output.