I'm also looking for soft start and speaker protection circuit for my impending PassDIY F5 build. How much did one cost you?
I'm also looking for soft start and speaker protection circuit for my impending PassDIY F5 build. How much did one cost you?
:lol: My stories (and they do read like looong stories) are more the struggles of a confused and ambitious experimenter than any accomplished designer. Getting inspired by me may make you as confused as I.Your builds are truly inspirational!
Subscribed to your new project.
A more accurate description would be "vast effort to share limited knowledge and experience." :lol:I've been following your blog for quite some time now and really appreciate the efforts you make to share your vast knowledge and experience.
Thanks a lot for the encouragement. And yes, i intend to keep writing, because it helps me structure my own thoughts.All the very best with the build, mate. I shall be following this thread eagerly. Do keep us posted and keep up the good work.
Thanks a lot! Will definitely read this. The MiniDSP side is a huge unknown for me, and I've jumped in with both feet (I've actually completed the purchase of a MiniDSP 4x10HD already, so the deed is done), but I am sure there will be pitfalls and challenges. God only knows, I will need all the help I can get.Oh yeah, Tarun, do make it convenient to go through this 2channelaudio article on miniDSP before you bite the bullet on the miniDSP
I read the article. I really appreciate your passing on the pointer to me.Oh yeah, Tarun, do make it convenient to go through this 2channelaudio article on miniDSP before you bite the bullet on the miniDSP (John K of 'Nao' speakers fame may not agree).
One tries to be of help. What else are friends for, eh?I read the article. I really appreciate your passing on the pointer to me.
Hmmmmmm . . . .I saw the following points which may give me a different conclusion than his:
- He tried the 2x4 module, which operates at a smaller word length and 48 KHz, while I am using the 4x8 module, which operates at a larger word length and 96 KHz
- I am hoping (without any quantitative evidence) that the ADC/DAC of the 4x8 module will be better than the 2x4 module, simply because they are more expensive overall
- He used the analog inputs, while I intend to feed digital inputs straight into the MiniDSP for all sources other than (someday) a turntable. This will eliminate the AD conversion at the input
- He tried using a volume control before feeding the analog input to the MiniDSP. This is completely a bad idea -- you cannot feed attenuated signals to an ADC if you want clean digital data. I intend to use the MiniDSP volume control at its output, which MiniDSP itself provides. (Of course, he says he even tried without the volume control, so maybe there is a problem independent of the low signal amplitude and quantisation noise.)
The test of the machine is the satisfaction it gives you.Let us see how my experience fares. Maybe my ears won't be as good as his, in which case I will live in ignorance-is-bliss Land. :lol:
I am actually quite hopeful about the MiniDSP because Uncle Seigfried The Man himself has blessed a miniDSP crossover for his latest OB speakers. (Of course, it uses different DA converters, but then those DA converters are not particularly fantastic either.)
Last but not least: I am admitting a secret thought: I had decided from Day 1 that if I don't like the sound of the MiniDSP, I will use it for getting my crossover right, and then will build an analog active crossover with exactly the same transfer functions. That way, MiniDSP will be used for experimentation, and when I'm ready to cast the xo in concrete, I will switch to an analog xo. Am keeping all bases covered.
"Watch this space!" as they say.![]()
Too generic and open-ended for me to be able to agree with it.The test of the machine is the satisfaction it gives you.
There isn't any other test. :indifferent14:
If the machine produces tranquility, it's right. :licklips:
If it disturbs you, it's wrong :sad: :annoyed:
UNTIL, either the machine:argue:
OR
your mind is changed.![]()
hyeah:
--- Robert M Pirsig
Yes, this was SL's original thinking on the subject. But he has accepted DSP based crossovers for his LX521 which someone has submitted to him for his evaluation: see here"Q5 - Will you offer a DSP based crossover/equalizer?"
"A5 - A DSP based xo/eq makes a lot of sense for manufacturing and cost reasons. For the DSP to become the sonic equivalent of the LX521 ASP will require a lot of attention to design and performance details in the digital and analogue bowels of the beast. I am not an expert in DSP application programming or even know how to chose the most suitable DSP engine. It is not on my priority list at this time."
FAQ LX521
You don't have to.Too generic and open-ended for me to be able to agree with it.![]()
Yes, this was SL's original thinking on the subject. But he has accepted DSP based crossovers for his LX521 which someone has submitted to him for his evaluation: see here
I presume you've seen this page on SL's site? it sings quite a different tune from his earlier FAQ entry you have quoted. He tested this MiniDSP rig with an interesting multi-channel volume control from MSB, but if I use the 4x10HD, I don't need any separate box for volume control. In fact, he himself mentions that Tony @ MiniDSP suggested the use of the 4x10HD, which SL accepted. I don't think SL could be any more explicit in his approval: he says "The miniDSP 4x10 Hd strikes me as more than an alternative to the ASP, accommodating both analog and digital sources and also providing volume control."
If you see his initial opinion about the feasibility of a DSP based crossover, you can clearly see that the issues he points out are genuine, but not at all unforeseen or insurmountable -- a lot of modern DSP platforms are tackling them well. A proper set of biquads on top, some attention to levels to ensure no clipping, and a good DSP xo can be made. For instance, it's hard to justify that levels are a challenge when floating-point DSP chips now exist. So, frankly, SL's comments about the DSP challenge had left me quite underwhelmed (if there is such a word). And as a demonstration of his true greatness and humility, he has later accepted (with the MiniDSP xo for LX521) that a DSP-based xo is perfectly feasible as an alternative.
I agree that it's hard to design a crossover. But if you can do all the hard work to design a good analog active line-level xo, then converting that to a digital line-level active xo is not a serious challenge -- I refuse to believe otherwise.
I contacted the seller and got details of this circuit. He is using the Omron G2R relay, a DPDT model, which gives him a 5A rating per contact. I think this is okay for my current build, but some of us may want a higher-current rated relay, just for improved contact. Just in case you were thinking about getting this one...I'm also looking for soft start and speaker protection circuit for my impending PassDIY F5 build. How much did one cost you?
I think most of us have read ZAAMM -- goes well with the DIY credo, I feel. I also read, last night, the other article you pointed me to from skeptic.com. ZAAMM has its limitations, and this skeptic's article too has too much of the tilting-against-windmills attitude for me to like it beyond a point. More importantly, I can't figure out why we need to discuss ZAAMM before we can discuss your choices of analog versus digital crossovers.Please do read what I had written in the other linked thread in reply to a FMs query about why I am (seemingly) hell-bent for analog.
Very smart! :lol:i am certainly not keen on getting into an analog versus digital debate
I did a bit of poring over the PDF files that the manufacturer offers, and also looked through the biquad generating spreadsheet (open source stuff for minidsp hardware) but that's about all. I did not read through the forums on the minidsp site.but Re: the minidsp platform, I have a few comments. I do not know if your purchase of minidsp was serendipitious, or well researched...
I completely agree. (I rarely do, with anyone, on any issue, BTW :lol... the minidsp platform is one of the most tweakable ones that a DIYer has access to. And I do not mean only from a processing horsepower point of view.
Agreed.1. Analog to digital conversion to native DSP sampling rate (potential source of degradation)
or
(For a digital source) Sample rate conversion from the input rate to the Native DSP rate (another potential source)
I have thought about this matter, and I am of the opinion that I will step out of the IIR arena only when I clearly identify a situation where I am unhappy with IIR. Currently, I am nowhere near that point -- I would be happy with analog active, and I am using digital xo only to simulate the analog active xo.2. the processing - Minidsp uses IIR filters which are pretty efficient. (I do not want to get into an FIR versus IIR debate, since I have no experience of FIR filters)
Agree 100%.3. the analog conversion - the DACs and their implementation defines the final output and in a DSP board that packs 8 within a price of 299USD, it probably is just not the latest cutting edge.
I intend to feed it just 44.1K/16bit to start with, and I'll see what I get.IMHO & experience with the 2x4 minidsp (and 2x8 which the 4x10 retail box is based on is a more modern platform) the ADC is not the best. but the minidsp accepts digital inputs and you can bypass the SRC issues by feeding native resolution stuff (trivial if you are on computer audio).
In fact, I intend to explore all that once this project is done. The most serious limitation and irritant of the 4x10HD is that it allows only one SPDIF input and one set of unbalanced analog inputs. In reality, I need to use that box as a preamp, and I would like to feed it at least three SPDIF inputs and three sets of analog RCA. It's a pain that I can't do it. So, for this reason alone, if not for anything else, I think I'll have to move to a naked miniDSP board a couple of years down the line, with pure digital ins, and then build my own ADC and channel selector layer upstream of it.finally, the minidsp accepts i2s inputs so you can do your choice of ADC if you so choose, later.
There be dragons down that route my friend.regarding the DAC conversion; there is a provision of i2s output, which allow you to connect up your choice of DAc...
minidsp itself is now selling a Sabre 9023 kit for all of USD40. you could choose to just change the mid channel or do all 3/4 pairs for usd 120/160. and if you want esoteric, you can get a a sabre 9018 kit for less than USD 100 (DIYin HK). i am sure even cheaper if you shop around. And with a 9018 you are talking cutting edge.
Hey, do you mean the official Adobe-AIR based interface and its GUI for setting filters and parametric EQ? That won't be of much use at all -- I might as well use a Behringer or something in that case. I intend to use the community-created spreadsheet to generate biquads and then feed them to the MiniDSP... but from what I've read on the net, the Minidsp interface does not fully exploit the DSP's full capability...
If you suffer from insomnia, you may try reading my account of the Asawari Mark I, and you will then see how long I take to do the most inane of things. Patience -- I don't have an option but to be patient!@#!Essentially, what I am saying is that unless you are a rather impatient person (and you do not strike me as such); I'd recommend you give minidsp a rather fair shake.
Noted, sir. I intend to get rid of it only when I finish a custom-built preamp for the Darbari with a naked MiniDSP board inside it and my own AD and DA layers upstream and downstream of it. May be another two years. Don't wait up.finally, if you tire of it quickly, I will be happy to take it off you hands.
How does the 9023 compare with the 9018? (I, given my "vast knowledge", was under the impression that 9023 would be better, because it's a larger integer. :lolHow do these various DAC boards compare? Is there any difference between these Chinese DAC boards and the one from Twisted Pear Audio, for instance? Do all these super-fantabulous ultra-linear regulated power supplies and rhodium-filmed platinum-plated capacitors make an audible difference to the sound? Any inputs would be really great.
A thought just occurred -- why would an upsampling done in my computer in software be better than upsampling done by the digital input of the MiniDSP?... the minidsp accepts digital inputs and you can bypass the SRC issues by feeding native resolution stuff (trivial if you are on computer audio).
Would you hazard a guess about the relative quality of the Buffalo 8-channel DAC vis-a-vis the native MiniDSP 2x8 DAC? Will the Buffalo be audibly better?...The 9018 is now nearly the de-facto choice to USD800+ DAC manufacturers looking for an all-in-one chip solution, including the guys at Twisted Pear Audio who use it in the Buffalo...
To be taken with a huge pinch of salt and based solely on market trends, but the ESS9018 DAC on the Buffalo does see more application in "audiophile DACs" than the Cirrus Logic DAC on the MiniDSP. Plus, the Buffalo is a one-trick pony, so I would assume that the design is more optimised for audio than the MiniDSP (though the specs on the 2x8 datasheet do look quite good).Would you hazard a guess about the relative quality of the Buffalo 8-channel DAC vis-a-vis the native MiniDSP 2x8 DAC? Will the Buffalo be audibly better?
To be taken with a huge pinch of salt and based solely on market trends, but the ESS9018 DAC on the Buffalo does see more application in "audiophile DACs" than the Cirrus Logic DAC on the MiniDSP. Plus, the Buffalo is a one-trick pony, so I would assume that the design is more optimised for audio than the MiniDSP (though the specs on the 2x8 datasheet do look quite good).
Cost-wise, MiniDSP is more attractive and its a Swiss army knife. I'm sorry if this was more confusing than helpful :lol:
Went through their products, but can't find it. Can you please help? :sad:... minidsp itself is now selling a Sabre 9023 kit for all of USD40...