The "concerns" have been clearly addressed and rejected in the last few pages by several members. I wonder if you have another motive.
Ian has kindly refused your request, maybe you should respect his decision.
Careful when quoting ppm values - most clock system ppm values are measuring short-term stability, as that is what matters for jitter and circuit function, but long-term absolute values are usually a different matter.
Such as?
As I wrote (for some reason you seem to have quoted something completely unrelated): "As it is your design, you are of course free to do whatever you want", so yes, I respect his decision.
I have discussed one design decision and the implications of that decision. The consensus seems to be that while it is agreed that there is a theoretical issue, it won't be a problem in practice. I have no need to take this discussion any further.
ryan, I do share some of your frustration here that its being carried on soooooooooo long again, when its all been discussed several times, but no, I think Julf is just interested in a solution, its a 'problem' so he wants to solve it or at least discuss what the solution may be. hey thats fine but it is taking it a bit far to expect any changes or hardware updates to adress a problem that hasnt had any effect on anyone at all as yet.
Ian obviously tested all this long ago and simulated bad clocks etc, I gather he got bored before he was even able to force a problem with realistic values of input clock jitter. of course he could have taken the mythbusters approach and placed more and more unreasonable/unrealistic demands on the system till it breaks, but why?
as hinted at, I gather it either needs hardware modification, or exposing the source code in some way to display the information, so its totally Ians call on if/how to deal with that.
actually its my experience that most clock ppm specs are long term so thus have in the past been pretty useless to indicate close in phase noise. long term jitter is the easy one to get right afaik.
Ian obviously tested all this long ago and simulated bad clocks etc, I gather he got bored before he was even able to force a problem with realistic values of input clock jitter. of course he could have taken the mythbusters approach and placed more and more unreasonable/unrealistic demands on the system till it breaks, but why?
as hinted at, I gather it either needs hardware modification, or exposing the source code in some way to display the information, so its totally Ians call on if/how to deal with that.
actually its my experience that most clock ppm specs are long term so thus have in the past been pretty useless to indicate close in phase noise. long term jitter is the easy one to get right afaik.
Thanks Jeremy,
I'm usually on the side lines observing very closely as i let the experienced members discuss problems and solutions. But even myself as an intermediate amateur I can understand that it is not a problem if its not a problem.
Lets continue discussing meaningful problems and solutions instead. Ill be on the bench with my mouth shut.
I'm usually on the side lines observing very closely as i let the experienced members discuss problems and solutions. But even myself as an intermediate amateur I can understand that it is not a problem if its not a problem.
Lets continue discussing meaningful problems and solutions instead. Ill be on the bench with my mouth shut.
hehe no worries Ryan, agreed sticking to the discussion of meaningful problems is a good idea.
as hochopepper said, it is true that if we had access to this register that knows how full the memory is, it should be abe to be leveraged to determine the delay vs the input clock and thus provide matching delay for a video feed sync. thats a much more meaningful use of the information, for knowing how full the fifo is at any time, its a curiosity at best, I know from experience that it simply never happens given my use patterns.
we were talking about this some time ago wrt to the video sync possibilities and the possibility of adding such functionality to the si570, given it probably has access to the mentioned info on the fifo, as well as knowing what the current master clock frequency is, as well as having the spi ports needed most likely
as hochopepper said, it is true that if we had access to this register that knows how full the memory is, it should be abe to be leveraged to determine the delay vs the input clock and thus provide matching delay for a video feed sync. thats a much more meaningful use of the information, for knowing how full the fifo is at any time, its a curiosity at best, I know from experience that it simply never happens given my use patterns.
we were talking about this some time ago wrt to the video sync possibilities and the possibility of adding such functionality to the si570, given it probably has access to the mentioned info on the fifo, as well as knowing what the current master clock frequency is, as well as having the spi ports needed most likely
Last edited:
Indeed. As I explained earlier, my interest was driven by the fact that the buffer under/overrun situation is the argument usually quoted to justify the need for a samplew rate conversion, something I really want to avoid.
I have no better solutions or competing designs to push.
And I wasn't asking for any hardware changes - the discussion we had pretty much resolved my concerns, but as Ian was writing:
I was assuming that the functionality was already there, and no significant changes would really be needed.
I take it that the calls are there and if we are lucky available on one of the headers, but that doesnt mean its just as flicking a switch. not everyone in the GB is capable of mods on SMD, or writing code to display or react to changes of status headers. At the same time it may be a function that many feel they may need, even though it isnt actually a problem in any reported case; so Ian may wish to avoid a rush of people wanting to send their fifos back for the update, for essentially nothing.
so even if it doesnt require a hardware respin, it can be more of a hassle, because people would BUY a hardware update, while firmware updates have so far been free....
so even if it doesnt require a hardware respin, it can be more of a hassle, because people would BUY a hardware update, while firmware updates have so far been free....
The biggest "issue" brought up in the last few pages was saying the FIFO is not DIY comparing it to a guy who creates a preamp and only gives the schematics to buyers of his PCB.
This FIFO is competely different. As a Diyer we should regulate the opportunity to have such techology. As a DIYer just think of this as a CS8412 or WM8804 DONE RIGHT !
It is no different than going to Ti and buying a spdif receiver chip, or going to Xmos to buy an asynchrous USB reciever. Do Does Ti or Xmos release their "code" or structure?, no.
So if we were to say this is not DIY then any CPDL, FPGA, basically any silicon for a digital signal is not allowed in a DIY endeavor, we must be stuck with crummy poor performing discreet synchronous +vcxo solutions that have been done before.
This is a new direction for DIYers we should embrace the new technology not fear it because it works better than anything we can buy off the shelf or build.
This FIFO is competely different. As a Diyer we should regulate the opportunity to have such techology. As a DIYer just think of this as a CS8412 or WM8804 DONE RIGHT !
It is no different than going to Ti and buying a spdif receiver chip, or going to Xmos to buy an asynchrous USB reciever. Do Does Ti or Xmos release their "code" or structure?, no.
So if we were to say this is not DIY then any CPDL, FPGA, basically any silicon for a digital signal is not allowed in a DIY endeavor, we must be stuck with crummy poor performing discreet synchronous +vcxo solutions that have been done before.
This is a new direction for DIYers we should embrace the new technology not fear it because it works better than anything we can buy off the shelf or build.
Last edited:
Who is to say what is DIY and what isn't? The International DIY Committee? And who cares anyway?
Not sure it is a new direction in any way - DIYers through the ages have always embraced new technologies. I built my first microcomputer in 1977, but no, I didn't put together each gate from discrete transistors...
This I agree with 110%!
The other issue is the concern that documentation is littered through this thread. It is dead easy to use the search function and isolate only post's by Ian, they are the only source of 100% truth, though the rest of us know a few things from experience, Ian has taken the time to respond to everything himself to at least confirm what others have said before he's had a chance. At that point the post count is far more easily consumed.
Another tip, use technology to make it easier to keep track of notes/details that are important. I use Evernote to save important posts and have posts tagged with what project(s) of mine that they relate to, so that I can find the interesting stuff quickly (I did this with Si570 serial protocol as just one example). I make this even easier for myself with the Evernote and Clearly extensions for Chrome. With a few of these extra tools (and dropbox to store datasheets/documentation) I am able to quickly refer to important details when I need them. This is really no different to the way most projects evolve on DIY forums where development is ongoing and using a few tools means you can easily keep track of just the information that is important to you.
With the above (and calculations in spreadsheets in Google Docs) I am able to access all of my notes, calcs, datasheets from practically any internet connected device. Using a tablet on my workbench for circuit schematics and datasheets makes working on these projects a lot simpler for me I find.
Where and when was this?
I read back 10 pages of the most recent discussion, and couldn't find anybody saying anything like that.
the one from percy I replied a bit tersely to that was getting a bit bent out of shape because apparently there wasnt a complete rundown on development including schematics in the thread.... it did happen.
Julf, are you actually interested in this as anything more than a curiosity? its just you do seem to be spending a lot of space basically questioning everything posted...
Julf, are you actually interested in this as anything more than a curiosity? its just you do seem to be spending a lot of space basically questioning everything posted...
Last edited:
So we're way off course here, and while I usually make it a point to stay out of the "skirmishes" - Im gonna toss in another view. Maybe it's because I don't want our Australian blokes to be painted as perennial grouches (though one could make a compelling argument with qusp) Cheers mate!
I must confess to being a bit annoyed by percy's post. It's a bit hard to articulate, as there's a subtle line between reasoned and researched curiosity and a query that feels a bit too demanding, or challenging (for justification), or "everybody - send your time/energy gathering together the information/measurements that I need to be satisfied that this project is worth my time".
I don't know. As I said, I usually stay silent with these things because it's all too easy in this one dimensional communication 'channel' to misconstrue. It may not have been (likely wasn't) intended to come off that way, but it felt that way to me - and it bothered me.
I'm generally not a sensitive guy, but this project (and one or two others) stirs up a protective reaction. I think it's because what we're witnessing and are fortunate to participate in is very very rare. A brilliant gentlemen is (quite against the norm) pouring his time, knowledge and funds into a project that has elevated my audio chain(s) and enjoyment beyond what I could have purchased. Seriously. I've learned alot, and I'm hugely appreciative.
And Ian keeps pushing; designing, sending out proto boards, helping out. Honestly, I don't get it. It's so above and beyond what's usually done that it leaves me a bit protective.
I'll stop rambling with a request. Respect this community. Respect the minority few that pour their time/effort/resources for the good of all. We are owed nothing, and nobody has to prove anything. Tread lightly, and be willing to do the work for yourself and others.
"Can you say lecture? I know you can!"
Those of you who know MisterRogers will get that
I must confess to being a bit annoyed by percy's post. It's a bit hard to articulate, as there's a subtle line between reasoned and researched curiosity and a query that feels a bit too demanding, or challenging (for justification), or "everybody - send your time/energy gathering together the information/measurements that I need to be satisfied that this project is worth my time".
I don't know. As I said, I usually stay silent with these things because it's all too easy in this one dimensional communication 'channel' to misconstrue. It may not have been (likely wasn't) intended to come off that way, but it felt that way to me - and it bothered me.
I'm generally not a sensitive guy, but this project (and one or two others) stirs up a protective reaction. I think it's because what we're witnessing and are fortunate to participate in is very very rare. A brilliant gentlemen is (quite against the norm) pouring his time, knowledge and funds into a project that has elevated my audio chain(s) and enjoyment beyond what I could have purchased. Seriously. I've learned alot, and I'm hugely appreciative.
And Ian keeps pushing; designing, sending out proto boards, helping out. Honestly, I don't get it. It's so above and beyond what's usually done that it leaves me a bit protective.
I'll stop rambling with a request. Respect this community. Respect the minority few that pour their time/effort/resources for the good of all. We are owed nothing, and nobody has to prove anything. Tread lightly, and be willing to do the work for yourself and others.
"Can you say lecture? I know you can!"
Those of you who know MisterRogers will get that
I would second that, I have been watching this thread with great interest for many months, as it is one of the more advanced and interesting DIY projects.
One question I would ask is if you had the schematics what would you do with them? Lay out a board!
In the real world of electronics a project of this size would probably involve 3-4 people and cost thousands so I think that we are all getting value for money here.
One question I would ask is if you had the schematics what would you do with them? Lay out a board!
In the real world of electronics a project of this size would probably involve 3-4 people and cost thousands so I think that we are all getting value for money here.
Well, here's the point where I could walk through my post, quote out what I said... and around we go. That's not what I intended to say, or what I said. I think you're a bright guy, so you can discern my intent, if you choose to. I know it's a subtle line, and I honestly gave you the benefit of the doubt that yours was an honest reasoned question, not intending to be callus.
I was simply trying to shed some light on why quite a few took issue with your post. As a part of our community, you'll either care to hear, learn and understand... or not. I've got nothing against you mate, and again - I don't believe you intended to chafe. When I do and it's more than one, I'd like to know and understand why. That was my intent.
I'm pretty sure that if you stick around, you'll see me unintentionally do the same
Peace.
I was simply trying to shed some light on why quite a few took issue with your post. As a part of our community, you'll either care to hear, learn and understand... or not. I've got nothing against you mate, and again - I don't believe you intended to chafe. When I do and it's more than one, I'd like to know and understand why. That was my intent.
I'm pretty sure that if you stick around, you'll see me unintentionally do the same
Peace.
Saw your edit. I trust you do appreciate Ian's efforts. Honestly - you'd have to be pretty daft not to. This isn't about seeking information; at least not to me. I don't read anything in the other members response that indicates that you, or anyone are wrong to seek information.
It's about the how. That's either something you care about, or not. Let's move on.
It's about the how. That's either something you care about, or not. Let's move on.
