advertisement


$13500 Dac shocker

Keith,
Do you have answer for us yet?

The answer is clearly in your avatar for someone who regular uses jitter and eye pattern measurements it is pretty obvious, so an answer who be appreciated.

Updated

So you have posted a print out from the measurements taken from a test sequence from the piece of equipment using a USB2 interface test programme at the end of production before being QA passes it fit for consumer land. Not what I asked for Keith at all.

For example fora USB3 high speed interface measurement, you would need a much higher specification of the test equipment with over 8Ghz BW coupled with a sample rate nicely over 10G/s to capture that 5Mps signal rate. You would be looking around the £45K+ mark for that level of capability. The Lecroy machine it was measured on are excellent, personally I use a mix of R&S and Tektronix which also excellent. However I do not bang on about at every opportunity.

My question was just by looking at the eye diagram it self what does it indicate to you, also could using a histogram in conjunction with said eye diagram give us any extra information as to the outcome at all?

As Graham intimated, you use measurements to undermine other equipment's capabilities, what about your own capabilities in this area of test and measurement and genuine relevancy to said product (s) and how they REALLY sound ?

Instead of posting the report sheets, how-about being a little more constructive realism with your constant self promotion and defection techniques for avoiding questions
 
I didn’t realise you designed electronics Tony, could you provide a link to one of your designs?
Regarding my avatar, eye diagrams just allow a means to quickly and accurately measure signal quality.
Keith
 
eye diagrams just allow a means to quickly and accurately measure signal quality.
Keith

What the eye diagram indicates is certain parameters of a digital signal:
Normally these diagrams are used to look at a received digital data stream (e.g. data received on a USB input) to check for distortion or corruption of the data stream for potential signal errors (distortion of "1's" and "0's")
The sensitivity of the system to timing (e.g. distortion of the timing of the data stream)
Plus any margin of noise (width of eye-opening) where noise could have an influence on the quality of the data stream.

This is about a digital datastream, not analogue signal quality as your reference above might indicate.
 
<moderating>

I have removed some personal/potentially libellous content. Keith, you big yourself up with measurements etc, so it is only fair that people should be able to challenge your personal knowledge and grasp of these things (burping-up specs without understanding is religion IMHO). You do not get to lash out at people in such ugly ways on this forum in response to fair questioning.
 
I didn’t realise you designed electronics Tony, could you provide a link to one of your designs?
Regarding my avatar, eye diagrams just allow a means to quickly and accurately measure signal quality.
Keith

Morning Keith

They are many projects I am involved with that are in the not public domain or audio related.

Back to matter in hand it is quite obvious that you are failing to grasp the basic fundamentals of the eye diagram's ability to deliver a quality indicator of the potential issues associated with high speed digital signal transfer.

For example just by looking at the shape of the 'eye' itself (without using the results tables either) will able to to determine the size and shape of the eye pattern which can indicate issues with bandwidth, amplitude issues, signal timing, noise, rise and fall times etc.

So the eye pattern in your avatar instantly gives the impression of a very well sorted and under control Jitter spectra. A large wide open eye with smaller intersecting lines and large angles is generally accepted as being well though out, though they are always exceptions.

The size and shape of the eye will virtually instantaneously tell you if you need to look deeper into any issues that may be prevalent from the eye pattern itself. Which will involve histograms in conjunction with the eye pattern itself, plus jitter analysis to include clock recovery, TIE (Time interval error) jitter trends which helps with periodicity and distribution measurements, cycle to cycle plus pulse width both negative and positive measurements. These would be a bare minimum to help you debug and remove the vast possibilities with any jitter that would be evident.

The ability to 'add' a PLL (Phase lock loop up to 3rd order) into the software on the device can help greatly with helping with jitter causes, for example by adding a loop filter (corner stone) to the equation via the math's function on the device you can 'see' in virtual time what direct effect than can have, this can reduce the amount of time in debugging or realizing that the original layout design needs to be reworked,

By digging deeper in the jitter analysis you can then determine whether you have Deterministic or Random jitter (which is more Gaussian), with deterministic you are then segmented into two further groups Periodic and Data Dependent types which are related to the inherit design of the equipment, circuit board layout, data transmission, clock performance, any cabling RF interference etc.

Quite a straightforward answer for any competent electronics design engineer with high speed signal analysis experience Keith

A nice little white paper on basic jitter theory and practice by On semi conductor one of the worlds leading semi conductor manufacturers

Eye Diagrams for uninitiated
 
What a pity I am sure we would all be fascinated to see your designs.
Keith

Well quite a few have and oddly enough quite like them.

Why do never give a straight answer in reply to a question?

So cutting through your deflection techniques yet again, why not admit you really have no idea on how to respond to the genuine question asked let alone more basic electronic principles? Admit the obvious to the forum Keith. Demonstrating rhino hide to questions will not save your reputation other than make you look anything other than looking like a plonker Rodney.

So my suggestion would be either go back to colleague and undertake a basic 'A' level foundation course in electronics, you will find it quite helpful.
Before trying to beat potential customers around the head with your internet cut and paste specification antics on subjects you have very little understanding or grasp of. Think of what it is actually doing to your reputation and ultimately to your potential income.

This is all I could find near Muswell Hill for basic electronic courses

Electrical Engineering Courses
 
Thank you Tony ,if I ever consider a career in electronic engineering I will look them up, how are TAD sales going?
Keith

Keith, every HiFi dealer should have a good knowledge of audio, as well as a basic understanding of electronic engineering that is used in audio. It's one of the reasons this industry has gone down the pan, with so few good dealers, as well as and too many individuals thinking "That looks easy, I could become a HiFi dealer too" without any understanding of the subject...

Tony makes a very valid point and you could actually benefit from his advice.
 
Blimey, that's a big shovel Keith.
He’s probably digging to Australia so that his supplier can come to his rescue ;)

So, looking at this thread am I to understand that Keith isn’t qualified to call my MBL speakers “effects boxes” :)

That’s a relief I was thinking I might have to buy a pair of speakers from him that makes music sound like hifi!
 
Poor Keith has been Bullhorned.


Bet he’ll never, never, ever do it again ... until the next time.

Joe
 
You are most welcome Camverton, may I suggest your 'effect boxes' would be alien to our Keith as his concept of musical reproduction is limited to his non understanding of how emotion is key part of audio replay. Mind you I am sure he can 'produce' some technical specification to further any argument he has for pushing product 'x' today.

Keep up the good work sir
 


advertisement


Back
Top