my thoughts on hdmi cables, and the randomness of potential data loss...

[Clare Newsome]

I am getting so fed up of HDMI 'testing' that doesn't bring sound into the equation....

As you mention, our own "Big Question' feature where we asked three readers to blind test HDMI cables resulted in more noted differences being perceived in audio than video.

Personally i've participated in blind tests where we've experienced a difference in both picture and sound quality.

 

[Anonymous]

Most of the answers are subjective so how about a more scientific approach?

Take a known source, say the fist 1GB of any film you like and save as a file on disk on computer 1.

Read the computer 1 file and send via HDMI to computer 2.

Computer 2 captures the data and saves to disk.

Now compare the file on computer 1 with that on computer 2.

Can not be that hard can it?

Given the same equipment and changing only HDMI cables will allow the relative performance of cables to be evaluated. Who knows we may get manufacturers putting their Mr Mellie Benchmark results on their packaging.

 

[idc]

Clare, would you post the protocols for and results of your blind tests?

The issue here is randomness. Does WHFs blind testing find that all present identify the same thing (whether that is image, audio or both) with the same cable using the same track or film clip? If they do, there may well be something in the cable that is making the difference. If they do not, the cable cannot be cause.

 

[Deleted member 2457]

Clare Newsome:

I am getting so fed up of HDMI 'testing' that doesn't bring sound into the equation....

As you mention, our own "Big Question' feature where we asked three readers to blind test HDMI cables resulted in more noted differences being perceived in audio than video.

Personally i've participated in blind tests where we've experienced a difference in both picture and sound quality.

Same here

 

[professorhat]

Mr Mellie:Most of the answers are subjective so how about a more scientific approach?

Take a known source, say the fist 1GB of any film you like and save as a file on disk on computer 1.

Read the computer 1 file and send via HDMI to computer 2.

Computer 2 captures the data and saves to disk.

Now compare the file on computer 1 with that on computer 2.

Can not be that hard can it?

Finding two computers which communicate not only via HDMI cables, but which can also communicate using the same protocols that Blu-Ray players use to send data to TVs - that may be fairly hard, yes.

 

[idc]

Would it work to use a DVD to record to a PC using two different HDMIs, then measure the two files to see if there was any difference?

 

[hammill]

professorhat:Mr Mellie:Most of the answers are subjective so how about a more scientific approach?

Take a known source, say the fist 1GB of any film you like and save as a file on disk on computer 1.

Read the computer 1 file and send via HDMI to computer 2.

Computer 2 captures the data and saves to disk.

Now compare the file on computer 1 with that on computer 2.

Can not be that hard can it?

Finding two computers which communicate not only via HDMI cables, but which can also communicate using the same protocols that Blu-Ray players use to send data to TVs - that may be fairly hard, yes.

Indeed. Also, this method does not necessarily show if the data would have been displayed in a timely manner.

 

[The_Lhc]

It would have to be a highly customised PC that had an HDMI input (cannibalise an AV receiver perhaps) and could stream the raw data to disk (or SSD more likely, not sure a traditional HDD would be able to keep up). I'm not aware of any such device and it doesn't sound like something your average user could cobble together in a weekend, especially given that the PC would have to identify itself as an acceptable display device for the player to even talk to it(including HDCP, unless you meant record FROM a PC to DVD? But that almost certainly wouldn't be fast enough).

 

[v1c]

Found an interesting HDMI ebook pdf

Page 11 section 3.1 "It's just a cable isn't it" is were it starts to get interesting.

Yes you could make the case that is from a Cable store link however can you dispute the accuracy of the information ?

 

[Anonymous]

You guys may want to read this:

[EDITED BY MODS: Much thought they may, the links contravene House Rules]

The article demonstrates the difference between cheap and expensive cables at different lengths and bitrates (video resolution).

A resume for lazy ones: HDMI only carries ones and zeros, but a good cable will output the same ones and zeroes it had as input. Cables longer than 2m at 1080p, or short cables with higher than current 1080p bitrate may introduce errors. Even though most of these errors will be corrected by the receiver (proyector/TV), the image will differ from the original. Good error correction helps to hide the errors, but doesn't fix them.
Snow or no-image only happen in very rare situations.

They don't go into jitter.

 

[Anonymous]

there's some good ideas and info being aired here guys, do keep it coming

..

and we're on the third page and yet no argumentative posts so far, just good debate, thats great

 

[Anonymous]

for me the question that i'd love answered is how a supposedly inferior cable (in comparison to one deemed better) can seemingly lose data but not image consistency? if said data loss is significant enough to be noticed (remember, it's a supposedly inferior cable), and data loss is random, image consistency would be random, and noticed, to the same degree, imo..

if data loss is not random, i'd love someone to explain how or what, enables prioritisation of high value data whilst seemingly discarding data of a lesser value? as i assume would be neccessary to get a consistent picture, albeit one of lesser quality..

ta..

 

[idc]

Cant link (try googling HDMI test results) but another forum did a test and here is a summary
So HDMI is digital, it either works or does not. Not quite. You can get sparkles, lines or snow due to a partial degradation of the signal or a complete break down in the signal. In a test that took six months, 60 HDMI cables were tested to see what differences there are and how important quality and specification are. The cost of the equipment used to measure the cables was $205,000.

The measurements showed that as a HDMI cable gets longer it is more likely to fail. It did not show that as a HDMI cable gets more expensive it more likely to pass. Fail here meant that the eye pattern had been corrupted by jitter. In other words the 1s and 0s were not being transmitted clearly and become more of a blur. That can cause sparkles etc.

Then to some real world testing. Two of the worst performing cables were connected in to an AV system and, there was no sign of the fail any more. No snow, no crackles, no nothing. After connecting two cables together so they were 65 feet long, there was signal degradation. That was put down to the knowledge that such can happen and so the sender and receiver (say DVD and AV amp) are made to cope.

The final conclusion was that up to 4 metres you are fine, no matter what cable you use. Over 4 metres and it may matter, but there is no need to pay silly money.

 

[aliEnRIK]

If (as expected....no suprise there) blind testing is being brought to the table then we need to include blind testing in everything.

Has anyone blind tested their amp before they bought it? I put it to the group that all transistor amps sound the same

Blind testing has told us this. If people disagree then we must put away blind testing for this discussion. Its one OR the other. Not make up the rules to suite your answer

 

[aliEnRIK]

idc:
Cant link (try googling HDMI test results) but another forum did a test and here is a summary

So HDMI is digital, it either works or does not. Not quite. You can get sparkles, lines or snow due to a partial degradation of the signal or a complete break down in the signal. In a test that took six months, 60 HDMI cables were tested to see what differences there are and how important quality and specification are. The cost of the equipment used to measure the cables was $205,000.

The measurements showed that as a HDMI cable gets longer it is more likely to fail. It did not show that as a HDMI cable gets more expensive it more likely to pass. Fail here meant that the eye pattern had been corrupted by jitter. In other words the 1s and 0s were not being transmitted clearly and become more of a blur. That can cause sparkles etc.

Then to some real world testing. Two of the worst performing cables were connected in to an AV system and, there was no sign of the fail any more. No snow, no crackles, no nothing. After connecting two cables together so they were 65 feet long, there was signal degradation. That was put down to the knowledge that such can happen and so the sender and receiver (say DVD and AV amp) are made to cope.

The final conclusion was that up to 4 metres you are fine, no matter what cable you use. Over 4 metres and it may matter, but there is no need to pay silly money.

That same test goes some way to prove my theory -

2 of the cables failed 720P eye tests. And yet, they couldnt physically see anything 'obvious' onscreen. No sparklies or snow were visible, and yet errors were 'measureable'.

This can only mean errors were onscreen BEFORE sparklies etc appear as I originally posted

 

[Lee H]

idc:

What is about cable Superdooper that makes its blacks blacker than cable Niceynicey?

Ooooo, where can I buy these?

 

[idc]

aliEnRIK:idc:
Cant link (try googling HDMI test results) but another forum did a test and here is a summary

So HDMI is digital, it either works or does not. Not quite. You can get sparkles, lines or snow due to a partial degradation of the signal or a complete break down in the signal. In a test that took six months, 60 HDMI cables were tested to see what differences there are and how important quality and specification are. The cost of the equipment used to measure the cables was $205,000.

The measurements showed that as a HDMI cable gets longer it is more likely to fail. It did not show that as a HDMI cable gets more expensive it more likely to pass. Fail here meant that the eye pattern had been corrupted by jitter. In other words the 1s and 0s were not being transmitted clearly and become more of a blur. That can cause sparkles etc.

Then to some real world testing. Two of the worst performing cables were connected in to an AV system and, there was no sign of the fail any more. No snow, no crackles, no nothing. After connecting two cables together so they were 65 feet long, there was signal degradation. That was put down to the knowledge that such can happen and so the sender and receiver (say DVD and AV amp) are made to cope.

The final conclusion was that up to 4 metres you are fine, no matter what cable you use. Over 4 metres and it may matter, but there is no need to pay silly money.

That same test goes some way to prove my theory -

2 of the cables failed 720P eye tests. And yet, they couldnt physically see anything 'obvious' onscreen. No sparklies or snow were visible, and yet errors were 'measureable'.

This can only mean errors were onscreen BEFORE sparklies etc appear as I originally posted

I think that what that means is that the errors fail the test of transferring theory into practice. Cable makers make all sorts of claims as to why their cables are better, some of which contradict each other. They find a 'difference' and then say you will see or hear that difference. But the reality is that you will not. They are just using the power of suggestion, in other words, sales and marketing.

 

[v1c]

maxflinn:
if data loss is not random, i'd love someone to explain how or what, enables prioritisation of high value data whilst seemingly discarding data of a lesser value? as i assume would be neccessary to get a consistent picture, albeit one of lesser quality..

ta..

Wouldn't it be the same way as compressed data we watched before HD ?

 

[aliEnRIK]

idc:
I think that what that means is that the errors fail the test of transferring theory into practice. Cable makers make all sorts of claims as to why their cables are better, some of which contradict each other. They find a 'difference' and then say you will see or hear that difference. But the reality is that you will not. They are just using the power of suggestion, in other words, sales and marketing.

I disagree with what you think about the test

 

[Chewy]

This is an interesting discussion - and for once it has not degraded into a slanging match! . .Which is nice!!

. . .However given how hard this whole debate rages, not only on this forum but every where else, what still amazes me is that the creators of the whole HDMI design, tech, and specification have not come forward and confirmed definitively whether or not signal 'anomolies' in the way often described (blacker black, brighter colours etc) are physically possible within the design they have created?!?

These guys (and I don't know who they would be mind) will surely know if the construction of the cable beyond the basic specifications has technically any potential to impact on image or sound quality or not?

Why don't they speak up damn them!!

 

[v1c]

idc:. The cost of the equipment used to measure the cables was $205,000.

.

The cost of the equipment used to measure the cables is relevant ?

But apparently not the cost of the cable.

Did either the input source or the output source have any bearing on the delivered audio/video signals in said test ?

 

[Anonymous]

aliEnRIK:
If (as expected....no suprise there) blind testing is being brought to the table then we need to include blind testing in everything.

Has anyone blind tested their amp before they bought it? I put it to the group that all transistor amps sound the same

Blind testing has told us this. If people disagree then we must put away blind testing for this discussion. Its one OR the other. Not make up the rules to suite your answer

rik, do you have any views on whether or not data loss is random?

and if you believe it is not, then i'd love to hear your views on how it is prioritised?

ta...

 

[Anonymous]

v1c:maxflinn:
if data loss is not random, i'd love someone to explain how or what, enables prioritisation of high value data whilst seemingly discarding data of a lesser value? as i assume would be neccessary to get a consistent picture, albeit one of lesser quality..

ta..

Wouldn't it be the same way as compressed data we watched before HD ?

possibly, i don't know, and possibly similar to compressed hd video, similar to what one might find in an mkv file?

 

[Anonymous]

aliEnRIK:
If (as expected....no suprise there) blind testing is being brought to the table then we need to include blind testing in everything.

Has anyone blind tested their amp before they bought it? I put it to the group that all transistor amps sound the same

Blind testing has told us this. If people disagree then we must put away blind testing for this discussion. Its one OR the other. Not make up the rules to suite your answer

i think the subject of blind testing has been exausted on this forum, we are all aware that a very large majority of blind tests of hdmi cables find no significant differences..

the crux of my op was whether or not data loss is random, and the consequences..

if we can find that out then we can dig deeper, i agree, discussing blind tests is going off track..

edit.. i don't think it's fair to suggest idc is making up rules to suit his answer rik..

..

 

[Anonymous]

why do we need to do a blind test ...

surely with some planning, a test could be where a "cheap" decent - say fiver HDMI cable was used, and then say a real top end £100 one.

then with a high-end DSLR - with all settings set to manual to ensure consistancy, take a picture of both, and one should be able to show the difference for all to see ?

if its night and day as some people suggest, even if its a fairly subtle difference a decent camera should be able to pick up the difference.