Jump to content
disolusiond

HDMI cables

Recommended Posts

I just got TiVo for my birthday and I want to connect it to my 40 inch Bravia with a HDMI cable, the only frik'n cable it didn't come with. So I call WoW sight & sound to get a price on a cheap HDMI cable and the salesperson I speak with starts asking me what size TV I have, is it LCD or Plasma, is it full HD etc. I tell him it's a full HD 40 inch Bravia to which he replies "well you could buy a cheap $30 cable but I wouldn't, that's like putting retreds on a ferrar. What you need is one of the $80 cables".

 

Now this sounds like salesman on commission bullshit to me but I know not of these things. So you tech savvy bunch, do I need the $80 cable and could someone give me a better explanation than this "putting retreds on a ferrari" bullshit.

Share this post


Link to post
Share on other sites

Just buy a $20 cable. I must be a retard but I cant tell the different.

Share this post


Link to post
Share on other sites

$20 one, preferably from another store which isn't run by airheads.

 

LOL. Good to know, I thought it sounded sus. Got a mate at EB who can do me one for $15. Thanks for the speedy replies both of you, I can now go get my cheap cable.

Share this post


Link to post
Share on other sites

5 bucks from ebay, still the same thing.

 

Digital signal cable, so shielding and gold connecters = dick

Share this post


Link to post
Share on other sites

I'd be interesting to hear the salesperson's justification for the $50 additional cost.

 

Yes, cables do vary in their quality. And they deteriorate with age. I discovered this recently with a couple of my antennas.

 

At home I have two antennas for the 2m amateur band (144-148MHz), one vertical groundplane, which I feed using N-type terminated HDF400 coax. HDF400 is about as thick as one's index finger, and this 3m length cost me around $50 or so… can't recall the exact figure. It was expensive nonetheless. On 2m, with this coax, that 2m groundplane screams out, as this type coax has practically no loss at VHF, and is pretty good at 2.4GHz. (I originally bought it for my Brisbane Mesh node.)

 

I regularly participate on a 2m SSB net. For FM, the standard antenna polarisation is vertical, so the groundplane works well. On SSB however, the standard is horizontal, and so there's a 6dB cross-polarisation penalty.

 

Idea: stick a dipole up on my 9m squidpole — it'd be up nice and high then and correct polarisation. All I had was a 10m long length of RG-58 with BNC terminations, that used to pass Ethernet frames in a past life. No worries, I wound that around the squid pole up to the antenna, and gave it a try one Thursday evening (144.300MHz 7:30PM if anyone is interested, there's a net then run by Rob VK4ZDX).

 

Signal report was S4/Q5… yes it was making the trip to Ipswich (from The Gap), but normally I'm a S8/Q5 or better signal. Switched to the vertical and discovered just how poorly that antenna was doing. I had previously used said antenna at JOTA to make contact into a repeater at Mt Crosby (from Robertson Scout Den) so I knew the antenna worked, first suspicion was the coax.

 

I wound up spending about $50 on a 30m roll of RG-195. Not as good as say RG-213 or even the HDF400, but at least it wouldn't be so heavy as to make the squid pole bend over, and was better than the RG-58. I cut off a length of it, terminated one end BNC for the antenna, and the other PL-259 to suit the radio and gave it a try that Thursday evening. Signal report, S9 +10dB/Q5. Much better.

 

Relating this back to HDMI. Three factors are at play here:

  • The bandwidth of the signal
  • The centre frequency
  • The length of cable being used

All cable incurs loss. The loss comes about due to the capacitive coupling between the conductor(s) and the shield, and the inductance in the conductors themselves. In a coax cable, the capacitance will appear to connect the centre conductor to the shield. In multi-core cables, it does this and can connect to each and every other conductor in the cable unless they're individually shielded (making the cable like a bundle of coax cables — extra cost).

 

Shunt capacitance and series inductance is a recipe for a low-pass filter. Hence, loss increases with frequency.

 

Posted Image

The longer the cable, the more of those you've got in the circuit, hence the loss increases with length too.

 

Better constructed cable will incur less loss per metre than a cheaper cable. HDMI signals, being digital and high definition are very broadband signals, and likely are modulated at frequencies well into the UHF (and likely SHF).

 

The difference between a cheap cable for short lengths, and a long cable is very small however.

 

Play around with http://www.timesmicrowave.com/cgi-bin/calculate.pl and that might give you an idea.

 

For a 10GHz signal: 3ft of RG-58 gives a loss of 1.7dB. The same signal passed through the same length of LMR400 gives a loss of 0.4dB. 1.7dB for most purposes isn't that noticable, and so for that short a run, I'd be sticking to the much cheaper RG-58. For a long distance though, I'd go with the higher quality cable. I only use HDF400 on the vertical 2m antenna here because I had it laying around.

 

What HDMI will tolerate and the loss per metre for each cable, is unknown. Likely the salesperson won't know and the manufacturer won't say, but it's this rule of thumb. If it's going to be a longish run of cable, go for the higher quality cable, otherwise, likely the cheaper stuff will work fine.

Edited by TheFrunj

Share this post


Link to post
Share on other sites

Wow. Thank you for taking the time to pass on all that info Redhatter, very helpful. I only needed a 1.5m cable so the $15 one I ended up getting suits me fine. I compared it against a 1.5 Monster brand HDMI cable I use with my PS3 that I was foolish enough to spend $90 back when I purchased my tv and there is absolutely no difference in picture or sound quality.

Share this post


Link to post
Share on other sites

Well, no you can't because a single coat hanger wire wouldn't be enough to provide the signal and a reference voltage. You'd need at least two, then you've got to keep the buggers apart to prevent shorting.

 

Not to mention the fact that HDMI cables have several pins on the connectors with terminals fine enough to make termination a royal pain. ;-)

 

But yes, 1.5m I think even the cheapest possible cable will perform satisfactorily.

Share this post


Link to post
Share on other sites

I'd be interesting to hear the salesperson's justification for the $50 additional cost.

 

Yes, cables do vary in their quality. And they deteriorate with age. I discovered this recently with a couple of my antennas.

 

At home I have two antennas for the 2m amateur band (144-148MHz), one vertical groundplane, which I feed using N-type terminated HDF400 coax. HDF400 is about as thick as one's index finger, and this 3m length cost me around $50 or so… can't recall the exact figure. It was expensive nonetheless. On 2m, with this coax, that 2m groundplane screams out, as this type coax has practically no loss at VHF, and is pretty good at 2.4GHz. (I originally bought it for my Brisbane Mesh node.)

 

I regularly participate on a 2m SSB net. For FM, the standard antenna polarisation is vertical, so the groundplane works well. On SSB however, the standard is horizontal, and so there's a 6dB cross-polarisation penalty.

 

Idea: stick a dipole up on my 9m squidpole — it'd be up nice and high then and correct polarisation. All I had was a 10m long length of RG-58 with BNC terminations, that used to pass Ethernet frames in a past life. No worries, I wound that around the squid pole up to the antenna, and gave it a try one Thursday evening (144.300MHz 7:30PM if anyone is interested, there's a net then run by Rob VK4ZDX).

 

Signal report was S4/Q5… yes it was making the trip to Ipswich (from The Gap), but normally I'm a S8/Q5 or better signal. Switched to the vertical and discovered just how poorly that antenna was doing. I had previously used said antenna at JOTA to make contact into a repeater at Mt Crosby (from Robertson Scout Den) so I knew the antenna worked, first suspicion was the coax.

 

I wound up spending about $50 on a 30m roll of RG-195. Not as good as say RG-213 or even the HDF400, but at least it wouldn't be so heavy as to make the squid pole bend over, and was better than the RG-58. I cut off a length of it, terminated one end BNC for the antenna, and the other PL-259 to suit the radio and gave it a try that Thursday evening. Signal report, S9 +10dB/Q5. Much better.

 

Relating this back to HDMI. Three factors are at play here:

  • The bandwidth of the signal
  • The centre frequency
  • The length of cable being used

All cable incurs loss. The loss comes about due to the capacitive coupling between the conductor(s) and the shield, and the inductance in the conductors themselves. In a coax cable, the capacitance will appear to connect the centre conductor to the shield. In multi-core cables, it does this and can connect to each and every other conductor in the cable unless they're individually shielded (making the cable like a bundle of coax cables — extra cost).

 

Shunt capacitance and series inductance is a recipe for a low-pass filter. Hence, loss increases with frequency.

 

Posted Image

 

The longer the cable, the more of those you've got in the circuit, hence the loss increases with length too.

 

Better constructed cable will incur less loss per metre than a cheaper cable. HDMI signals, being digital and high definition are very broadband signals, and likely are modulated at frequencies well into the UHF (and likely SHF).

 

The difference between a cheap cable for short lengths, and a long cable is very small however.

 

Play around with http://www.timesmicrowave.com/cgi-bin/calculate.pl and that might give you an idea.

 

For a 10GHz signal: 3ft of RG-58 gives a loss of 1.7dB. The same signal passed through the same length of LMR400 gives a loss of 0.4dB. 1.7dB for most purposes isn't that noticable, and so for that short a run, I'd be sticking to the much cheaper RG-58. For a long distance though, I'd go with the higher quality cable. I only use HDF400 on the vertical 2m antenna here because I had it laying around.

 

What HDMI will tolerate and the loss per metre for each cable, is unknown. Likely the salesperson won't know and the manufacturer won't say, but it's this rule of thumb. If it's going to be a longish run of cable, go for the higher quality cable, otherwise, likely the cheaper stuff will work fine.

 

Bloody hell mate. You would have to be one of the most helpful people that have ever been here.

Awesome stuff.

Keep up the good work. People like you continue to make the tech forums a great source of info for us all.

 

*takes hat off to redhatter*

Share this post


Link to post
Share on other sites

Redhatter, if you're going to include images, please don't make the background transparent. With black on black, I'm sure I'm not the only one who just assumes you have included arbitrary spaces between paragraphs.

Share this post


Link to post
Share on other sites

Well, I can't help others' colour schemes. The image is from the Wikimedia Commons, so I don't control the colouring.

 

Sadly, there's no way to tell the forum software to put a white box behind it either even though it's possible in HTML.

Share this post


Link to post
Share on other sites

Well, I can't help others' colour schemes. The image is from the Wikimedia Commons, so I don't control the colouring.

 

Sadly, there's no way to tell the forum software to put a white box behind it either even though it's possible in HTML.

Just putting the image in quote tags makes it visible enough.

Share this post


Link to post
Share on other sites

Ah, sorry, didn't check the source of this one.

 

But yeah, putting it in quote or code tags is enough (to know it's there, anyway!)

Share this post


Link to post
Share on other sites

yeah all well and good, but the elitists will tell you its digital...you can just use coat hangar wire...

err, you have that backward. the Elitists will tell you "Its terrible if you dont have the BEST cables, man, what are you doing?"

 

people wil COMMON SENSE will tell you 'any will do'.

 

As i've written several times. Digital is a lot more 'simple' than analogue, in that the data consists of 0 or 1, no 'frequencies' etc. And a slong as the signal reaches the other end, its fine. even if it drops a bit of data, most things can 'guess' what it was and you often wont even see.

 

Eg.

CLEAN: 10101010101010

DIRTY: 10I0|olO1@10|0

LOSSY: 10101010_01010

 

The clean is obviously easiest to read, the dirty is 'messy' but, just like a 'weak signal' in digital, still pretty damn easy to 'guess' which is a 1 or a 0, and the lossy has missed a '1' as easily noticible by a pattern.

 

it then takes this digital pattern and makes an image, sound, whatever. There is no 'quality' there is no 'sharper image' from better cables. As long as its good enough for each end to 'get the jist' (without losing data) it'll work. perfectly.

Share this post


Link to post
Share on other sites

I sell both $20 and $90 HDMI cables.

 

The way I tend to explain it to customers is like transferring a photo from your camera to your computer.

The cable has no impact on the quality of the image.

 

The only reason I suggest the more expensive cables are for longer lengths.

Share this post


Link to post
Share on other sites

For anything under 1 metre, any cable will do.

 

1-3 metres, I'd just want to make sure the thing is going to stand a bit of pulling and twisting. Basically this is a connector thing rather than a shielding thing.

 

pretty much the next step after than is 10+ metres, in which case I would recommend the "better" or "best" cables as signal drop-off becomes an actual problem at those lengths, and typically they'll be pulled through a wall or ceiling, or on the floor where the durability of the shielding will be a factor.

 

 

Although there's no disputing that there is a difference between "good" and "shit" cables, the fact is that this only becomes actually apparent either at long run-distance or when the connectors are shite.

Share this post


Link to post
Share on other sites

As i've written several times. Digital is a lot more 'simple' than analogue, in that the data consists of 0 or 1, no 'frequencies' etc. And a slong as the signal reaches the other end, its fine. even if it drops a bit of data, most things can 'guess' what it was and you often wont even see.

That's a complete load of nonsense. Fourier would be spinning in his grave reading that!

 

The only signal that has "no frequencies" is flat plain DC with absolutely all noise filtered out. Everything else, has a frequency, and anything more complex than a theoretical sine wave will have a bandwidth. Ergo, all signals have a bandwidth.

 

In the case of the "digital signal" you described earlier… it's a whole heap of these concatenated together, correct?

 

Posted Image

 

That, dear viewers, is the Gate function or Rectangular function. It's comprised of two Heaviside step functions subtracted from each-other, one delayed relative to the other. The width (in seconds) is the reciprocal of the frequency (in Hz). So a repeating signal where each one of those pulses is 500µs wide, spaced apart 500µs (thus: 1ms period) equates to a 1kHz square wave.

 

Take the Fourier transform of that, or set up an oscillator and throw the resulting signal at a spectrum analyser, and you'll see this:

 

Posted Image

 

Ohh look at that, not just one frequency, there are thousands of them. Lots of juicy harmonics too. In fact, it's for this reason that high-speed signals are never sent this way. The pulses are shaped to more closely resemble sine waves.

 

Raised Cosine filters are one such example of this. GSM mobile phones use Gaussian Frequency Shift Keying, if you were to "tune in" on the FM carrier generated by one of these, you'd see the pulses have a Gaussian shape to them. There are numerous of others too, some use a truncated/delayed version of the normalised sinc function shown above.

 

Yes, you're right that it's digital, but all signals with any interesting content occupy frequencies and bandwidth. And no, it's not more simple than analogue, it's just easier to filter noise out than with analogue.

Edited by Redhatter

Share this post


Link to post
Share on other sites

Redhatter is correct.

 

It was unwise of me to simplify things so far and generalise so much.

 

All i was trying to say really, is that Digital is interperated as 1's and 0's. As long as each end can tell which one its meant to be, it'll be OK. The rest is complete bunk and was being used as an illustrating piece rather than scientific fact.

Share this post


Link to post
Share on other sites

The simplest test of a cable for digital video is whether it works or not, it will work or it won't

 

I remember seeing gold plated TOSLINK cables at jb-hifi when i was looking for some, what a wank. They cant possibly have any bullshit to cover that

Share this post


Link to post
Share on other sites

Gold plated tos..... ow my head!

I'll have you know that gold plated TOSLINK cables inspire an Electromagnetic Propulsion EffectTM which accelerates data up to 5x faster than a standard TOSLINK cable. Also, it's shiny.

 

>_>

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×