Expensive HDMI cables make no difference - the absolute proof

Comprehensive, proper testing answers the question once and for all

Michael
Passingham
22 Jul 2015

We’ll admit it: we’re a little annoyed. A few years ago we sat down and definitively proved that expensive HDMI cables don’t make any difference (see the Original article page in this article). We seem to have helped stop the deluge of pointless HDMI cable reviews online, but we don’t seem to have completely won the war yet. Despite our scientific testing, we’ve seen editorial in Hi-Fi magazines extolling the virtues of HDMI cables and staff explaining how they can see and hear differences with different digital cables. We’ve even had our work cited by a reader who filed a complaint with the press complaints commission against a magazine over its coverage of digital cables. The magazine's lawyer’s response was that our original testing didn’t use motion (we had the video paused), therefore the results were invalid. It’s a stupid technicality, as paused video outputs as many frames as normal video (the screen is refreshed as often), but the complaint wasn’t pushed.As a result, we’ve decided to revisit this subject for the absolutely final time, testing moving video. To make sure we cover absolutely everything, we’ve split this article into two sections: the theory of why HDMI cables don’t make a difference (a FAQ of common questions), which includes the science of why HDMI cables don’t make a difference, and the testing proving that HDMI cables don’t make a difference, which we've put on page two of this article.

FAQ - Why don’t expensive HDMI cables make a difference?

Q. HDMI cables can make a difference can’t they – reviewers have noticed that grass is greener and flesh tones are better between models?A. This is impossible and suggests that an HDMI cable has intelligence built into it. As a result, the reviewers are at best delusional and, at worst, lying to you. Inside, an HDMI cable has 19 individual wires connected to 19 pins, each designed for a specific job, but these are effectively just bits of metal designed to conduct an electrical signal. It’s important to note that there’s absolutely no processing in the cable and, as far as the wires go, they could be carrying a picture, some audio or anything else.For an HDMI cable to make flesh tones better, for example, it would have to decode the video signal, process where people are and then tweak the image all before re-encoding it. Think about a cable that can apparently make foliage better – how would it cope with Kermit the frog standing in a green field? Would it make his green tones better, or would it be able to discern the grass and just make that better? Of course, the answer is neither, as cables don’t have processors. ^If HDMI cables can improve foliage, how can they tell the difference between the trees, the blankets and Kermit?Think of it another way, if HDMI cables can improve the quality of a picture, could a more expensive SATA cable make a word document better to read? Imagine Dan Brown buying a £2,000 SATA cable and opening up his latest novel to find out that it had been transformed from his usual quality into Dickensian beauty. Of course, that’s nonsense and we’re stuck with Dan Brown’s prose and HDMI cables that can’t touch image quality.Q. If HDMI cables don’t have any processing in them, how comes you can buy active cables, which can only be plugged in one way?A. Active cables draw power from the HDMI port to power a signal booster. This helps with longer cable runs (say more than 5m) or lets you use a thinner HDMI cable. Some of these active cables have to be plugged in one-way round to work properly, as plugging them in the ‘wrong’ way round will introduce some errors. Note that an active cable doesn’t boost image quality but cuts down on errors.Q. So reviewers definitely can’t ever see differences?A. Differences can be seen in testing, but this is down to the kit used. Every TV will interpret the signal slightly differently, displaying different colours. Every Blu-ray player outputs a slightly different picture, too. This is the reason that calibration is recommended. However, it’s definitely not the cables causing these issues.Q. You’re saying that all HDMI cables are identical and one can’t be better than another?A. We’re not. The issue we have is with the word ‘better’. This implies that a one HDMI cable can make an improvement over another, which it can’t. With a digital signal, everything is sent as 0s and 1s. You get an error, if a 1 is received as a 0 or a 0 is received as a 1. In this way, an HDMI can either correctly transmit everything or it can introduce errors. It’s technically possible that a poor HDMI cable will transmit more errors than an expensive one.Q. So errors do mean differences and you can see them, you’re backtracking aren’t you?A. Certainly not. We freely admit that a digital signal will occasionally cause an error in transmission, but this has to be put into context. The HDMI standard allows for one error per one billion bits, which is known as the Bit Error Rate (BER). Assuming that a picture is transmitted using 24-bit colour (8-bits each for RGB) at 24fps at a resolution of 1,920x1,080 (2,073,600 pixels), that’s a total of 49,766,400bits per frame. At 24fps, the film standard, that’s 1,194,393,600bits (1.1bn bits). In other words, the HDMI standard allows for the worst cable to have a single error in one pixel, in one frame per second. You’re going to have to have incredible eyesight in order to spot that mistake. Besides the one-bit error could occur in the HDCP copy protection or the audio track, which would mean that you wouldn’t even see it. Finally, the video picture has error detection to look out for these kinds of things.Q. Ah, but HDMI video doesn’t have error correction, does it? So, there could be differences in the picture.A. Assuming that your one-bit error causes a problem in the picture and one pixel is incorrect, error detection on the TV will let it know where the problem is. The TV can analyse the surrounding pixels and make an educated guess as to what colour the error pixel should be. So, you may get one pixel per second that is ever so slightly the wrong colour. If you can spot that, you have the best eyesight of anyone that has ever lived. Well done you.Q. What about HDMI standards and certification, surely different cables make a difference?A. Although there are different HDMI standards, with HDMI 2.0 being the most recent, there are only two HDMI cable standards: Standard and High Speed. Standard is out-of-date now and supports lower resolutions; High Speed supports everything, including the HDMI 2.0 standard, which gives you 4K TV at 60fps (Ultra HD).Q. What about audio, though; you said that you could get errors here, too. Surely that can make a difference?A. Again, the one-bit in a billion steps in here. Realistically, it means that the occasional error will pop into the audio track. This kind of error could cause a noticeable pop or blip if it wasn’t corrected; fortunately, audio has error detection and correction. This means the receiving equipment can detect an error and correct it, as though it never happened. In other words, there is no longer an error. In the event that there are too many errors you simply won’t get any sound, as receiving kit is programmed to shut down audio rather than output potentially damaging and irritating sounds.Q. What about jitter? HDMI’s really bad for this, right?A. Jitter is the described as the deviation from true periodicity of a presumed periodic signal, which probably doesn’t mean a lot to most people. In simpler terms it's describing how a signal might not be properly in sync. This is because all digital data uses a clock to synchronise transmission, with each clock cycle (a tick if you will) used to send a bit of data. How the data is sent differs from system to system, but a simple view is that when there’s data, the voltage rises to maximum; when there’s a zero, voltage falls to 0. When plotted on a graph, you get a square graph with the values going up and down.When the clock is accurate, each bit is sent at a perfect time interval, but the timing is often not as accurate as you may think. For example, rather than sending data every second, you may find that the first signal goes out at 0.99s, then 1.12s then 1.05s. We’ve exaggerated the example to make a point, with digital transmissions occurring much faster and with less error, but it shows how timing affects the signal. ^In the graph at the top, the digital signal is sent perfectly with the vertical lines matching up with clock (the dotted lines) completely; the graph at the bottom shows what happens is the clock isn’t so regular, with the graph distorting.With jitter distorting the transmission, two things can happen. First, repeated errors can make the receiving device believe that is has a 0 instead of a 1 (or vice versa), introducing an error. As we’ve discussed, these errors are corrected to the point where you don’t notice them for both audio and video.Secondly, if you’re outputting in real-time (or close to that), the mistiming can mean problems with audio, as a sound may occur too early or too late, distorting the analogue waveform as it's converted from digital. However, receiving devices buffer some audio to help eliminate these issues and minor differences in timing can’t be heard.More importantly, as far as this article is concerned, it’s not the HDMI cable that’s at fault for jitter, but the HDMI standard. In other words, the HDMI cable can’t and doesn’t make any difference to jitter. A bigger difference is made by the quality of the Digital-to-analog Converter (DAC), which takes digital sounds and converts it back to the analog sounds we hear, but even this pales in comparison to the quality of your speakers and AV receiver.

1435256222257Testing cables

Scientific explanations are all well and good, but it's practical testing where the talking stops and the evidence starts. To prove the doubters wrong, we upped the ante and decided to test full motion video to prove that changing cables makes no difference. In order to test scientifically, we turned to our Digital Foundry TrueHD card, which captures the RAW and uncompressed HDMI signal. Crucially, it performs no error correction, so we can accurately compare the output from different cables and spot any errors.As an input, we connected a laptop via HDMI to the TrueHD card. We set Windows to use a resolution of 1,920x1,080 with 24-bit RGB colour, at a refresh rate of 50Hz (PAL). Note that the playback device doesn’t make any difference, as the HDMI signal is the same regardless of whether it’s a Blu-ray player, games console or PC. For test footage, we used the open-source film Tears of Steel. As our PC was set to 50Hz, the video played back at 25fps (25p), which is the PAL standard.In order to compare different cables, we needed to ensure that the captured footage was the same regardless of the cost of the cable. To do this we used the TrueHD capture card to capture a few seconds of Tears of Steel for each cable, saving the results in an uncompressed video file. We then used a frame grabber tool to take the same 50 frames (two seconds’ worth) from each captured video file, saving them as uncompressed BMP files. To ensure that we captured the same frames, we started on the first frame of the spaceship taking off; it’s easy to spot as the frame before this is completely black.Once we’d got 50 frames from our range of test cables, we needed to compare them. Our rational was that if frame one from HDMI cable one was the same as frame one from HDMI cable two (and so on), then the cables had made no difference. In order to compare the files we generated an MD5 hash of each image, which can be thought of as a digital fingerprint, as is represented as a 32-digit hexadecimal number, such as 6add8f2f6edee97b042fcb629fe2f5e7. In short, if the MD5 hash of two files is the same, the files are forensically identical.In the event that we found any captured frames that were different, we had a second method to check for differences, ImageMagick’s Compare tool. This tool is designed to mathematically and visually show the differences between two files. It creates a new image, where pixels are colour red to show that there’s a difference between the two images. This allows us to see where the error is.Our first comparison was between an expensive 1m HDMI cable that costs £70 and a 1m cable that costs £4.99. In our first set of tests, 49 of the 50 frames we captured from the expensive cable were completely identical to the 50 produced from the cheap cable. With the one frame that was different, we got two completely different MD5 hashes.

Cheap cable MD5 Expensive cable MD5
6add8f2f6edee97b042fcb629fe2f5e7 bc5de71326a6025410e710b1e25ae300

We needed to find out where the error was, so we used Compare to highlight what the error was and found that there was a one-pixel error. To see if you can spot where the error is take a look at the images below and see if you can spot the difference: the top image is from the cheap cable, the middle image is from the expensive cable and the bottom image is from the expensive cable (click any image to view it full-size).Hard to tell, right? If when we crop in to the image (see below) to show the error, spotting a single-pixel error is incredibly hard. Look at the top-left of the right-hand image and look for the red pixel. This is the error, which is the difference between the cheap cable (left) and expensive cable (right). As if to prove our point about cable quality not making a difference, the single pixel error we found wasn't from the cheap cable's output; it was a dark grey spot on a single frame of the expensive cable's output. Keep in mind that this is a single pixel amidst two million other pixels, displayed for 1/25th of a second. This is barely noticeable in the extreme circumstances we set up in our Labs, so the likelihood you'll spot it on a TV that has built-in error correction is next to zero. What's more, one frame in 50 is well within the error allowance of the HDMI standard.Next, we decided to compare a cheap 5m cable (£5.99) versus an expensive one (£130), as the argument is that cheap cables suffer over length. This time around we didn’t find one error on either cable. Throughout all of our testing, mathematically and scientifically, bar that one-bit error, we didn’t see any differences between expensive cables and cheap ones, proving that price doesn’t influence quality.

Cable care

Doing a bit more testing, we found that the only reason you may want to pay more for an HDMI cable is for better build quality. This isn’t because out-of-box you’ll see any difference, but because a more expensive cable will be more rugged and have a better-made connector. As proof, we tested a few cables that had been hanging around for a few years and had been rather mistreated and stored badly. These cables were showing signs of physical wear and, when we plugged them into our capture card with no error correction, there was an extreme number of errors in every frame. As such, this cable would no longer pass the HDMI certification, and so it is effectively broken.Dozens of sparkling, coloured and white dots dancing across the screen showing just how badly damaged the cable was. However, even a cable in such poor repair would be hard to spot on a TV, as error correction will make those pixels impossible to spot even if they are technically wrong.Even so, this gives us two important bits of information. First, treat your HDMI cables with care and make sure that connectors are securely plugged in and there’s nothing putting pressure on them. Secondly, if you're buying an HDMI cable for a piece of kit that you move around a lot, such as a projector or a laptop you're constantly unplugging and plugging back in, it will pay to buy a cable with a more rigid cable and a stronger, less flexible joint between the cable header and the wire itself.

Conclusion

We were told that we hadn't tested for motion when we last ran our test, but we've since done that and still proved that there's no difference in quality between expensive and cheap HDMI cables. Our one concession on this matter is that more expensive cables have better build quality and are more rugged, so are a better choice if you're constantly unplugging a lead. Even so, there's no need to go crazy and spend a fortune. We're also happy to run a blind test for anyone that still believes that they can see a difference, although we don't think that anyone will be brave enough to do this.1437556218009This is the original article that we published, when we first tested HDMI cables. When we did our testing in 2012 we used a paused video for video capture, which some people said was cheating and didn't prove anything. While our new testing proves the same thing, we've left this article up, as it goes into more detail about the science of HDMI.Whether or not to buy an expensive HDMI cable is one of the most contentious subjects on the internet. On pro the side of buying them are home cinema magazines and audiophile websites, which sing the praises of more expensive cables, heralding how they can do remarkable things, such as boosting flesh tones.On the other side, you have the sceptics that say HDMI uses digital, so all cables are equal. In fact, James Randi has even extended his $1 million prize for anyone that can prove paranormal activity to include anyone that can prove that expensive HDMI cables make a difference.We've always been sceptical about the benefits of more expensive cables, but the problem was always proving it in a scientific, repeatable way that removes any argument. Now, we've found the way to do that and can once and for all prove beyond a doubt if expensive cables make any difference.The only way to answer the question of whether or not expensive cables really make a difference is to test image quality objectively. Subjective testing using a group of people is inherently flawed, as the average person will almost always pick a side when given two options to choose from – even if they are identical. HDMI cables are no exception to this rule.Using identical kit is no guarantee of identical results and subjective testing is inherently flawedThere are other problems, too. Using identical TVs, Blu-ray players and discs still can’t create a level playing field, because of minor differences between the colour casts of each set. Even identical devices have the potential to affect the results of a blind test. When we compared two PlayStation 3 consoles, we noticed a distinct difference in the colour reproduction of a Blu-ray disc - independent of the TV or HDMI cable we used. Unless you colour calibrate each device first, a blind test can’t produce reliable results.
1437556199336

Definitive testing

To objectively test HDMI cables we first bought the Digital Foundry TrueHD capture card. This high-quality card allowed us to capture the HDMI output from a device in a RAW uncompressed format with no error correction, using 24-bit RGB. This eliminates any colour cast or error correction from the equation, as we're dealing with raw frames captured.Next, we used a PC to output the Blu-ray version of Sintel, an open source film produced by the Blender Foundation. The choice of a PC may seem a little strange, and some people would argue that we should have used a Blu-ray player instead. However, we're not interested in the argument of which device produces the best-quality video, but rather if HDMI cables produce different results. To this end, the PC gave us one important advantage: we could screenshot a frame from the film on the PC and compare this to a frame captured on the TrueHD card. If the two were the same, the cable had made absolutely no difference; if the two differed, the cable was affecting quality.To ensure that we managed to get the same frames at the PC and TrueHD ends, we paused the film and jumped to chapter two. It may seem strange to pause the film, as we weren't capturing motion, but as far as the PC's graphics card, cable and TrueHD card are concerned, pausing has no effect. The PC still has to output at the same rate, but rather than outputting 24 different frames per second for motion, it's outputting 24 identical frames per second.As an HDMI-certified cable allows for a maximum of 1-bit error in a billion (roughly one pixel incorrect per second), we captured 24 frames of the paused Sintel (one second's worth) and saved them as uncompressed bitmap files that we could then compare to original frame we'd screen-grabbed on the PC. As the TrueHD doesn't apply any correction to the source, a single pixel's difference would be highlighted.Sintel is an open source Blu-ray film, which we used to test if HDMI cables make any difference at all

Compare and contrast

Once we had our captured frames, we needed an objective way of comparing the captured frames to the original frame we'd captured on the PC. For this we decided to use two methods. First, we used the ImageMagick Compare command. This takes two images and creates a third picture, highlighting any pixels that are different in red. Next, we created a MD5 hash of each file we captured. An MD5 hash is a unique fingerprint of file: if two files have the same MD5 hash, they're physically and scientifically identical, no argument.By also creating an MD5 hash of the ImageMagick-created compare files, we could prove that they're all identical, too. In essence, then, we ensured that our captured frames were identical and that our comparison frames were identical.ImageMagick's Compare command can detect a single pixel error and highlight it.In both of our verification tests, a single shade of colour in a single pixel will be enough to trigger a difference. To test this level of accuracy, we edited a screenshot in Photoshop, changing the red colour value of a single pixel by one point before running our comparison tools. The red highlight in the comparison image and different MD5 hashes indicate a difference in the two images.
1437556199525

Proof that image quality can't be improved by HDMI

This method of testing not only determines the number of errors transmitted by a particular cable, it will also indicate if colours change at all when switching between cables - an idea frequently used to explain why certain cables are “better” than others in high-end home cinema reviews. Basically, if the cable does anything at all, the captured frame will be different from the frame screencaptured on the PC.We looked at a range of cables, from the bargain basement cables that come bundled with Blu-ray players and TVs, to incredibly expensive high-end ones with huge feature lists covering the packaging. There was a vast difference in terms of price, ranging from £5 to an eye-watering £150, which could easily confuse anyone looking for a new cable.The list of cables that we used for our testing

Absolutely no difference

After methodically testing each cable, our comprehensive results revealed the truth. We couldn’t find a single error in any of the frames of video we captured using our tests, regardless of the cable we used. Even allowing for a single error per billion bits, none of our captured frames contained as much as a single pixel out of order by a single colour shade.This image is the ImageMagick Compare-created image. If there were any errors, the pixels would be highlighted in redThe MD5 hash values backed up our findings – the captured frames were identical to the original ones. As verification, the MD5 hash of the ImageMagick Compare files were also identical. For reference, here are the MD5 hash values generated.The MD5 hash proves that the captured files are all identicalWe had to force an error by reversing a uni-directional cable, which generated several incorrect pixels. Because our capture card doesn’t apply any error correction, these pixels simply failed to render.Comparing our captured images to one another also proves that individual cables have no impact on colour reproduction. Because each capture was identical, with no differentiation between the colours of individual pixels, there can be no question that the final picture displayed on a TV is unaffected by your choice of HDMI cable. With no active circuitry inside them, HDMI cables don’t know what format the data it transmits will get re-encoded into when it reaches its destination, so there’s no way it could adjust colour values, improve skin tone or create better black response.More expensive cables may have been engineered to eliminate timing jitter or reduce binary errors, but we didn't get any errors with our Lindy 10m cable, which costs less than the Atlas 1m cable. What's clear from our tests is that we got no errors on any of the cables, and more expensive cables did not improve image quality at all. Given our results, we'd definitely buy cheaper HDMI-certified cables.

Read more

In-Depth