If the old SCART lead was only partially wired or you were using something other than RGB (remember that SCART can also carry S-video and CVBS/composite) then changing to RGB is bound to make a difference.
I'm openly willing to accept than when unusually long distances are involved between equipment then there is indeed gain to be had by using better quality cables. What I'm not convinced of is any difference between a standard, fully-wired SCART lead that's a metre or so long behind your TV, and a product ten times the price with gold plated connections and teflon insulated oxygen free copper.
We'll all have to agree that A/V enthusiasts always have and will claim vast improvements in their system in line with the amount of cash spent, and that's fine. In my personal opinion, though, there comes a point where you're spending money on products that make absolutely no discernible difference to sound or picture quality.
Of course, if you're willing to spend insane amounts of money in the first place then you're bound to conceive a difference when you insert your new £1000 speaker leads. My bone is that no manufacturers, retailers or enthusiasts are interested in providing any meaningful science or testing to back up their claims. For example, it would be extremely easy for a manufacturer to set up a test rig with a PC based oscilloscope and a signal source, then plot the signals at both the source and receiving end of their cable. They could then compare this with a typical out-of-the-box SCART lead. If it's such an easy test then why don't they do it? Simple - they would prove that most (if not all) of the marketing ploys of their product(s) are, in fact, snake oil. Not to mention their excuses when blind tests prove the pointlessness of their products...
Isn't that the same quote that most AV enthusiasts use?
You're right on that one. Although the data is digital, the signal can become corrupted in one way or another between devices. A couple of out of place 1's and 0's here and there won't necessarily mean the all-or-nothing that people typically talk about when referring to digital signals. The sheer amount of bandwidth on a HDMI connection means that small amount of lost data would be hardly noticeable. However, it could be enough to cause artifacts on the screen.
The 'cliff effect' referred to by Flameport is referring more to radio signals such as digital terrestrial. This problem is caused because a digital tuner requires a set signal strength to be able to operate effectively. Once the signal strength falls below that level, it is no longer possible for the tuner to operate. However, this is NOT the same issue as I'm discussing above.
Again, my bone is when people claim improvements such as "much more vivid colour" when they've just parted with £200 for a HDMI lead. Following similar logic, does that mean that if I upgrade the network lead to my computer, I might be able to get a better picture on BBC iPlayer? Please...
Besides, for short leads such as those found in a typical home theatre setup, I still don't believe that upgrades to HDMI cables will make any noticeable difference whatsoever. The problem only comes with longer cables where capacitance and issues with signal timing/convergence can hack up the signal pretty bad. That's the same with any high bandwidth digital interconnect, and why cat5 has over time been superseded by cat5e, cat6, cat7 and even cat8, although I'm not sure the latter is a formal standard.
