HDBaseT has been around for a few years now and, to the likely disappointment of the initial HDBaseT consortium, appears to have settled into a role as a midpoint; it still isn't a format likely to be seen as the input or output to a device, but is widely used in distribution and switching. I had the chance to ponder and discuss this with some of my colleagues in the commercial AV industry at AMX's showroom during the two-day certification class for their HDBaseT system: Enova. It's striking to see how far this technology has come, and how different manufacturers can take the same underlying chip set to create very different-feeling solutions.
The idea for HDBaseT is a grand one; to have one and only one cable connected to each display. That cable, a standard network cable with standard RJ45 connectors, would carry uncompressed high-definition video, audio, control, and even power. Yes, they expect to power your display via the same Cat5 jumper that carries everything else. One cable in this case literally means one cable. In real world applications, that isn't what's happened. For one thing, native HDBaseT inputs and outputs. Even if manufacturers want to use this kind of system, even mid-size displays in the 50 to 65 inch range outdraw HDBaseT's upper power limit by an approximate factor of 10. So, like a phone company running fiber to the curb before converting to old-fashioned copper, we're left with easily-pulled Cat5e cables to a receiver-box giving way to good old HDMI cables for the proverbial last mile. We're not quite where we someday can be, but it's an improvement. Crestron's Digital Media is an HDBaseT solution, as is Extron's XTP and AMX's DXLink, the heart of its Enova solutions.
How do solutions compare? Not only does Enova have a different feel from Crestron, but AMX's DVX (a family of all-in-one presentation switchers) behaves differently from their DGX (more conventional digital matrix switchers). One thing I hadn't known - because I'm not in the habit of removing the case on expensive pieces of electronics unless I have to - is that while it looks like one seemless piece of hardware, the DVX is, in fact, card-based. My notes include a glimpse at its innards, highlighting the HDMI input cart swappable with a DXLink card for another model. It's a clever design philosophy which allows AMX to inexpensively and reliably use one platform to produce a suite of units with slightly different features.
So how does Enova compare to Digital Media? At a glance they're similar; each has a presentation-switcher with built-in control processor, audio mixing, and an amplifier. Each line boasts a modular matrix switch capable of handling different video formats. Each has a series of Cat5 transmitters and receivers. Each handles HDCP key authentication, effectively eliminating any risk of running out of keys in large systems. Looking a bit closer, one sees differences.
The first distinction - of which AMX is justifiably proud - is that each output on an Enova system has a built-in "smart-scaler" which reads the EDID from a display and scales the image to fit. This means that even in a large system with many different displays each device will get an image at its native resolution. Their contention is that other practices, like choosing "best common", not only leave adjusting resolution to a device's onboard scaler, but fails to take advantage of the higher resolution in the largest display in a system. It adds a certain measure of cost, but AMX feels that they give value for it.
The second distinction is that each AMX matrix switch has a control processor built into it. This makes for a neater and more compact installation, but isn't a tremendous improvement over the inclusion of a standalone processor, and burying it in the switch gives you the new problem of having no local control ports. Unless everything in your system has IP control, you'll have to build out the system with varying add-ons. In all fairness, AMX has a nice suite of IP-based control port expansion modules, and seems to have a philosophy of preferentially using IP-based controls.
What about the distinction within the Enova line? At Enova training in AMX's New York showroom we got a demonstration of both the DVX presentation switcher and DGX matrix switcher. The performance in switching speed is markedly different, with the DVX switching, I would estimate, twice as fast as its cousin. The difference was explained as a result of different engineering teams working on the two devices, and upcoming firmware updates to let the DGX catch up were hinted at. (it was further explained that the slower switch time - one and a half to two seconds - was a result of the switcher dropping sync when it changed sources. It is very interesting how the same hardware can have different performance given different firmware.
What about venerable switching manufacturer Extron? I don't have much to say about their XTP line; certainly until they start shipping the Cat5 input and output cards for their switcher, they can't be said to have a real solution available. Even given that, they're still behind the curve in not having a DMPS300/DVX style presentation switcher available. They have added power injectors to their XTP line, but it is, at the very least, unclear that these would work with other HDBaseT solutions.
Long story short? It's an exciting time in the commercial AV industry as we finally seem to have the hardware, software, and expertise to start making digital video work closer to the way it should; many of these solutions are, if anything, easier to use and more flexible than old analog solutions.
A side note: my adventure at AVI-SPL has come to an end, so I am open to new opportunities in the AV field. Anyone reading this is welcome to comment or drop me a line with any openings in the New York metropolitan area.
Stay tuned for a book review later this week.
The idea for HDBaseT is a grand one; to have one and only one cable connected to each display. That cable, a standard network cable with standard RJ45 connectors, would carry uncompressed high-definition video, audio, control, and even power. Yes, they expect to power your display via the same Cat5 jumper that carries everything else. One cable in this case literally means one cable. In real world applications, that isn't what's happened. For one thing, native HDBaseT inputs and outputs. Even if manufacturers want to use this kind of system, even mid-size displays in the 50 to 65 inch range outdraw HDBaseT's upper power limit by an approximate factor of 10. So, like a phone company running fiber to the curb before converting to old-fashioned copper, we're left with easily-pulled Cat5e cables to a receiver-box giving way to good old HDMI cables for the proverbial last mile. We're not quite where we someday can be, but it's an improvement. Crestron's Digital Media is an HDBaseT solution, as is Extron's XTP and AMX's DXLink, the heart of its Enova solutions.
How do solutions compare? Not only does Enova have a different feel from Crestron, but AMX's DVX (a family of all-in-one presentation switchers) behaves differently from their DGX (more conventional digital matrix switchers). One thing I hadn't known - because I'm not in the habit of removing the case on expensive pieces of electronics unless I have to - is that while it looks like one seemless piece of hardware, the DVX is, in fact, card-based. My notes include a glimpse at its innards, highlighting the HDMI input cart swappable with a DXLink card for another model. It's a clever design philosophy which allows AMX to inexpensively and reliably use one platform to produce a suite of units with slightly different features.
So how does Enova compare to Digital Media? At a glance they're similar; each has a presentation-switcher with built-in control processor, audio mixing, and an amplifier. Each line boasts a modular matrix switch capable of handling different video formats. Each has a series of Cat5 transmitters and receivers. Each handles HDCP key authentication, effectively eliminating any risk of running out of keys in large systems. Looking a bit closer, one sees differences.
The first distinction - of which AMX is justifiably proud - is that each output on an Enova system has a built-in "smart-scaler" which reads the EDID from a display and scales the image to fit. This means that even in a large system with many different displays each device will get an image at its native resolution. Their contention is that other practices, like choosing "best common", not only leave adjusting resolution to a device's onboard scaler, but fails to take advantage of the higher resolution in the largest display in a system. It adds a certain measure of cost, but AMX feels that they give value for it.
The second distinction is that each AMX matrix switch has a control processor built into it. This makes for a neater and more compact installation, but isn't a tremendous improvement over the inclusion of a standalone processor, and burying it in the switch gives you the new problem of having no local control ports. Unless everything in your system has IP control, you'll have to build out the system with varying add-ons. In all fairness, AMX has a nice suite of IP-based control port expansion modules, and seems to have a philosophy of preferentially using IP-based controls.
What about the distinction within the Enova line? At Enova training in AMX's New York showroom we got a demonstration of both the DVX presentation switcher and DGX matrix switcher. The performance in switching speed is markedly different, with the DVX switching, I would estimate, twice as fast as its cousin. The difference was explained as a result of different engineering teams working on the two devices, and upcoming firmware updates to let the DGX catch up were hinted at. (it was further explained that the slower switch time - one and a half to two seconds - was a result of the switcher dropping sync when it changed sources. It is very interesting how the same hardware can have different performance given different firmware.
What about venerable switching manufacturer Extron? I don't have much to say about their XTP line; certainly until they start shipping the Cat5 input and output cards for their switcher, they can't be said to have a real solution available. Even given that, they're still behind the curve in not having a DMPS300/DVX style presentation switcher available. They have added power injectors to their XTP line, but it is, at the very least, unclear that these would work with other HDBaseT solutions.
Long story short? It's an exciting time in the commercial AV industry as we finally seem to have the hardware, software, and expertise to start making digital video work closer to the way it should; many of these solutions are, if anything, easier to use and more flexible than old analog solutions.
A side note: my adventure at AVI-SPL has come to an end, so I am open to new opportunities in the AV field. Anyone reading this is welcome to comment or drop me a line with any openings in the New York metropolitan area.
Stay tuned for a book review later this week.
Nice, round up. Just an FYI we are beginning to see a limited number of products that natively support HDBaseT like the new range of Panasonic projectors (PT-RZ470/430 and the PT-RZ370/330) in the next few months we will see a number of products in the prov AV world support HDBaseT natively. Though the consumer world if a different question all together.
ReplyDeleteI should revisit this at some point as HDBaseT endpoints become more common. There are also some flatpanel displays from Primeview that follow the dream of using one cable for power/video/audio/control. As you see, I live in the professional world. There might be a post this week on where I am now.
ReplyDelete