Next Generation Panels
jjames
Posts: 2,908
So what's the hold up do you suspect? This is sort of a "what do you expect and when?" type thread. Just something to get talking about the next generation panels which hopefully are being developed as we speak.
Let's face it, the G4 engine is getting outdated. While you can still do some "fancy" stuff, it's not even comparable to an iPhone or iTouch's capabilities. Speaking of phones, I just recently got a new one - a Sprint HTC Hero, which runs on Android. Naturally, I wanted to peak around and see what'd be required to do a few apps for it. The requirement: Java. Who would have thought that these cool looking interfaces run on Java (I didn't - to be honest I was completely oblivious to the Android engine even though it's been out for a while.) I started thinking: "Why couldn't AMX do something along these lines?"
Sure, the argument that a cell phone has much more "buy-ability" over a touch panel that costs ten times can be made, and I think we all understand that one. But, thinking about it from a semi-technical standpoint, I keep asking: "Why not?" Sure, that would require all of us needing to learn Java to use the "nextgen" panels, but with the push of Duet (i.e. new modules only being written for it), I don't see this as too far fetched.
So what's your thoughts on AMX's future next generation panels and what should they be like? What do you expect to have with it? And lastly, when do you expect to see some "knock-your-socks-off" new panels?
Note: It may seem like I know something, but I *REALLY*REALLY* don't. I'm just throwing out the question: "If you could run a touch phone on Java, why not a touch panel on Java to control AMX?"
Let's face it, the G4 engine is getting outdated. While you can still do some "fancy" stuff, it's not even comparable to an iPhone or iTouch's capabilities. Speaking of phones, I just recently got a new one - a Sprint HTC Hero, which runs on Android. Naturally, I wanted to peak around and see what'd be required to do a few apps for it. The requirement: Java. Who would have thought that these cool looking interfaces run on Java (I didn't - to be honest I was completely oblivious to the Android engine even though it's been out for a while.) I started thinking: "Why couldn't AMX do something along these lines?"
Sure, the argument that a cell phone has much more "buy-ability" over a touch panel that costs ten times can be made, and I think we all understand that one. But, thinking about it from a semi-technical standpoint, I keep asking: "Why not?" Sure, that would require all of us needing to learn Java to use the "nextgen" panels, but with the push of Duet (i.e. new modules only being written for it), I don't see this as too far fetched.
So what's your thoughts on AMX's future next generation panels and what should they be like? What do you expect to have with it? And lastly, when do you expect to see some "knock-your-socks-off" new panels?
Note: It may seem like I know something, but I *REALLY*REALLY* don't. I'm just throwing out the question: "If you could run a touch phone on Java, why not a touch panel on Java to control AMX?"
0
Comments
But still something we need on it.
The wireless touch panels should have a built in Mozzila Firefox or any other browser, I think the table/wall panels don't need this.
Talking about a Java virtual machine running on it, could be nice make small apps inside the touch panel. And very important, the program could modify the UI, make the button disappear, change oppacity. The java apps could be triggered by some events like on/off of a button, level, and by send_commands like:
SEND_COMMAND dvtp, "'button fade in".
Just use a tablet PC, netbook, or laptop in combination with the new NXV-300. If your clients want a computer, give them a computer
Jeff
Personally, if a customer insisted, I would install a dedicated rack PC and open a computer control window to it; if it were a wired, desktop panel, I would (and have, in the past) go with a TPI.
Here are my guesses for a "G5" panel:
1) Native Flash support. The new DVX-2100-HD has Flash pages for setting up the video scaling and audio mixing. I expect that this will spread to other AMX products, and pretty soon. Touch panels are the obvious destination.
2) Transitions/zoom. The ability to programatically (or via "TPDesign5") shift buttons/images around with ease. And not just relocate them, but make the button travel along a path. I predict lots of cheesy "curtain" effects when you start a movie.
3) Some form of MultiTouch. Again, due to the iPhone being common-place, AMX really can not ignore this and remain a high-end product. "MonoTouch" is so last-century. (Not trying to be offensive here. Just presenting an "Average Joe" view.)
4) Replace the surface capacitive overlay with a better form of touch. This is probably a requisite for multitouch. Projected capacitance (like iPhone) is a much nicer surface to interact with. Overlays are really starting to feel retro.
And here are my wish-list ideas:
5) A waterproof panel (not resistant... proof) that can be installed in showers/pools/etc. Again, getting away from the surface capacitive overlay may make this more achievable. Should come with a complimentary tube of silicon.
6) Fail-over NIC. (Okay, I don't expect AMX to do this one, but it would be awesome.) Given the haphazard nature of WiFi (due to telephones, microwaves, solar flares, etc) it would be great if panels could utilise a wired connection when docked. Perhaps the easiest way to do this would be to put the second NIC in the cradle and use a proprietary protocol between the panel and the cradle. The panel could have some smarts that disable WiFi transmission (not Rx) while docked. The panel places it's request via the cradle, which then does the heavy network lifting of the data. The cradle and panel could do a run-time "bind" so the master knows to route data to the cradle while the panel is docked.
Just my guesses. Let's revisit this thread in a year to see how wrong I am!
Roger McLean
Swinburne University
I'm sure we've all got a wish list for G5.
I think being able to run apps (like a mac dashboard) would be awesome. I only say this with some exposure to S**T. If the panels do end up running off Java it opens the platforms and gives us all an extra skill to learn...Java. Web browser isn't much of a feature anymore in small devices but a requirement. Its silly that my Blackberry can surf the web, but my 5200 is land locked. I have looked at the Android devices and they are freaking cool. I wouldn't have suspected they are just running on Java.
As far as overpriced panels in an undervalued world. I think AMX could wake up from its massive profits from a 17" touch panel and realize less is more. We can put more panels in a job if they cost less. I am working on a project right now that just so happens to be perfect from AMX Home and the DAS Experience Kit, for $11,800 retail you can get a whole lot of stuff. I haven't fully used AMX Home yet, but I ran through a few of the options and it seems rather interesting, all that stuff said.
I think Control in general should have more mid-level market appeal. If it doesn't we'll see smaller cooler devices, that can't quite get the job done attempt to and a client won't be thrilled, but what happens if they are okay with it? If they tell a friend how little something cost, but how much it can do? There have been quite a few of the fly by nights that for now have gone by the way side if we can it would be great to capitalize on highly profitable small jobs.
Sorry for the rant. I think there needs to be a "control system revival". I'll be in attendance.
1) A command set to control the NIC card or and/or a way to embed some kind of command set to run upon loosing connection. (Rollover would be nice)
something along the lines of " hey touchpanel, reset your NIC card to these settings and reboot. If after some time you don't find a connection, go back to your previous settings and report back."
2) gesturing (swipe, pinch, etc..) Something like the iPhone
3)changing in-wall TPs screen viewing angle orientation. What I mean: on most all displays the horizontal viewing axis is much wider than the vertical. Obviously, for a TV this makes sense since most people sit at the same height but will be splayed around the TV at a fairly wide angle.
For an in-wall this almost makes no sense. Very rarely do you stand more than 10 degrees off of the horizontal axis center. However, the vertical axis is very wide since people stand within arm's length of the panel. And people are all sorts of heights. I'm tired of seeing my clients stooping down to see the panels.
4) More graphic formats. (motion and still)
5) Wireless N and G
6) Bring back some G3 panels. There's no mid-level anymore. It's either keypads or Moderos. My vote is a small in wall, small table top and large table top. Offer model(s) with the wireless option. I have on in my house that works like a champ.
But I would like to see:
1. Thinner panels. (Depth and bezel) More like the 5" wireless. The 8400 is starting to look outdated.
2. Better table top docking stations. I want to just drop it in and forget it.
3. In the corporate world, wireless "A" on the panels would be AWESOME!
4. This one is a two part for tabletops:
a. Annotation built in
b. Video output. That would allow you to annotate on ANY source! (Cr3str0n has this already
and it's great for education.) PS, don't mention the API becasue you can only annotate on a PC.
- Thinner bezel, especially on the 5200i since you have to press in a weird angel if you want to do anything on the touchscreen near the edges now.
- Gestures would be great.
- Clearer display.
- More colours. (24-bits instead of 16/18-bit)
- better image format support.
And I agree with the points the others made so far.
- what's not midlevel about the new NXD-435 panel? It's small, cheap, but looks awesome.
- multi-touch would have no extra benefit for AMX touchpanels (read my "rant" in some other post on the forum)
- flash support would be ok, but then again, would you develop in flash? (PS. i'm pretty sure the DVX-2100 doesn't have flash pages. They sure look pretty "flashy", but I think it's something like Ajax (they are all .xml pages
- Why wireless N? for extra speedy button presses? Wireless A, that would be awesome.
My extra's"
- Tabledock with wired ethernet!
Just a few thoughts:
Wireless N - Why go for A when N give you more freq flexibility and more bandwidth? For those that like streaming lots of IP security cameras to their wireless panels (Dr Evil?) the extra bandwidth would be useful. Multiple frequency would allow panels to be installed into older non-A/N environments without having to change existing WiFi infrastructure.
A Multi-touch "button" over the top of a video stream (IP camera) would be an excellent way of controlling the camera. I presume the joystick control is okay for pan/tilt, but MT "pinch" would give you zoom as well. I'm sure there are a lot of other useful applications of MT that I can't think of right now. (Some form of drag-n-drop layout for MultiSource-to-MultiDestination selection?)
Flash - I'm just looking at the DVX-2100 pages again. The Flash pages are for the DVX-2100HD-SP device (typically 5002, selected from the device drop-down in the top right corner). Once the page for the device has loaded, right-click->About Adobe Flash Player 10.... As for whether I would develop in Flash or not, if there are no additional licensing requirements (i.e. Duet) then why not?
Roger McLean
Swinburne University
I think the in wall TPs are out of style and I would like to see either extremely small bezels or preferrably trimless TP's that would require a mounting box that gets plastered in. The TP would in affect be w/o a case then slide & snap into place. This would probably need a push release latch for servicing.
Dynamic Images slots.
Ummm... I thought N was both 2.4GHz and 5GHz. http://en.wikipedia.org/wiki/802.11#Protocols I was suggesting that N allows you run in 5GHz exclusively if you wish, or in 2.4GHz for backwards compatibility. When using dual radio APs you can support both freqs simultaneously, (N devices run on 5GHz while B/G run on 2.4GHz), which allows for more simultaneous connections. Where customers may not be ready to update to 5GHz networks, at least an N product falls back to 2.4GHz.
I would contest that N is is "a single AP per installation device". This may be the case if your APs are low quality or do not support dual radio, but is certainly not a limitation of the spec.
If the N's do have 2 or 3 RF channels per device that makes adding multiples a problem since you're not supposed to repeat channels with certain exceptions like extreme distances where they are out of overlapping range. Now if you can't get far enough a way to add additional AP's I guess you could turn off channels in the N as required but then you might as well just install a single channel AP then.
I also wonder how configurable the RF channels are in the N AP's. I would almost hink they'd be fixed just to prevent users from setting them to channels that overlap. I guess they could have rules that pop up telling you tha't not allowed but that could only be on device with 2 channel that you would even get that choice.
Maybe I paranoid but they kinda scare me, can you imagine resi or commercial condos where multiple owners installed them. At the 300' indoor range and using either 2 of the 3 or 3 of the 3 channels recommended they'd be stepping all over each other.
I think I argued against you in the other post, but just to be a pain in the arse, multi-touch would be incredibly useful on AMX panels. Think in the case of a multi-touch 5200i in a home theatre setup. It would provide an excellent means of in depth, gestural interaction that doesn't require the user to shift their focus to the touch panel. These things are meant to make it easier for people to use their tech. If someone can pick up there standard remote and locate the control they are after by feeling for it then control the device without having to stop what they are doing (even if it is just watching tv), shift their attention to the control device, look at where there hands are in relation to said device and perform the interaction, using an AMX system to do the same task is a severe step backwards in my books. Just look at the increased usability of macbook's with the multi touch trackpad, or apple's new multitouch mice. There's been a hell of a lot of research and usability studies done into this and its proven to be a good thing.
Agreed flash support would be 'ok'. I think there's more a need for a more powerful UI toolkit, flash is just one of the available techs that would provide this.
For integration with existing wireless N infrastructre. It similar to saying why UTP for the wired ethernet instead of coax. It also provides plently of bandwidth for IP based media streaming directly to your wireless panels.
And if I were to research improvements for the current generation, then I would definitelly channel my research in the two directions: connectivity stability and extended battery life.
maybe i need to change my attitude against multi-touch.
It's a nice way of controlling a device, but I just think there has been too much "Multi-Touch" madness going on.
Like i said in the other thread, if you want a good way of controlling a home theater withouth losing focus of the image, use an R4. That way you can "feel" the buttons.
A 5200i has to be held with two hands, that's enough reason to not use multitouch on a wireless touchpanel...
Multitouch on a NXD-1500 or something would be nice to see though...
If you have an existing wireless N infrastructure there should also be a B+G infrastructure, no problems there i guess...
There is research and prototyping for such interfacing experience. Nobody said it will happen today or tomorrow, but I believe in imagination...
Agreed devices that vibrate when you touch them (by that I mean UI devices, get your mind out of the gutter) make most people want to through said device at the closest hard surface. Unfortunately that implimentation which is common amoungst cheap, widely distributed mobile devices has created a pretty bad impression of haptic interfaces to the general public.
Thankfully though even haptic interaction in its current state is far beyond the vibrating lump of plastic stage that most people have been exposed to. Chris Harrison has an awesome interface prototype that in my opinion would be ideal for AMX interfaces. It provides a multitouch display that has an inflatable latex overlay which provides a dynamic physical interface. There's heaps of other incredible research being done in this field as well. I would love nothing more that to get all touchy feely with AMX.
I think multi-touch could be "OK" for some panels, but "gestures" are the missing piece to the puzzle.
The R4 is a horribly engineered remote for residential use and in no way fixes the problem of needing to take your focus off the TV because it is lacking transport buttons, and numeric buttons are too useful to trade for transports. Not to mention the abysmal touch screen.
Gestures would solve the transport problem with just a little training of the homeowners (if they actually use the system). Swipe:
right -> left = rewind
left -> right = fast forward
top -> bottom = stop
bottom -> top = play
double tap = pause
done that already.
There is also a demo on the AMX website that kinda displays this possibility.
Don't need multitouch for that.
check this:
http://amx.com/ui/file.asp?uiType=gElements&item=Sliding%20Globe%20Navigation
http://amx.com/ui/file.asp?uiType=gElements&item=Sliding%20Popup%20Page%20Navigation
http://amx.com/ui/file.asp?uiType=gElements&item=Side%20Slide%20Navigation
As yuri said linear gestures can already be done and can be implimented easily with bargraph / joystick controls. Unfortunately due to NetLinx's lack of math functionality the processor load required to process non-linear (think cirlce left/right for scrubbing media playback etc) puts the response time through the roof.
Also, if you start implimenting gestural interfaces you run out of interaction methods very quickly if you can only track a single point of contact. When implimented with multi touch capable hardware you can, for example, distinguish if the user swiped with two fingers instead of one which can be utilized to provide a lot a very natural feeling interactions.
Thanks for the links, but they all require the user to look at the panel to start the gesture in the proper spot on the touch panel. I was thinking of gestures at a hardware level so you can swipe anywhere on the screen without looking. I would also like to be able to fill the screen with all the normal buttons and not need to leave a big area that would just be used for gestures.
I also worry about too much focus on gestures and/or multitouch because the homeowners are bound to forget all of the hidden interaction features.
Although you can't currently gesture anywhere easily (you could have a transparent joystick button overlayed and trigger events based on push coordinates) the most efficient way I've found to impliment basic gestural input on current gen TP's is to sit a transparent joystick control behind all of the other buttons. That way as long as the gestures begins in 'blank' screen real estate you can still track it, even if the movement covers areas where other controls are positioned.
Gestures don't have to be hidden, you can impliment them with accompanying visual clues on the interface. For example, say you have a panel located in a publically accessable area. It might be a nice idea to have screen locking so that the greater unwashed can't play with it. At the same time, if the room is is bookable for events etc you may not want the panel to be locked accidentally via a single button push (which would then require a staff member with access to come and unlock it). You could pull up a confirmation dialog alerting the user they are about to lock the system to give them a chance to back down if it was pushed in error. Alternatively you could use something such as a swipe gesture which only requires a single interaction, and is difficult to accidentally trigger.
I agree 100% with you Gary. Maybe some opions for nicer/cooler bezels for the panels..such as brushed metals (brass, alum. etc) or even wood for those high end exec rooms. I'd also think it would be cool if the top/bottom/sides would have a scroll feature like the i-phones. A few extra functions in the bios that would allow me to reboot the panel if it looses its connection for x time period and the number of reboot tries.
--D