Savant, current opinions/thoughts?
vining
Posts: 4,368
For those who either work directly with Savant or work in a shop where someone else does it what are your current opinions or feelings of the product?
How much programming can the dealer acutally do? Can the dealer create his own drivers or just modify existing drivers as mentioned in another thread? Is it just drag and drop and data entry?
How customizable is it? Is it possible to get a Savant system to do what we can do with an AMX system? I'm sure there's some cool things Savant can do that AMX can't but what would they be?
I have a client who now is shifting towards Savant and he's really only considering the UI and the GUI's which I don't think we can touch with AMX gear. I'll admit my GUI's are a bit dated and boring by today's standards so maybe if I spent more time on GUI's then I do programming I could stay closer to the curve. He actually wants me to become a Savant dealer and experiment on his house but I really have no interest in learning a completely new system even though there's no real love to be lost with AMX after these last few years. If it's awesome and I'd be a fool to stick with AMX I'd consider it (a little) but if I can't create the code and do what ever customization that my little pea brain can come up with I'd rather stay in my comfort zone.
How much programming can the dealer acutally do? Can the dealer create his own drivers or just modify existing drivers as mentioned in another thread? Is it just drag and drop and data entry?
How customizable is it? Is it possible to get a Savant system to do what we can do with an AMX system? I'm sure there's some cool things Savant can do that AMX can't but what would they be?
I have a client who now is shifting towards Savant and he's really only considering the UI and the GUI's which I don't think we can touch with AMX gear. I'll admit my GUI's are a bit dated and boring by today's standards so maybe if I spent more time on GUI's then I do programming I could stay closer to the curve. He actually wants me to become a Savant dealer and experiment on his house but I really have no interest in learning a completely new system even though there's no real love to be lost with AMX after these last few years. If it's awesome and I'd be a fool to stick with AMX I'd consider it (a little) but if I can't create the code and do what ever customization that my little pea brain can come up with I'd rather stay in my comfort zone.
0
Comments
AMX is still WAY more customizable but I'm no longer sure that's necessarily an edge. Clients don't seem as impressed by quirky features. It's old hat. 'Working' and 'The Trades are out of my house' seem to be much more valued at this time. Add to this an overall perception that AMX stuff isn't as reliable as it once was and you've got trouble. We had a few clients get bit by issues with the AMX HDMI switcher issues recently and it didn't turn out well. The client was not impressed.
There is a bit of a learning curve to Savant that will drive a programmer like yourself insane. You find yourself saying, "I could do this so much more easily if I could just program.". But other things are faster. On balance it's about the same, maybe a little faster to get up and running.
I've had a lot of dealings with it over the years since it first came out.
My previous employer was an early adopter and I even did all the trainings. As we started Certified Cyber Solutions I had even more dealings with them and the engineering people there. I learned a lot more about how they work and get things done. My current employer is also in the same boat. It's honestly pretty disheartening after working in the industry for over 14 years. But, I'm not going to start whining...
To answer your question more succinctly: it's a lot more stable than in the past and works as advertised. But it really steers your business down a whole new path. That may or may not be what you want or planned.
I absolutely hate it. Don't get me wrong, it's a good product, and getting better all the time. It's quick and easy to set up a system. It's easy for the client to use, and most people can walk right up and jump in with no instruction. It's intuitive and clear. It has nice graphics.
But it's utter hell to customize, and their philosophy seems to me to be, "if you don't do it our way, that's your problem, not ours." You can write your own drivers for it, if you are a masochist. It's not so bad if you start with a working profile and modify it to your device, but the XML scheme is very unforgiving, and has zero error-checking. You make a mistake, and either the entire program crashes, or your device just disappears from the project (as in, it won't recognize it as a viable source). You are on your own for finding where the problem lies. If you add a function that isn't in their template for that type of device, you have to create a custom Automator script to add it to your controllers, and it doesn't automatically propagate to all like devices unless you only have one. Their control screen paradigm is every device has it's own screen for every room in which it is useable. Which doesn't sound so bad until you have a dozen rooms, and half a dozen thermostats, for example ... that gives you 72 thermostat screens. You can link them together to a degree, so changes on one are reflected on the rest, but not all the properties carry over, and then you have to do a search and replace; and it is so easy to make a minor slip that screws up everything, it raises my blood pressure just thinking about it.
And you cannot completely do without customizations. They use the same UI screen template for every device in a given class. All their disc players, for example, have a chapter/track/time indicator, and those screens get generated even if you are using an IR device and no such information is available. In fact, I have yet to see a disc player where they did populate. I delete those buttons on *every* job. They also put disc selection buttons on every disc player, whether it's a changer or not; they use the same template.
It's also a really bad choice for very large jobs. I have one job that I inherited from a company that went under, and they have a system with more than a dozen TV rooms and maybe 25 rooms or so of audio. It takes forever to build and load the project. One button is out of place, or one tiny function doesn't work right ... 10 seconds to fix and 45 minutes, at least, to load, more than that if there is much network traffic. It doesn't run slow, the mac mini they use for a master is very robust responsive. But their project files are simply enormous, and they don't do incremental loads of any kind.
Updates are another nightmare. They seem to push a new revision every 2-3 weeks, and once you update your programming computer, you have to update your projects too. Some are painless enough, but some require OS updates and all manner of patches, and every now and then, an update will utterly break your old UIs and you have to create new ones from scratch, all your customizations included. Their official solution is to create a new user on your programming computer for each revision and install to that, so you can edit older projects without updating them. This gets really unwieldy after some time ... and, of course, none of the other tools or settings you painstakingly set up for the original user carry over to the new one. I've had times when I was logged into three users at the same time, just to be able to use a VPN connection on one, have an old file open on another for comparisons, and my current project.
So my advice is, if you have a boilerplate type of system that you need to get up and running fast, it's just fine. Unless it's a simple IR device, don't spec in any equipment they don't have a fully tested profile for (which raises another point, I have yet to see a system where every device profile worked perfectly ... they are good at getting them fixed for you, but it's dang aggravating). Anything large, or needing lots of custom stuff, stay away.