![]() |
Quote:
We should be shouting this from the mountain tops. Proprietary software/hardware only benefits the manufactures IMHO, not the users. As I understand things,ARINC 429 is already the standard of the industry for the big boys, and it is used in some of our stuff-------why not go all the way?? GRT has done a great job dealing with outside vendor compatibility as can be seen here. http://www.grtavionics.com/compatible.html |
Quote:
|
The internet is largely based on open standards, and yet the companies that make products for it largely do well from it.
Arguing in favour of closed systems only works if you don't have good interoperability. The problem is no one is arguing in favour of more open systems ... that would need to come from the experimental avionics users first and foremost ... it's tougher to do in certified avionics because of some of the standards that need to be met. ARINC 429 is ANCIENT. CANBus and ARINC 825 is where it's at for our market. Most of the modern EFIS' use CANBus' just their own special sauce version of it. I've actually got some ARINC 825 code running on arduino and such at home, I've been playing with this, seeing what's needed. With the right open specs and tools, there's NO reason we couldn't have truly interopoerable "LRU"s ... I've been dreaming of drafting an open ARINC 825 profile and creating supporting tools ... but ti's a long road ... any software engineers here? :) Not to mention that the competition would bring costs down (and/or features up) and really help bringing aviation to more people in the long run. Many of the best open standards started either in academia, or as grass roots efforts ... seems to me that would be the way to go here too, don't expect established vendors to just start playing nice with each other. |
The big gorilla is sitting quietly in the corner. Our cool screens are expensive to design. Santa Claus does not contribute. Each manufacturer is required to issue paychecks twice monthly to the engineers and software writers. If their $ are over at the competition building... the paychecks bounce.
Then the Blue Mountains are just a pretty place and not a source of modern glass design. I get to see behind the scenes each week in the avionics industry with everything from ADS-B, EFIS, AHRS, P25 digital com radios, CO detectors and Iridium com boxes. It is all very expensive to design. It is compromised in modern times because the components come primarily from the consumer electronics industry. ie. cell phone technology. Whereas... in the past, the military and industrial demand for components drove the designs to a higher level of robust nature. Tolerance for heat, reliability of discrete components, long term availability of IC's is all a changing landscape these days. Having said all that.... this week I launched into marginal icing conditions and drizzle in Nevada at 12K feet. The turbo 210 was doing ok performance wise... but the iron gyro that decided to have an insidious failure in the form of pitch up first... followed by slow roll to the right... while in IMC---- At that moment... I was missing the solid state MEMS derived attitude in my RV big time. Thankfully the S-TEC rate based gyro didn't care what the attitude indicator was doing. Center gave me lower... I dialed in VS command of -400fpm and kept my hands off the yoke. I am going to work on convincing the aircraft owner to look into Aspen.... rather than spend further dollars overhauling the old gyro. We are all very lucky to be flying in an era where the Cheltons and Garmins and Free Flight GPS receivers are giving us situational awareness that was sorely missing for many years. |
Nick makes a very excellent point, as do several others here. The BIG problem with the costs and associated recovery of said costs is just in the scale of what we are dealing with. Compare that it took Garmin over 10 years to sell 100,000 GNS units (which seems pretty darned good)...until you realize that it probably cost many, many millions to develop. You have to spread that NRE cost over a VERY small (in the consumer electronics type world) number of units, over a very long timeframe. Most folks with that kind of money don't want that sort of ROI horizon. Couple that to the fact that IC's and other discrete components are rarely designed specifically for our components (means the mfgr's have to use COTS stuff that was/is developed for other markets) and some of it will be beyond their control.
Contrast this to the fact that the famous fruit company sold more new phones a couple weeks ago in their first HOUR of sale than Garmin sold GNS's in EVER in 10+ years....that's why costs are what they are. You simply cannot in any way/shape/form compare avionics costs (and NRE and such) with anything else consumer related (internet, software, etc..) that much larger audiences use. The costs are spread out amongst many more users in a much more compressed timeframe. Regarding lifespan, many other very good points have been made. Open source common interoperability at this point is a pipe dream in our world, but in the heavy iron world I come from it's a much different story. MTBF's, MTBR's, etc.. are calculated down to the minutes, but conversely these units may see many thousands of hours per year and many hundreds of cycles. The Airlines and Boeing have a lot more leverage than we in GA do. I think we're starting to see systems as a whole maturing a but more, with more improvements coming from software and ancillary devices over core system changes. CCFL's are pretty much long gone in this world, as are other bits of technology (Smart Media, Compact Flash, etc..). I suppose some of which we use today will face a similar fate as newer technologies come on board, but at least it's not as bad as phones or PC's. Just my 2 cents as usual! Cheers, Stein |
Stein, as always you bring a POV that is both the high road and educated truth.
If I assume an 8 year lifespan on an EFIS with data subscriptions and compare that to 8 years of "apples" plus service it becomes an interesting perspective. |
Give me a CANbus and we can work miracles!
Your car has been running this technology for years and it has stood the test of time. |
Quote:
|
There are plenty of industrial rated ICs with extended temp support and such ... The main CPU manufacturers have long-term supported chips, including Intel.
Some technologies like CANBus actually matured in larger markets (i.e. automotive and the liked) so the prices and economies of scale are already there. There's no doubt design can be very expensive, but you have to separate the certified and non-certified world. Having seen what it takes to create a certified products, it's absolutely mind-boggling. I understand why a GPS costs the price of a small car, and there's little to no way around it. Part of the problem is that entire systems, not just components, have to be certified together (for now anyways). With experimental avionics, this is not required, so our market is much more primed for the use of things like open interfaces, protocols, etc. Create a common CANBus-based protocol (full stack, not just the bottom 2 OSI layers!), create some test tools and such, develop against the standard, certify against the standard, and voila. Give the DIY nature of home builders, you'd think you could get a community growing around this and kit-build avionics :) |
I certainly can't argue with the previous two posts, both make good points!
I'm not trying to be a negative nelly or a pessimist, but I am trying to note the realities. I know of at least a half dozen people who have "rolled their own" open source EFIS with good success (with plans to perhaps sell it), ultimately to end up just buying a system on the market today. You still have to remember that AT MOST, we're talking about a market (in it's entirety) is maybe hundreds of units per year...certainly not thousands. Fun for hobby and DIY projects (certainly doable), but still quite difficult for anyone to make much of a real business case out of from a long term point of view....and for it (whatever that is) to be a real product that folks will adopt and ultimately pay money for, it'll need to be something more than that; which is the old chicken and egg problem all over again! :) What has happened to many folks is that they come up with something neat, then find out that the customer support in this business will bury you quickly...so they have something that is becomes too big to be a hobby/part time, but not big enough to support as a job. There are dozens upon dozens of said examples out there of this happening. I don't know the answer at all, but I do agree that leveraging current COTS stuff is a start - and I do see some of the mfgr's now dipping their toes in. Cheers, Stein |
| All times are GMT -6. The time now is 01:30 AM. |