LX Navigation Introduces the New LX 9090 Zeus

The LX 9090 Zeus features a 5.5 inch colour screen and runs a new LX
Navigation glide and nav software. It has the ability to show either
a 2D or 3D terrain moving map on screen.

For more details visit the LX Navigation section at http://www.foxonecorp.com

or

http://www.lxnavigation.si/avionics


Delivery of the new Zeus are scheduled for the end of May 2012.

Dave Springford wrote:
The LX 9090 Zeus features a 5.5 inch colour screen and runs a new LX
Navigation glide and nav software. It has the ability to show either
a 2D or 3D terrain moving map on screen.

For more details visit the LX Navigation section at http://www.foxonecorp.com

or

http://www.lxnavigation.si/avionics

Is this the successor of the LX9000?

The screen shots look very much like the LX9000 software (by LXNav),
however LX Navigation claims to have developed the software
"completely".

This is confusing. And the LXNav vs. LX Navigation story is already
confusing enough.

Max

Dave Springford wrote:
The LX 9090 Zeus features a 5.5 inch colour screen and runs a new LX
Navigation glide and nav software. It has the ability to show either
a 2D or 3D terrain moving map on screen.

I sent a few questions about the LX 9090 Zeus to LX Navigation, in an
attempt to resolve my confusion, here are the answers from Crtomir
Rojnik:

Q: Is this a team effort by LXNav and LX Navigation, or did you adapt
LK8000 to look like the LX9000, or did you really write a new software
from scratch?

A: "LX Zeus is using LX Navigation original SW, there is no adaptation
of LK at all."; "Yes, we did new SW for LX Zeus."

Q: Are customers free to replace your new software with a custom one
on the hardware they own?

A: "at the moment we don't plane Zeus running under SW like XCsoar. MM
family will continue on free ware SW."

Q: Does it run Linux or Windows?

A: no answer, question was ignored.

Q: I heard the LX Zeus runs Linux, is that correct?

A: no answer, question was ignored.

Q: The MiniMap and the LX8000/9000 are all based on the old ARMv5
architecture. Did you choose a modern CPU architecture for the Zeus?
(i.e. ARMv7 with NEON and OpenGL)

A: no answer, question was ignored.


Unfortunately, nothing new. All interesting technical questions were
ignored, not even the choice of operating system is disclosed at this
point.

Can't wait to analyze the firmware image when it's out, to see if the
software was really developed from scratch. I don't think there's a
market for yet another commercial gliding software - writing one costs
millions, but the market is too small. I don't think any company can
afford to write a new one from scratch, it'll never pay off. It would
need to be a "killer" app that everybody wants to buy. The new 3D
terrain display is not even something I find even remotely useful.

Max

Dear Max et all,

I am the lead developer of the Lx Zeus product. My colleagues have
informed
me that some misleading/wrong is being shared on various glider
forums.
Just to clear things out, I would like to point out a few things:

* Zeus is a multiprocessor platform. The main processor is a high
performance
RISC Cortex A8 - 750 MHz (1GH comes soon). OpenGL is also used for
graphics
rendering (essential for 3D). At least two more high performance ARM7
micro-
controllers are essential for correct operation. One takes care for
the
peripheral interaction, especially for the CAN - 485 interaction,
while the
other is used for inertial (AHRS /vario) calculation. It depends on
the Zeus
version. More uCs can be added later via CAN bus connection.

* Zeus runs Linux. Pure, vanilla Linux was taken and then heavily
patched in
order to get it running. Some in-house developed drivers were
necessary to get
the openGL running and the CAN messages into Zeus.

* Zeus is pure CAN system. Can bus is the main bus. It is loosely
based on the
CANAerospace formulation, which was significantly extended to cover
additional
needs. Complete 485 bus (old LX bus) is also fully supported and one
of the uC
roles is to take care for the communication between both buses.

* The software in Zeus was written from scratch. It is completely in-
house
developed. It is coded in C++. Significant portions of the
mathematical model
logic is written in pure C++, which is not platform dependent. This
means that
a significant portion of code is the same in 7007C and Zeus. Please
note that
7007C code was also written from scratch and not a single line of code
from
old 7007 was used.

* The graphical interface in Zeus is build around Qt library. In
7007C, we
developed own library.

* Although I am familiar with the XC soar and LK, not even a line of
code was used
from there. The programming approach, abstraction and platform
differences were
simply too large for this to be effective.

* Zeus as such is pretty a complex product and it is difficult to
handle on OS
level. It requires special patches and drivers. For this reason, I
find it
difficult (but not impossible) to be a host for LK or XC soar systems.
We
definitely do not have enough man power to dig into this.

* You will notice firm Kanardia d.o.o. to which I belong. Mr. Rojnik
(head of LxNavigation) is my
lifetime friend and after the issues with the lxnav, we are working
together
with Lx Navigation on development of new products. Most of technology
used in
Zeus was developed and tested by Kanardia in previous years.

Note: Almost identical email was sent to Max.

With very best regards,

Aleš


--
Dr. Ing. Aleš Krajnc
Kanardia d.o.o.
Ulica heroja Rojška 70
SI-3000 Celje
Slovenia

Mob: +386 40 360 512
Tel: +386 590 12 521
Fax: +386 590 11 416
URL: www.kanardia.eu

Ale¨ Krajnc wrote:
The main processor is a high performance RISC Cortex A8 - 750 MHz
(1GH comes soon).

Oh, my other post was written before I took a phone call an hour ago,
didn't see your RAS post - you previously wrote it was ARM11 (=
ARMv5), but if it's really Cortex A8, then it seems to be ARMv7 after
all.

Max

Max Kellermann wrote:
Dave Springford wrote:
The LX 9090 Zeus features a 5.5 inch colour screen and runs a new LX
Navigation glide and nav software. It has the ability to show either
a 2D or 3D terrain moving map on screen.

I sent a few questions about the LX 9090 Zeus to LX Navigation, in an
attempt to resolve my confusion, here are the answers from Crtomir
Rojnik:

Just in case anybody else is interested in the technical details, I've
got an informative email from Ales Krajnc, who was involved in the
design of the LX 9090 Zeus.

Q: Does it run Linux or Windows?

A: no answer, question was ignored.

The LX Zeus indeed runs Linux. It is a custom Linux distribution, and
the application is based on Qt.

The firmware image from a similar product by the same company
(Kanardia Nesis, http://www.kanardia.eu/products/nesis) suggest that
the LX Zeus runs a Debian-like distribution, Linux kernel 2.6.24, Qt
4.6.2. Qt runs on top of the frame buffer.

Q: The MiniMap and the LX8000/9000 are all based on the old ARMv5
architecture. Did you choose a modern CPU architecture for the Zeus?
(i.e. ARMv7 with NEON and OpenGL)

According to Ales, the LX Zeus has an ARMv5 CPU with 750 MHz (the
firmware has XScale drivers). That leaves me unimpressed, because
this is only slightly better than the ancient hx4700 (ARMv5 620 MHz).
Contemporary mobile phones have dual-core ARMv7 CPUs with 1200 MHz or
more.

Max

Strange thinking Max!

I do not know why you compare procesor in Zeus with new procesors in Mobile Phones.
I believe that they wont run android + hundreds of application on this device.

If you have written application for specific hardware it does not matter what processor is inside, if peformance is good.

So if they run 3D on 750MHz processor and works good, I would say that they are good programers.
Also on 1200Mhz dual core you can have non optimized code and can run slow.

I say that we wait until is out and see what its performance!


I do not know where your problem is Max...?

wrote:
I do not know why you compare procesor in Zeus with new procesors in Mobile Phones.

If you're interested, why don't you ask why I do that?

If you had asked, here's the answer I would have given:

ARMv7 is state of the art, it's common and cheap. When ARMv5 was
sold, desktop computers had PentiumII CPUs. It's deprecated
technology, just like the P2 is deprecated.

A product based on ARMv7 may be overpowered now, and ARMv5 may (or may
not) be enough now; but in two years, you may want new firmware with
new features. New features will consume more CPU, and the old ARMv5
may not be powerful enough. With an ARMv5, you'll have to buy a new
instrument, while the ARMv7 will survive a few more years. Don't
build new products with deprecated technology.

So why do I compare mobile phones with the Zeus?

- CPUs in mobile phones are a good showcase for current mobile CPU
technology, and CPUs in mobile phones are designed for similar tasks
as the Zeus (GPS navigation).

- my mobile phone runs XCSoar, which makes the Zeus and my phone very
comparable.

- there is no technical reason why a flight instruments for thousands
of dollars should have ridiculously outdated hardware compared to
contemporary phones.

(Note that Ales has clarified meanwhile that the Zeus does have an
ARMv7 CPU, contrary to what he wrote first; what I know for sure is
that the LX MiniMap has ARMv5)

If you have written application for specific hardware it does not
matter what processor is inside, if peformance is good.

Will the performance still be good after the next firmware update?

I do not know where your problem is Max...?

Is that a question? And why do you think I have a problem?

Max

Max,

On Feb 14, 10:25*am, Max Kellermann wrote:

A product based on ARMv7 may be overpowered now, and ARMv5 may (or may
not) be enough now; but in two years, you may want new firmware with
new features. *New features will consume more CPU, and the old ARMv5
may not be powerful enough.

Perhaps because it's not part of their business model.

The navigation devices we're talking about offer a number of
advantages, such as
running as-is (out of the box), tested, serviced and guaranteed
function of hard and software combination and high reliability. If
these features can be realized, using newer, more expensive hardware
will reduce the profit margin. Newer devices will be sold in the
future, with new hardware and software. How else would they profit
(and fund further developments)?

It seems to me that the number of makers of great soaring instruments
fails to realized the efficiencies of mass-production.

The availability of Android devices with absolutely fantastic
displays, and my desire to be able to upgrade the software in the
future, and the great functionality and flexibility provided by XCSoar
proves the advantages of the alternative. But that's not a
"business" rather than a dynamic eco-system of talented software
developers and mass-market hardware. I hope that more integrators
will be able to market and service devices that use cheap mass-market
hardware, have optimized sensors, and run upgradable, free and open-
source software.