But it's not all about density, but power consumption starts being the issue of the day. Anything to compete with the other guy and make a name in the market that really doesn't have any differences in product...

But as another poster noted, there isn't any difference in how AMD and Intel processors operate in a data center. If the machine runs the software you want, most of us who are buying servers don't really care all that much. Today's machines are faster, smaller, and consume less power so they fit in the hole left by the server we just replaced without running any new wires, such operators don't care about the CPU vendor, they are just happy the new server fits easily so we will buy the cheapest hardware that fills the need and fits in the space.

The only players that care about any of this are the server farm operators, but even then, getting another two CPU's in a 128 CPU rack is but a marginal improvement. What they really care about is cost, and there are only a few areas where real estate has pushed floor space costs high enough to really make the marginal density improvements worth the investment.

So density is way overrated as a differentiator in the server market. It doesn't really matter to the bulk of the customers anyway and people that fall for the whole blade server thing but buy server chassis that are not totally full to start, are nuts. IF you cannot afford the blades now, trust me, you won't be able to get them in 2 years after they are EOL'ed by the vendor. Just buy separate servers and keep the upgrade path as simple as possible. So none of this has anything to do with the CPU vendor in the first place and AMD was barking up the wrong tree trying to play in this market.

I used to think this too, but I'm not so sure this is totally true today. There are more CPU makers than Intel and AMD, although these are the two players in the PC/server market, the PC/server market is starting to show a decline. People are moving away from the desktop/laptop in favor of their smartphones and handheld devices and the CPU's in these devices are usually not AMD or Intel made.

I used to think that Intel had to keep AMD going to avoid anti-trust problems, but these days that issue is really less and less important. If AMD went away, I think you would see the rise of the companies doing mobile processors as these devices got cheaper and laptops/desktops got more expensive. Push this far enough and these manufacturers could easily step up and into the small server market and hold Intel's prices in check. Yea, Intel would dominate in the "large iron" CPU market where absolute single thread performance was paramount, but this market segment is vanishingly small because most server farm operators are about footprint and power consumption, not brute force and if you have enough power to run a few virtual machines on a host, it's enough.

Of course the fly in all this ointment is Microsoft and the Windows operating system and it is their choices which really drive this market. If we start to see their server offerings being offered on alternate architectures, you can bet the end is near for AMD. But as long as Microsoft keeps AMD support alive, they will at least have a subsistence market to keep them alive. Well, unless this really *is* the year of the Linux Desktop and Microsoft goes bankrupt, but a snowball in a really hot place has a better chance than that.

So the grammar police are still out in force.... OK, what ever floats your boat.

True, but if you have good enough instrumentation and data collection built into your system it's not *that* hard to figure out what exactly happened. I'm sure they have this system well instrumented so it won't take them long to narrow down on a solution to the problems they are seeing. My guess is that they just need to tune their control systems a bit to account for the valve movement delays and all will be better. They might need to work on the actuators for the valves some if they cannot smooth things out in the software, but it looks like a small software change is all they need IMHO.

And when they happen, you loose the vehicle...I think you are trying to address the wrong problem. IMHO the smart money is on making what they have work though some software adjustments or if they really can't do it that way the minimum adjustments in the hardware to smooth out the valve responses. They are really close here, and from what I just was reading they are a software change away from making this work. Just remove the oscillation problem in software by tweaking the delay calculations in the feedback and I'll bet it's near perfect...

You seem to want to redesign the whole thing to increase the system's thrust and overcome it by brute force... I think a small bit of tweaking would be enough and cost a lot less.. But what do I know, I'm not a rocket scientist, just a EE software guy...

Ok, replace "and" with "or" and read it again.. But EVERYTHING Boeing does fits into one of these areas in some way... Well that stuff with safety and regulations too.

They have touch points physically... There are some things the WiFi systems just need to know that come from the flight control systems, or systems which are attached to systems which are attached to... (You get the idea)

Where I seriously doubt there is a direct logical connection, where some hacker just needs to know the right IP address to adjust the auto pilot or something, there are physical data connections between the wifi network and the flight controls.

Physical separation does make a logical connection harder, but it still does not mean a logical connection is impossible. But my point that network security is about logical connections is still valid. You can have two networks that are not physically separated, that are not logically connected, and that's all you need for security. Yea it's harder when you don't have physical separation, but not impossible.

The reason Boeing went for this was to reduce weight, power consumption and complexity. You can save a boat load of wiring if you can use the same ones for multiple networks. Wires are heavy, especially when you add in all the connectors, lacing and clamps required in an aircraft. So you use the minimum number of wire runs that gets you the required redundancy, bandwidth and latency, put everything going from point a to point b on the wires going that way and then keep the networks that go over the wires logically separate. It also reduces the complexity of the wiring, making it easer to install, maintain as well as allow more flexibility in the future for avionics changes.

All of the reasons Boing would do this are related to the cost to purchase and operate the aircraft. Less weight = less fuel burn. Less power = Less fuel burn. Less complexity = easier to maintain = less maintenance costs PLUS lower NRE costs for new components you want to integrate in the future. Need data to flow between here and there for that new package? Configure the back planes like this, use these IP addresses and slap'm into the empty spot in the rack and you got data. Oh I see plenty of reasons for Boeing to do stuff like this... Not that any of this is really new, except for perhaps using TCP/IP as a transport protocol...

You are correct.. They maintain LOGICAL separation very well. Most people confuse the LOGICAL separation issue with PHYSICAL separation and think that you need to have the latter or you don't have the former. But even with physical separation, you don't have logical separation all the time.

Security requires LOGICAL separation, but that does not require physical separation.

If you can maintain logical network separation, you have all you need for security. Physical separation is NOT required if logical separation is maintained.

For instance, Assuming my switches are properly implemented, I can maintain multiple VLans that flow over the same link and they are logically separate. One Vlan does not communicate with another Vlan. I have this very thing in my home, where I keep multiple logical networks. One Vlan that is my ISP's network segment (because where they enter my home is NOT where I wanted my router to live), one for my DMZ, and a couple for my equipment. All this is though my three managed switches which are trunked together. Doing this kind of thing on an aircraft would save weight (less wires and switches) power and a whole host of other benefits afforded you by being more free to put equipment where you wanted without much concern for how the wiring would need to change..

You'd think Boeing doesn't know how to do networking or something. Come on, who thinks the network in a 787 looks anything like what you have a home?

OF COURSE the flight controls are on separate NETWORKS from the in-flight entertainment systems. However, they do have equipment that is on multiple networks and I'm pretty sure there are places where separate networks may flow over the same physical connections.

What you need for security is LOGICAL separation between these networks, NOT PHYSICAL separation. Where physical separation does make logical separation a given, logical is all you need for security.

On the 787, the answers are NO, and NO. The aircraft has more than two subnets, but does have equipment that spans more than one subnet.... Don't know about the buggy firewalls though, but I assume the firewalls are not that buggy. This was all well understood before the FAA issued the type certification and the FAA required additional work by Boeing (beyond what the existing regulations required) before they would issue the certification. The GAO is nuts.

No, the most important part of a business is (1) to make enough money to stay in business, followed by (2) which is to manage the cash flow so you can pay your bills when you need to.

If you don't get 1 and 2 right, nothing else matters, not even your employee's happiness, because you will go broke and they will be unemployed which is decidedly NOT a happy place.

This most certainly IS a PR stunt. As I understand it, this guy's salary is only capped at $70K until the company grows to a predetermined point, at which time HE gets a raise, presumably to a salary north of what he used to make. Not to mention that he OWNS at least part of the company, so even if he's not collecting a salary he can still collect on the stock dividends if he wants..

Businesses don't run an a vacuum. The issue is that if other businesses can do the same thing significantly cheaper or better, your business isn't as efficient as it should be. Inefficiency, if not controlled, will eventually kill a business. This guy has higher than average labor costs now.... That's not efficient.

Detection error.... Please check your hardware...