Tired of foreign domination of its telecom market share China...

crobato

Colonel
VIP Professional
Sigh you really have no idea how military-grade electronics parts are qualified. The qualification process is completely independent from the design process, and it makes no assumption the designed temperature range. It doesn't matter if the part will only see one end of the extreme. It will get qualified from the very low to the very high, cycled 500 times or more.

And how does that change the algorithms, the basic construction technologies. Tell me which military electronic technologies that isn't made of silicon or silicon on copper? Tell how why and how should temperature requirements force changes in the architecture? Address the question please, because your reply says nothing.

Somehow the idea that China can't do a decent datalink to a plane when they have accomplished all sorts of datalinks to satellites, space probes and spacecraft, all of whom under harsh environments and motion stress, is all beyond me.


No I don't. Google qualcomm R&D and you'll see they are not an IP shell company. You point your finger at Qualcomm's acquisitions, but these are aquisitions outside of its core CDMA chipset business, e.g. bluetooth, WLAN, even Eudora. True they receive 6b a year from CDMA royalties, but this the exact same business model by ARM and MIPS. If you dislike patents so much, why do you use patent application as an indicator of Huawei's technological "dominance"?

Look. For 3G stations and handsets, hundreds of patents have to be made from all sorts of different companies. Did you check the 3G license page? Tell me what is the proportion of Qualcomm's patents to well over 500 basic patents alone, not to mention another equal amount for supporting patents.

Sigh. Qualcomm buys a company that was intended to do wireless streaming of movies, then somehow uses that company's patents to make a claim on OFDM?

Companies are using patents for positional stakes. Maybe some of them are true innovations but others are just a repeat of what others are also trying to do. You're using patents to make a claim "I did it first" which means nothing with regards to the capabilities of other companies to reach that goal. And quite frankly if other companies patented the same idea on other countries, Qualcomm would not be able to enforce its royalty claims on those other countries.

The fact is all these companies from Docomo to Siemens to Ericsson to Nokia could have and would have perfectly developed 3G W-CDMA without Qualcomm. In fact, its possible each and every company could have developed the technologies independently but instead, you got a situation where companies end up suing each other where the sets of intellectual properties would intersect with each other. Then Qualcomm took it to itself to charge royalties as other companies have more or less settled and shared the royalties. Then Qualcomm itself gets sued by Broadcomm.

Another example of the nonsense going on is so called "multitouch" technology Apple claims for its iPhone. Just because Apple did it, does not mean that others cannot accomplish the same thing. But now, you use the patent process to prevent others from incorporating the same feature on their own handsets.

Your entire point of bringing up Qualcomm is to suggest without Qualcomm, no one could have developed 3G technology. That's nonsense. Everyone could have developed it independently sooner or later. The patents is about the sooner guys trying to force the later guys to pay up for the disadvantage of being late and has nothing to do with development or technological capability.

Huawei's and ZTE's patent staking is of the same idea but they're using it to stake even future technologies from 4G to beyond. There is no point in reinventing 2G and 3G. The point of all those few thousand patents is to stake 4G so they don't have to pay royalties, or pay less royalties, or have others pay them for royalties or cross license them with other patents.

You are missing the point. Where are the real key components developed and made?

And you're missing the point too. Its an American jet. Tell me if it uses Russian engines or Japanese developed flight control systems either.
 
Last edited:

oringo

Junior Member
And how does that change the algorithms, the basic construction technologies. Tell me which military electronic technologies that isn't made of silicon or silicon on copper? Tell how why and how should temperature requirements force changes in the architecture? Address the question please, because your reply says nothing.

I have answered that before, you just keep missing the main idea here. What I basically said is that you can't use the cutting edge commercial parts to measure the amount of technology that goes on a military airplane. These commercial parts run too hot and are not as reliable as the parts fabbed using, say 0.5um process.

The answer has nothing to do with algorithms. Take a multi-ghz processor that can do blue-ray decoding on the fly. Chances are this processor cannot operate at 125C ambient. Now dial the frequency down to 133MHz. The chance of the process running is lot better now because of the reduced operating frequency. Can you expect it to decode blue-ray at real time running at 133MHz? The answer is no. It can still do simpler things, but probably not any better than a RAD750.

Somehow the idea that China can't do a decent datalink to a plane when they have accomplished all sorts of datalinks to satellites, space probes and spacecraft, all of whom under harsh environments and motion stress, is all beyond me.

I never said that before. All I'm saying is that what the commercial companies achieved with 3/4G wireless technology has little impact on the real military electronics.

Look. For 3G stations and handsets, hundreds of patents have to be made from all sorts of different companies. Did you check the 3G license page? Tell me what is the proportion of Qualcomm's patents to well over 500 basic patents alone, not to mention another equal amount for supporting patents.

Sigh. Qualcomm buys a company that was intended to do wireless streaming of movies, then somehow uses that company's patents to make a claim on OFDM?

My argument was never based on how many patents each company holds (that was your guage for how successful a company is). All I'm saying is that Huawei still relies on the chips from western countries when it comes to the key technology. Show me a real published ASIC implementation from Huawei to prove me wrong.
 

Schumacher

Senior Member
....... All I'm saying is that Huawei still relies on the chips from western countries when it comes to the key technology. Show me a real published ASIC implementation from Huawei to prove me wrong.

Hmm .... I think you have it backward, it actually should go something like you showing how "Huawei still relies on the chips from western countries when it comes to the key technology." to prove you right.
 

crobato

Colonel
VIP Professional
I have answered that before, you just keep missing the main idea here. What I basically said is that you can't use the cutting edge commercial parts to measure the amount of technology that goes on a military airplane. These commercial parts run too hot and are not as reliable as the parts fabbed using, say 0.5um process.

The answer has nothing to do with algorithms. Take a multi-ghz processor that can do blue-ray decoding on the fly. Chances are this processor cannot operate at 125C ambient. Now dial the frequency down to 133MHz. The chance of the process running is lot better now because of the reduced operating frequency. Can you expect it to decode blue-ray at real time running at 133MHz? The answer is no. It can still do simpler things, but probably not any better than a RAD750.

And so?

If you have to rely on chips using the 0.5um process, guess what, the technology level of the chips then are also much lower as a result. In which case the technology of the software would also have to scale down as a result. And because the overall technology that has to be implemented is generations behind, so what's the problem of down scaling from the more advanced technologies that you have?

Its either that or you add more chips.

But then again, figure out why militaries are now relying more and more on COTS. COTS as in Commercial Off the Shelf Technology. That's right they're relying more or less the same technologies and chips you get from the commercial areas. Look at the latest screens to be used on AEGIS and other air defense ships for example. These terminals are just running off Windows and Pentiums. The reason why there is more and more reliance on COTS is because militaries are finding out that their byzantine and bureaucratic process of certifying are overly redundant, complex and takes far too long, too long that Moore Cycles are happening multiple times on commercial technologies while a military part waits to be certified, leaving the military part, generations behind.

I never said that before. All I'm saying is that what the commercial companies achieved with 3/4G wireless technology has little impact on the real military electronics.

Um, wrong...and bingo big time.

Things like spread spectrum, frequency hopping, pseudo random, and LPI techniques are common and used in commercial telephonies every day, every second, and you don't need the radar off an F-22 Raptor to have a working sample of such technologies when you can already find it in the device inside your pocket.


My argument was never based on how many patents each company holds (that was your guage for how successful a company is). All I'm saying is that Huawei still relies on the chips from western countries when it comes to the key technology. Show me a real published ASIC implementation from Huawei to prove me wrong.

Show me published ASIC implementation from Alcatel backhaul or CISCO router or IBM server then. You won't find it. You only get published material if these companies are in the business of selling chipsets. You won't get it if these chipsets represent company secrets and their competitive edge. The chipsets Huawei are getting are most probably for products that are not their core businesses like handsets or wi fi routers. Want to show me who sells the chipsets for a backbone Internet super router then?
 
Last edited:

oringo

Junior Member
But then again, figure out why militaries are now relying more and more on COTS. COTS as in Commercial Off the Shelf Technology. That's right they're relying more or less the same technologies and chips you get from the commercial areas. Look at the latest screens to be used on AEGIS and other air defense ships for example. These terminals are just running off Windows and Pentiums. The reason why there is more and more reliance on COTS is because militaries are finding out that their byzantine and bureaucratic process of certifying are overly redundant, complex and takes far too long, too long that Moore Cycles are happening multiple times on commercial technologies while a military part waits to be certified, leaving the military part, generations behind.

Stop defining terms to prove you are smart. There is no such thing as Moor Cycle. Everyone knows what COTS is. Mentioning the acronym doesn't get you very far to prove you know something.

Show me published ASIC implementation from Alcatel backhaul or CISCO router or IBM server then. You won't find it. You only get published material if these companies are in the business of selling chipsets. You won't get it if these chipsets represent company secrets and their competitive edge. The chipsets Huawei are getting are most probably for products that are not their core businesses like handsets or wi fi routers. Want to show me who sells the chipsets for a backbone Internet super router then?

To both Crobato and Shumacher:

Google IEEE ISSCC. Most major ASIC players publish their new ASIC chips there. That's the big player conference to show off you latest ASIC developments. IBM, Intel, Qualcomm, AMD, nVidia can all be seen there. I don't find a single technical announcement from Huawei.

On the contrary to what you believe, even if you don't intend to sell the chips, you still publish material to establish your technical prestige.
 

RedMercury

Junior Member
Remember you are not the only one reading the posts and a little extra explanation does not hurt. As to Moore's law, well, I expect that to be basic knowledge if you're trying to talk about EE?
 

crobato

Colonel
VIP Professional
Stop defining terms to prove you are smart. There is no such thing as Moor Cycle. Everyone knows what COTS is. Mentioning the acronym doesn't get you very far to prove you know something.

If you don't know what the F**K Moore's Cycle is, don't bother responding to the topic again.

To both Crobato and Shumacher:

Google IEEE ISSCC. Most major ASIC players publish their new ASIC chips there. That's the big player conference to show off you latest ASIC developments. IBM, Intel, Qualcomm, AMD, nVidia can all be seen there. I don't find a single technical announcement from Huawei.

On the contrary to what you believe, even if you don't intend to sell the chips, you still publish material to establish your technical prestige.

WRONG. CISCO, Alcatel, Juniper Networks, Siemens, Ericsson, don't publish what is in the chipsets they use on their back end routers or their back haul wireless equipment. You only publish if you sell chipsets for use in end user equipment.

Neither does IBM. What IBM publishes are reference material for their processors and systems for use in programming, but I have never seen IBM ever published the chipsets they use in their high end mainframes and servers.

Intel, Qualcomm, AMD and nVidia are in the busy of selling chipsets. Not making end equipment themselves.
 

victtodd

New Member
If you don't know what the F**K Moore's Cycle is, don't bother responding to the topic again.



WRONG. CISCO, Alcatel, Juniper Networks, Siemens, Ericsson, don't publish what is in the chipsets they use on their back end routers or their back haul wireless equipment. You only publish if you sell chipsets for use in end user equipment.

Neither does IBM. What IBM publishes are reference material for their processors and systems for use in programming, but I have never seen IBM ever published the chipsets they use in their high end mainframes and servers.

Intel, Qualcomm, AMD and nVidia are in the busy of selling chipsets. Not making end equipment themselves.

My admiration for your knowledge and logic sees no end.....:coffee:

On a side note, even as a layman as ignorant as I am, I do have heard about Moore's Cycle.:D
 

alopes

Junior Member
I also know nothing about electronics.

But isn´t making a fighter plane radar and fire control system as complicated, in terms of engineering as making
a military datalynk system.

So, if China can develope and implement a fighter´s fire control system and radars, isn´t plausible that they can also
create a modern datalynk system?

:coffee:
 

crobato

Colonel
VIP Professional
They can. Much of the datalinks you hear about the military was based in 1980s technology. And they keep it so because you have a large established base of equipment that uses the datalink protocol. Going to a more advanced datalink protocol actually forces an expensive round of upgrades to all your legacy equipment. This is something many militaries cannot afford. And remember, if ain't broke, don't fix it. Militaries are by nature, very technologically conservative.

If you have an absence of existing legacy, its actually much easier to start with a more advanced system. This is the advantage of having a clean slate.

Among Chinese aircraft that uses Beidou, Beidou for example is more than just a satellite positioning system. Unlike GPS, Beidou also allows a measure of digital communications, such as SMS. It won't carry voice, but you can transmit digital data on it, which means it can be used for a computer to computer interchange. That's what a data link means, a computer to computer communication.
 
Top