Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Finance Predictions for 2017 and Beyond?

The majority of Regulation in the EU is Protectionism. E.g. Rather than allow the food mountains in EU warehouses be shipped to countries who are short of food - the EU pour petrol on the food and destroy it.

Some of the Regulation in the US is also Protectionism (e.g. suppressing China's ability to be a technological powerhouse). Other Regulation is just measures that the rest-of-the-world would consider common sense.

If Trump does 'trade deals' that selectively reduces Protectionism, he will accumulate a wealth of new and powerful international friends. If he reduces common-sense Regulaton, his existing rich and powerful friends will rejoice.

The small 'politically insignificant' companies or those with a MO aimed at being 'Green' / good / healthy / inclusive are the one that will loose out. E.g. The foody companies that want to show that their products are the only ones that don't include pesticides, growth hormone, GM ingredients...

AJ
 
Dan Kanes said:
Antony, can you explain to me how the US regulates Chinese technology expansion? I don't see it at all but I'm open to be enlightened.
Click to expand...

I think he's referring to how US companies are loathe to share tech secrets with Chinese manufacturers and military, for fear of it being stolen. It's almost become a defacto US policy. China's refusal to honor copyright is a major hurdle for it's global financial relations.
 
Dan,

Supercomputers are required to 'back' long running simulations to bolster materials science, superconductor simulations, magnetic confinement simulations, Chip design, medication research, and environmental studies (almost everything these days!).

When China started prioritising technical advance, they decided they needed computers as 'Fast' as the ones that the US builds and sells. The US response to this was to do what it could to slow this rate of China's advance, like:
+) Prevent supercomputer builders from selling to the Chinese
+) Then after China amassed enough knowledge to build their own machines - The US blocked high end CPUs (like Intel upper end XEONs) from being sold to them.
+) China responded by trying to use 1,000,000 chips + huge amounts of electricity to try and create their own super, even though they only had access to 8 year old process technology .. and in doing so, even created their own CPU design.
+) The US got wind of China trying to send a researcher to talk about the 28nm design Chinese chip at a US supercomputer event - and promptly blocked their SOC scientist from presenting a paper on in in the US, and made their travel the US impossible... even though they were a Speaker at the event!
+) China have been trying very hard to buy a more modern chip plant so that they can make their own chips. All attempts that involve any modern tech (e.g. Memory), or China investing in SOC firms ... so that they don't have to build every component from scatchs has be stopped by the US.

In the short term this strategy IS working.
Even though the Chinese supercomputer is 125+ Petaflops (with a gazillion Chinese chips) vs 27 PF for the top US Super, the Chinese is super inefficient and getting at getting real work done (I think it's about 100 times Less efficient with is PTs than the Japanese super).

In the long term, there is something interesting brewing.
+) China have decided to throw huge amounts of money into building their own fabs, something they weren't motivated to do before.
+) China are rapidally building knowlege on ARM designs for their supers + home market. This will ultimately hurt Intel.
+) The US money going into materials science / researched into confinement fusion reactors / green energy / green subsidies may under the new administration be 'more challenging' to get hold of, but at the same time, China is very keen excel in it.

AJ
 
Last edited:
Antony Newman said:
Dan,

Supercomputers are required to 'back' long running simulations to bolster materials science, superconductor simulations, magnetic confinement simulations, Chip design, medication research, and environmental studies (almost everything these days!).

When China started prioritising technical advance, they decided they needed computers as 'Fast' as the ones that the US builds and sells. The US response to this was to do what it could to slow this rate of China's advance, like:
+) Prevent supercomputer builders from selling to the Chinese
+) Then after China amassed enough knowledge to build their own machines - The US blocked high end CPUs (like Intel upper end XEONs) from being sold to them.
+) China responded by trying to use 1,000,000 chips + huge amounts of electricity to try and create their own super, even though they only had access to 8 year old process technology .. and in doing so, even created their own CPU design.
+) The US got wind of China trying to send a researcher to talk about the 28nm design Chinese chip at a US supercomputer event - and promptly blocked their SOC scientist from presenting a paper on in in the US, and made their travel the US impossible... even though they were a Speaker at the event!
+) China have been trying very hard to buy a more modern chip plant so that they can make their own chips. All attempts that involve any modern tech (e.g. Memory), or China investing in SOC firms ... so that they don't have to build every component from scatchs has be stopped by the US.

In the short term this strategy IS working.
Even though the Chinese supercomputer is 125+ Petaflops (with a gazillion Chinese chips) vs 27 PF for the top US Super, the Chinese is super inefficient and getting at getting real work done (I think it's about 100 times Less efficient with is PTs than the Japanese super).

In the long term, there is something interesting brewing.
+) China have decided to throw huge amounts of money into building their own fabs, something they weren't motivated to do before.
+) China are rapidally building knowlege on ARM designs for their supers + home market. This will ultimately hurt Intel.
+) The US money going into materials science / researched into confinement fusion reactors / green energy / green subsidies may under the new administration be 'more challenging' to get hold of, but at the same time, China is very keen excel in it.

AJ
Click to expand...

Antony, I think you are over-looking what China is trying to do in the South China Sea. (They are attempting to encroach on US allies' maritime boundaries by creating man-made islands for military purposes and changing freedom to navigate laws now in force in the South China Sea.)

And as for not sharing secrets and technology, US firms also don't do that with other US firms. It's just good business, especially in the case of China which has been engaged in Industrial Espionage for the purpose of stealing US technology. Even if they were granted permission to use the US technology if they paid for it, they might find it cheaper to just steal it. '-)

And as long as they remain a potential enemy, I think it would be foolish for any country to try to raise your potential enemy to your level of expertise.
 
Elsie,

I was not trying to defend China's chequered history. :-)

There have been countries that have shared their 'crown jewels' of knowledge with a China and got their fingers somewhat burnt (e.g. Japan).

I was trying to show that the 'good business practice' to make it more difficult for international competitors is Protectionism.

AJ
 
Antony Newman said:
Elsie,

I was not trying to defend China's chequered history. :-)

There have been countries that have shared their 'crown jewels' of knowledge with a China and got their fingers somewhat burnt (e.g. Japan).

I was trying to show that the 'good business practice' to make it more difficult for international competitors is Protectionism.

AJ
Click to expand...

Oh... sorry, I misread the intent of your post.
 
I know that the highest end Intel Xeon chips (engineering samples no less) are readily available in Hong Kong, and I would then posit also in Mainland China... Maybe there are some ITARS restrictions against Intel direct selling their absolute top chips in Mainland China? seems to be no problem putting an S-Ton of Cherry Trail x86 SOCs into dirt cheap tablets built in mainland...

What about Nvidia (Taiwan) and TSMC / Global Foundries ? I guess that is Taiwan...but Taiwan is ROC?

Arguments could be made that Nvidia with ARM licensed technology combination are pushing the boundaries of "Neural Networks"

I believe we live in a deeply interconnected world and there is great dependence on one another (for better and sometimes worse). I pray we don't end up in a "cold war" of technology because nothing will slow advancement of technology like isolationism.

I'd also say - China has long been a technology leader - but maybe not always on the forefront of human rights. Gunpowder, metalsmithing, Porcelain (china?) to name a few ancient advances, but some of the things perpetrated on China by its own people and by colonials has been very cruel (human nature the world over I suppose).
 
The CIA term for the negative actions (for US companies) caused by US government commerce interference is "Blowback".

There is a general opinion that the explicit reasons the export bans were brought into effect (by the US dept of Commerce) do not stand stand up to scrutiny, and caused some embarrassment for the US companies involved (that were in the last stages of upgrades).

+) Intel's & AMD are effected by the Ban.
+) IBM was supposed to be prevented ... but they published the Open Power architecture that allowed China to roll their own.

There has been a positive knock on effect of the Chinese rapid progress (ie now that the worlds two fastest supercomputers are both Chinese & their fastest is 5x as fast as the top US super):
+) The US is getting ready to accelerate their own plans and bring 2023 forward to 2021.
+) The US is looking for speculative (novel) designs to over-invest in (for costly expedited ramp up)

With Intel taking its foot off the semiconductor R&D gas pedal, its current stranglehold on the markets Intel works in, and the multitude of mergers to consolidate costs - the world may well benefit from additional competition and funding.

Intel is spending circa 10 billion USD / year on R&D.
China (since the ban) is expecting to spend over 150 billion USD in order to design and build its own chips, own infrastructure and chip fabs. With that sort of backing - we might even start to see technology flowing the other direction!

To answer part of Elsie's first post:
If the incoming US government follows through on some of the semi-engineering hints of the previous administration, I would guess that 'big' money will be made available to invest in internationally competitive US infrastructure.

AJ

Edit : Thought I'd give an example of the US dept of commerce in action:

+) Germany has a metalorganic company (Aixtron) that allows creation of semiconductors and other nano materials using vapour deposition.
+) The company makes most of its money from selling to Asia + Europe.
+) China wanted to invest or buy the company.
+) Europe was happy with this + Compny happy with this + existing investors also happy.

The US blocked the sale - as the company has 'US ASSETS'.
Reason given : China buying ths German company would 'damage US National Security'.

My thoughts on this are:
+) a great many companies have 'assets' in the US.
+) if the EU is happy for a European based company to be sold to / or receive investment from China (or the UAE), and the US department of commerce intends to block all such actions ... then the US government should buy the company, and/or perform capital injection (something that Aixtron looks like it needs!)

Here's a link to the current US thinking.
http://www.eetimes.com/document.asp?doc_id=1331080&
 
Last edited:
You must log in or register to reply here.
Back
Top