• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

News [CNBC] SoftBank hires Goldman Sachs to explore sale options for chip designer Arm. . .

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Doug S

Senior member
Feb 8, 2020
326
455
96
People have been saying the end of Moore's Law is on the horizon for the 30 years I've been involved in tech. Probably longer, I just didn't hear it then.

It is easy to look at where we are and say "we don't know what we're going to do" after 5-10 years in the future where the roadmap is blank. Luckily there are people a lot smarter than clueless forum posters who are actually doing the research required to keep the train going. Yes someday we may run into fundamental physical limits, but we are a long way from reaching that point.
 

soresu

Golden Member
Dec 19, 2014
1,420
591
136
We've failed pretty hard in finding a replacement for the paradigm around silicon transistors for the last 2 decades
You're confusing the industries collective buy in on CMOS silicon with it being the only technology in town.

Even now there are better materials used for various things.

Silicon Carbide and III-V materials for power electronics is one salient example.
In this regard regular silicon gets left in the dust for extremely power dense applications like electric vehicles which have been pushing greater and greater power density per kilogram for the past decade.

Whether they already get used in PC power electronics I don't know, but I would be surprised if not in the future considering even <1KW PSU's still need cooling today, to say nothing of voltage control circuitry on graphics cards and such.

With the ATX12VO standard requiring more power circuitry on the mobo we may see some movement there too.

The military and NASA also use III-V materials for various things because it gives better performance despite cost. This extra cost is a combination of both materials and the economies of scale relating to the difference in investment between silicon and III-V materials.

These materials especially find use in electro optical (optical IO) components as silicon does not at all perform well in these applications.

Germanium also has potential both on its own and hybridised with silicon.

The oft used buzz word is graphene, but due to mechanical difficulties (zig zag patterns, nature is weird) in processing semiconducting graphene it has more potential IMHO for metal layer circuitry on a node, as copper is also not scaling well at this point either.

In general no single material is a king of all trades replacement for silicon at the moment, and cannot be until massive investment has been made in a potential successor - the likely outcome is a short term hybrid replacement material after some decision has been made about the best direction to pursue on a more radical future device and materials.

My personal bet would be on some variant of spin wave logic there, it can process multiple wave frequencies at once - essentially allowing for multi computation logic gates, which could be a nice win for density on top of the likely logic die stacking or monolithic 3D layering that will come from any spintronic move.

Vertical density has to be a significant consideration for next gen technologies as area scaling basically runs dry.
 

soresu

Golden Member
Dec 19, 2014
1,420
591
136
Taiwan has to prepare for a backup plan (if they even have one) in case their wells start drying up because even if they do manage to change the paradigm how much could they improve upon it ?
Due to the sheer costs involved now at developing brand new nodes on established device strictures, let alone new devices and radically different materials, I think it altogether likely that no single fab will be a paradigm changer.

One may have made a good bet on the best future direction and have a head start on competitors, but it seems likely that some level of singular decision, information and cost sharing will be involved - if for no other reason than the crippling cost of it all.
 

soresu

Golden Member
Dec 19, 2014
1,420
591
136
People have been saying the end of Moore's Law is on the horizon for the 30 years I've been involved in tech. Probably longer, I just didn't hear it then.

It is easy to look at where we are and say "we don't know what we're going to do" after 5-10 years in the future where the roadmap is blank. Luckily there are people a lot smarter than clueless forum posters who are actually doing the research required to keep the train going. Yes someday we may run into fundamental physical limits, but we are a long way from reaching that point.
Yes and no.

Silicon is definitely on its last 10-15 years at this point, and even so it is unlikely that it will ever scale nearly as well vertically as current leakage related thermal problems are just magnified in a stack/layered configuration without resorting to radical mechanical cooling solutions like ICEcool (micro fluidic thermal TSV's that push coolant through a 3D die/layer stack directly).

I'm not sure if spintronics and photonics will eliminate the thermal issue in the further off future, but I can certainly say that I would not be sorry to see bulky and loud HS/F solutions become a thing of the past in high end systems.

With SSD's rising in popularity more than ever with the coming console generation, such a change would essentially make the PSU the only significant noise maker in any system, and perhaps even future power electronics improvements I mentioned earlier will eliminate that problem too in time.
 

Doug S

Senior member
Feb 8, 2020
326
455
96
Yes and no.

Silicon is definitely on its last 10-15 years at this point, and even so it is unlikely that it will ever scale nearly as well vertically as current leakage related thermal problems are just magnified in a stack/layered configuration without resorting to radical mechanical cooling solutions like ICEcool (micro fluidic thermal TSV's that push coolant through a 3D die/layer stack directly).

I'm not sure if spintronics and photonics will eliminate the thermal issue in the further off future, but I can certainly say that I would not be sorry to see bulky and loud HS/F solutions become a thing of the past in high end systems.

With SSD's rising in popularity more than ever with the coming console generation, such a change would essentially make the PSU the only significant noise maker in any system, and perhaps even future power electronics improvements I mentioned earlier will eliminate that problem too in time.
Silicon may run out of gas but will eventually be replaced (though may remain as the substrate with the other materials layered on it since we know it so well)

But I don't care what comes, we will never see the end of needing to cool "high end systems". If you can do x amount of work while dissipating 5 or 10 watts and don't need active cooling in a PC, then you can do 10x amount of work while dissipating 50 or 100 watts. By definition a "high end system" will always push those boundaries. If we were lucky enough to find a way to greatly reduce the joules per MIP we'd use the same number of joules to get a lot more MIPS.

It has been possible to make near silent systems for a long time though. I'm typing this on an i5-6500 with a heatsink/fan from Zalman that weighs like a couple pounds. There's a case fan that's temperature operated that I don't think is running (can't feel any air from it at least) and the power supply is fanless.

It is sitting on my desk three feet away from me and I can only hear it if I sit perfectly silent, don't have any windows open, AC running or anything like that. This is not a 'high end system' (no external GPU etc.) but it would be more than adequate for probably 90-95% of PC users.
 

Vattila

Senior member
Oct 22, 2004
541
540
136
Off topic, on Taiwan vs China:

China thinks that Taiwan is part of its control.
The People's Republic of China (PRC) government — the Chinese Communist Party (CCP) in control of the China mainland — considers Taiwan to be Chinese territory. But, surprising to many I guess, so does the government of Taiwan, formally the Republic of China (ROC) government. At least, it actively did in the past. Today, with democracy, there are diverging views on Taiwan's status in different political parties and groups in Taiwan:

"The ROC government has in the past actively pursued the claim as the sole legitimate government over mainland China and Taiwan. This position began to change in the early 1990s as democracy was introduced and new Taiwanese leaders were elected, changing to one that does not actively challenge the legitimacy of PRC rule over mainland China. [...] Different groups have different concepts of what the current formal political situation of Taiwan is."


The devolved government of Taiwan has its roots in the ROC government of 1949, which fled the mainland after PRC took control in an uprising after the Second World War. Before that, ROC had ruled China since 1912, when ROC replaced the Qing dynasty.

So, both disputing governments have traditionally held a "united China" policy. The dispute has been over who is the legitimate government of China. Most countries, the UN and many other international organisations now recognise PRC as the government of China. Should the ROC government proclaim independence for Taiwan, and by that formally abandon the "united China" policy, there would presumably be a major escalation of the conflict.

Until then, or alternatively, until a peaceful unification, Taiwan is a Chinese territory with devolved governance — much like Scotland in United Kingdom.

 

moinmoin

Golden Member
Jun 1, 2017
1,781
1,789
106
But, surprising to many I guess, so does the government of Taiwan, formally the Republic of China (ROC) government.
That's wrong, it's not "the government" that did this. You have to make a distinct difference between the different ruling parties involved. The Kuomintang (KMT) has always been a proponent of "one China". KMT ruled all of China before the Chinese civil war (1945-49) forced it to retreat to the island Formosa/Taiwan and the Chinese Communist Party (CCP) started ruling Chinese mainland. That's why Taiwan is formally the continuation of the Republic of China that existed for all of China before the CCP founded the People's Republic of China.

Taiwan became a democracy after the martial law era (which lasted until late in the 1980s), and the other (and current) ruling party Democratic Progressive Party (DPP) never followed the "one China" policy, instead emphasizing Taiwan's distinct history and population. In their eyes the rulings over Taiwan by Qing dynasty (1683-1895), Japan (1895-1945), KMT (1945-1987) as well as CCP's claims on Taiwan are all considered as foreign rule. Considering the Qing dynasty was formally a Manchurian foreign rule itself, Taiwan factually has no real common Chinese history of being unified with Chinese mainland.

For that matter: CCP likes to push the maximum possible expansion of Chinese territory but keeps relying on the historic coverage of the foreign rules by the (Mongolian) Yuan dynasty as well as the (Manchurian) Qing dynasty as historic justification for their claims. CPP's behavior could be seen as a belated drive for colonization which in Chinese history knows no parallel.
 
  • Like
Reactions: Tlh97 and Vattila

A///

Senior member
Feb 24, 2017
771
516
106
China thinks that Taiwan is part of its control. They are currently dismantling the "one country, two systems" agreement with Hong Kong and Macau and has indicated interest to include Taiwan in that as well. China is building up military in the seas. I don't think China wants a war, they'll probably do it through economic paths (coerced financial agreements). But we can't rule out war as a possibility either.
Ok. They take over TSMC. Then what? The engineers can quit or sabotage everything. How will China get repair equipment? Materials? New machines? Designs? And what will most of the world's companies and countries who rely on TSMC do? Idly sit by?

If China can't get jack now, what makes you think taking over TSMC will work out for them? They'll quickly find themselves in the same position with companies forfeiting their orders. They don't know if something was altered. Why take the risk? They'll just go to Samsung then, who don't have the best track record. China gets locked out once more, tech gets dated. Machines need repair, no technicians are allowed to help China. No new parts. No new machines. No new plans. No new research. Nothing.

If China were so adamant on having Taiwan, they would have done it a long time ago. Countries didn't care about Hong Kong because the trade and benefit to them was less than marginal. That is the sad truth about the real world.

If countries wanted to mess with China they would:

Ban all food shipments to them.
Suspend their SWIFT status and any country aiding them
Suspend all oil shipments to the country. Russia may still supply them but even they would be pissed off.

Russia has fabs and fabless companies, but they specialize mostly in micro controllers or very out of date hardware for certain applications. Like many countries, they rely on TSMC and Samsung. Or rely on fabless companies from outside the country who rely on Samsung or TSMC.
 
Last edited:

ThatBuzzkiller

Senior member
Nov 14, 2014
979
127
106
For that matter: CCP likes to push the maximum possible expansion of Chinese territory but keeps relying on the historic coverage of the foreign rules by the (Mongolian) Yuan dynasty as well as the (Manchurian) Qing dynasty as historic justification for their claims. CPP's behavior could be seen as a belated drive for colonization which in Chinese history knows no parallel.
More like reunification and if what you said were true then why haven't they've disputed the territorial status of outer Manchuria or Mongolia despite being historically ruled by past Chinese dynasties ?

They want to bring Taiwan under them not because they like to do conquests but it's because they desire racial unity among Chinese people. The same logic holds true for the Tibetans and the Uyghurs for which they want them to assume a Chinese identity as well to promote the ideology of racial unity ...

America should stop interfering with China's greatest project because all of it's idea about a liberal democracy does is promote division among it's own citizens so it's China's turn to potentially show the right way ...
 

DrMrLordX

Lifer
Apr 27, 2000
16,371
5,282
136
The engineers can quit or sabotage everything.
Not if they want to live very long.

There are some potentially-ugly consequences for any action the PRC takes against Taiwan, even before considering TSMC. And that goes somewhat outside the scope of this discussion. Do consider that China might benefit from having exclusive access to the world's best foundry tech. Even if their personnel become . . . less-than-useful after the acquisition, rest assured that SMIC will step in to study everything and start working feverishly to replicate (and improve upon) TSMC's successes.
 

sdifox

No Lifer
Sep 30, 2005
80,550
7,300
126
More like reunification and if what you said were true then why haven't they've disputed the territorial status of outer Manchuria or Mongolia despite being historically ruled by past Chinese dynasties ?

They want to bring Taiwan under them not because they like to do conquests but it's because they desire racial unity among Chinese people. The same logic holds true for the Tibetans and the Uyghurs for which they want them to assume a Chinese identity as well to promote the ideology of racial unity ...

America should stop interfering with China's greatest project because all of it's idea about a liberal democracy does is promote division among it's own citizens so it's China's turn to potentially show the right way ...
ahahahaha
 

A///

Senior member
Feb 24, 2017
771
516
106
Not if they want to live very long.

There are some potentially-ugly consequences for any action the PRC takes against Taiwan, even before considering TSMC. And that goes somewhat outside the scope of this discussion. Do consider that China might benefit from having exclusive access to the world's best foundry tech. Even if their personnel become . . . less-than-useful after the acquisition, rest assured that SMIC will step in to study everything and start working feverishly to replicate (and improve upon) TSMC's successes.
You underestimate loyalty and hatred Taiwanese people have against the PRC's gov.

Except SMIC is somewhat behind. And even if they study, they can't make the equipment or train on the new equipment that TSMC has over what they have at SMIC. ASML is the leading company in EUV equipment for a reason. And they source their parts from a number of countries, all of which can be blacklisted from the markets at will. But, as you say, China attempting to take Taiwan would lead to severe repercussions best left discussed in the politics forum and not take this thread further off topic than it already has.


Maybe JHH will get into leather boots or pants.
 

ASK THE COMMUNITY