AMD's Future Chips & SoC's: News, Info & Rumours.

[jaymc]

Spoiler

..just thinking its been about 10 years to get from 4.0 Overclocked to 5.0 overclocked.

That has a name: frequency wall

Not quite. It's a self-imposed wall. You can get higher speeds if you're willing to consume more power to achieve it.

I can totally see IBM going for 6Ghz if they so want to if the silicon allows within a reasonable power envelope.

Wrong. The frequency wall is a well-know limit of the laws of physics. Wait sitting for those 6GHz IBM chips. LOL

This is self fulfilling prophecy for everything known, and unknown is physics! As with anything, it's only a matter of time before what was once thought impossible is made reality.

The laws of physics don't stop working when someone desires to ignore them.

Au contraire ! Actually I believe they can.. An here's two examples:

Quantum Entanglement springs to mind. I couldn't find this one it goes something like this:
If you split a photon of light with a prism an shoot one half of the photon through a shudder.. .When you observe the experiment what you think should happen does an one shadow is cast but if you ignore and record it they both cast the same shadow.

Also Here's another experiment I did find that proves you can alter reality or physics just by observing it.. The double slit experiment.
To quote Morgan Freeman in the video "This suggests that we can change the way reality behaves just by looking at it"
https://www.youtube.com/watch?v=ayvbKafw2g0

It appears that what we think has an effect on our reality.. That's "spooky" Interaction for sure.
So that's why something is always in the last place ya look... It ran out of probable places to hide in. Ah that makes perfect sense now !
Stay positive people lol

 

[gamerk316]

Spoiler

..just thinking its been about 10 years to get from 4.0 Overclocked to 5.0 overclocked.

That has a name: frequency wall

Not quite. It's a self-imposed wall. You can get higher speeds if you're willing to consume more power to achieve it.

I can totally see IBM going for 6Ghz if they so want to if the silicon allows within a reasonable power envelope.

Wrong. The frequency wall is a well-know limit of the laws of physics. Wait sitting for those 6GHz IBM chips. LOL

This is self fulfilling prophecy for everything known, and unknown is physics! As with anything, it's only a matter of time before what was once thought impossible is made reality.

The laws of physics don't stop working when someone desires to ignore them.

Au contraire ! Actually I believe they can.. An here's two examples:

Quantum Entanglement springs to mind. I couldn't find this one it goes something like this:
If you split a photon of light with a prism an shoot one half of the photon through a shudder.. .When you observe the experiment what you think should happen does an one shadow is cast but if you ignore and record it they both cast the same shadow.

Also Here's another experiment I did find that proves you can alter reality or physics just by observing it.. The double slit experiment.
To quote Morgan Freeman in the video "This suggests that we can change the way reality behaves just by looking at it"
https://www.youtube.com/watch?v=ayvbKafw2g0

It appears that what we think has an effect on our reality.. That's "spooky" Interaction for sure.
So that's why something is always in the last place ya look... It ran out of probable places to hide in. Ah that makes perfect sense now !
Stay positive people lol

Not quite.

In your examples, the laws of physics aren't changing. What happens is the act of observing the experiment has an ever so slight effect upon it, changing the result.

 

[goldstone77]

Spoiler

..just thinking its been about 10 years to get from 4.0 Overclocked to 5.0 overclocked.

That has a name: frequency wall

Not quite. It's a self-imposed wall. You can get higher speeds if you're willing to consume more power to achieve it.

I can totally see IBM going for 6Ghz if they so want to if the silicon allows within a reasonable power envelope.

Wrong. The frequency wall is a well-know limit of the laws of physics. Wait sitting for those 6GHz IBM chips. LOL

This is self fulfilling prophecy for everything known, and unknown is physics! As with anything, it's only a matter of time before what was once thought impossible is made reality.

The laws of physics don't stop working when someone desires to ignore them.

Au contraire ! Actually I believe they can.. An here's two examples:

Quantum Entanglement springs to mind. I couldn't find this one it goes something like this:
If you split a photon of light with a prism an shoot one half of the photon through a shudder.. .When you observe the experiment what you think should happen does an one shadow is cast but if you ignore and record it they both cast the same shadow.

Also Here's another experiment I did find that proves you can alter reality or physics just by observing it.. The double slit experiment.
To quote Morgan Freeman in the video "This suggests that we can change the way reality behaves just by looking at it"
https://www.youtube.com/watch?v=ayvbKafw2g0

It appears that what we think has an effect on our reality.. That's "spooky" Interaction for sure.
So that's why something is always in the last place ya look... It ran out of probable places to hide in. Ah that makes perfect sense now !
Stay positive people lol

Not quite.

In your examples, the laws of physics aren't changing. What happens is the act of observing the experiment has an ever so slight effect upon it, changing the result.

The laws of physics as we understand them only change under extreme temperature and pressure, like in a singularity. But in general all physics, what we know and don't know, it is all still physics. Our understanding of it is what changes.

 

[goldstone77]

This guy does some great work creating these models.

 

[YoAndy]

This guy does some great work creating these models.

 

[jaymc]

Dell Announces it's Epyc Lineup:

https://dellemcpartnerprogram.emailsrvc.net/pages/12558-7746/en_US/

 

[jaymc]

Spoiler

..just thinking its been about 10 years to get from 4.0 Overclocked to 5.0 overclocked.

That has a name: frequency wall

Not quite. It's a self-imposed wall. You can get higher speeds if you're willing to consume more power to achieve it.

I can totally see IBM going for 6Ghz if they so want to if the silicon allows within a reasonable power envelope.

Wrong. The frequency wall is a well-know limit of the laws of physics. Wait sitting for those 6GHz IBM chips. LOL

This is self fulfilling prophecy for everything known, and unknown is physics! As with anything, it's only a matter of time before what was once thought impossible is made reality.

The laws of physics don't stop working when someone desires to ignore them.

Au contraire ! Actually I believe they can.. An here's two examples:

Quantum Entanglement springs to mind. I couldn't find this one it goes something like this:
If you split a photon of light with a prism an shoot one half of the photon through a shudder.. .When you observe the experiment what you think should happen does an one shadow is cast but if you ignore and record it they both cast the same shadow.

Also Here's another experiment I did find that proves you can alter reality or physics just by observing it.. The double slit experiment.
To quote Morgan Freeman in the video "This suggests that we can change the way reality behaves just by looking at it"
https://www.youtube.com/watch?v=ayvbKafw2g0

It appears that what we think has an effect on our reality.. That's "spooky" Interaction for sure.
So that's why something is always in the last place ya look... It ran out of probable places to hide in. Ah that makes perfect sense now !
Stay positive people lol

Not quite.

In your examples, the laws of physics aren't changing. What happens is the act of observing the experiment has an ever so slight effect upon it, changing the result.

The laws of physics as we understand them only change under extreme temperature and pressure, like in a singularity. But in general all physics, what we know and don't know, it is all still physics. Our understanding of it is what changes.

The act of observation changes the result of the experiment, explain that ?

Come on it's a head melter, I'm having some fun with you guy's... Merry Christmass an all that

 

[-Fran-]

The act of observation changes the result of the experiment, explain that ?

Come on it's a head melter, I'm having some fun with yis guy's... Merry Christmass an all that

In Experimental Physics, the "observing" part involves taking a look where light usually doesn't like traveling so easily, so your means of "observation" usually involve radiation of some kind. Hell, even using a simple light bulb implies you're irradiating energy of some sort into your observation strictly speaking, so scientifically, you're "dirtying" the measurement.

The simple example out teach (one hell of an experimental physicist back in Uni) told us all the time was low oxidation of plaques under some sort of current. If you wanted to see how the molecules had to re-arrange, you had to be careful on ALWAYS taking into account the type of microscope you used so you didn't, quite literally, broke the chain of molecules due to radiation.

Also, keep in mind the lower/smaller you go, subatomic particles are hell fast and hard to observe with any known means. So you have to develop (or use) mathematical models to predict where they might be at any point in time and if you try to observe, you will most likely affect trajectory for them. So you need to rely on the mathematical models more often than not and it's just experimentation to correct and complement these predictive models.

Cheers!

 

[jaymc]

Well put Yuka... But that doesn't explain why the particles go where we think they should when observed ?

 

[jaymc]

CES 2018 PC preview: What to expect from laptops, desktops and tablets:

https://www.cnet.com/news/ces-2018-computer-preview-what-to-expect-from-laptops-desktops-and-tablets/

 

[-Fran-]

Well put Yuka... But that doesn't explain why the particles go where we think they should when observed ?

It's because of those models. The models themselves don't give you a 100% certainty, unless you put a lot of assumptions, which are usually un-true anyway. So you have "probability clouds" for electron movement (instead of a single point, you have several per measurement tick - and also... time plays a role in the measurements, lol). Also, models take into account most known effects, but there's a lot of things we don't know yet or haven't "seen" to incorporate into them. That is what is holding advancement now and why they¿re experimenting with new composites. You can't correct for things you can't really "see", so you have to go with "what you know". Tunneling was theorized a long long time ago, but only proven recently and it's still a work in progress on how to keep it "under check", for example.

And in statistical terms, when you have 10 observations with only 1 that doesn't follow the pattern, you can decide whether or not the model is suitable for what you need to if you can find why 1 of them didn't go according to the model.

Cheers!

 

[jaymc]

Physicists have been debating over these questions for over a hundred years I don't think anybody has a definitive answer yet.
Quantum Mechanics is the mother of all modern technology an the make up of our reality an universe yet it seem's to behave very strangely under certain conditions. It apears that our reality is much more flexible than we had previously realized. I wonder if we can bend it or manipulate it as we do the particles. This is the burning question on everyones mind. Einstein refused to accept that two entangled paricles were not in a fixed state before observed an he died before it was proven to be true. It was an Irish Physicist Bell who developed a way to prove it. Einstein said "God does not roll dice" an "I want to know that the moon is still there when I'm not looking at it". But he was proven wrong an apparantly the entangled particles are in all proboble states until measured.. This would of melted his head.

 

[-Fran-]

Physicists have been debating over these questions for over a hundred years I don't think anybody has a definitive answer yet.
Quantum Mechanics is the mother of all modern technology an the make up of our reality an universe yet it seem's to behave very strangely under certain conditions. It apears that our reality is much more flexible than we had previously realized. I wonder if we can bend it or manipulate it as we do the particles.

“The more you know, the more you know you don't know.” .- Aristotle

Only really smart people knows they know very little in reality. Whoever tells you otherwise is a major idiot.

But to summarize it, experimentation is expensive, that is why Intel and other foundries have such massive budgets exclusive for R&D in these areas: you can only get better by experimenting. Theory only gets you so far and you destroy more theories than confirm.

Cheers!

 

[goldstone77]

Physicists have been debating over these questions for over a hundred years I don't think anybody has a definitive answer yet.
Quantum Mechanics is the mother of all modern technology an the make up of our reality an universe yet it seem's to behave very strangely under certain conditions. It apears that our reality is much more flexible than we had previously realized. I wonder if we can bend it or manipulate it as we do the particles.

“The more you know, the more you know you don't know.” .- Aristotle

Only really smart people knows they know very little in reality. Whoever tells you otherwise is a major idiot.

But to summarize it, experimentation is expensive, that is why Intel and other foundries have such massive budgets exclusive for R&D in these areas: you can only get better by experimenting. Theory only gets you so far and you destroy more theories than confirm.

Cheers!

The great thing is that we can simulate a lot more today. You can build and test circuits on your cell phone with an app. As technology grows it lifts up everything.

 

[goldstone77]

This guy does some great work creating these models.

It's about 1.69X more dense than Samsung and about 1.49X more dense the TSMC for 10nm. At 7nm Samsung looks like they will have the smallest process, followed by TSMC, and then GlobalFoundries.

 

[goldstone77]

IEDM 2017: GlobalFoundries 7nm process; Cobalt, EUV

There are 5 pages of information showing the progression of GlobalFoundries process. It really has improved considerably from 14nm.

 

[8350rocks]

Adding 50% moar cores and ~1GHz would increase the limit to around 350W.

This is what took you 5 minutes and is the basis for your argument for the slide being fake... .. . . . Care to explain? AM4 has a 140W+ TDP and the 1800X is a 8 core with base clock 3.6GHz and TDP of 95W.

That is only a part of my argument. The other part is where I mention the physical impossibility of 12LP reducing this wattage by a factor of ~3x to fit within the AM4 socket spec limits.

Per spec. the "AMD socket AM4 reference limit" is 128W.

140W is the rating for some coolers.

The '95W' of the 1800X is a marketing value that doesn't correspond to the real TDP. The real TDP is higher as confirmed by reviews.

TDP != Power draw

This has been discussed multiple times.

Does the 7900X have a TDP of 315W? No?

Case closed.

 

[adamsleath]

mmmmmm some great pics and info posted thanks for that!!!!

but the 7nm yields are really the issue yeah? whoever wins the yield race between 10nm intel and '7nm' glofo . and which will have the performance crown. if 7nm proves to be very expensive due to yield problems... anyway. i guess it will be what it will be.

great to see the info. keep it coming!

 

[jaymc]

Here's a great article from our very own Paul Alcorn very informative.

The Numbers Behind AMD's 2017 Ryzen Insurgency:
http://www.tomshardware.co.uk/amd-ryzen-intel-desktop-pc-market-share,news-57577.html

 

[goldstone77]

We have a 55% increase in fin height from 14nm(27nm height) and 7nm(42nm height)

Edit: Notice the much better rectangular shape of 7nm vs. the 14nm with slightly rounded corners at the top. That's what you want!

 

[jaymc]

Microsoft highlights the company's experience with EPYC:

https://twitter.com/AMDServer/status/944002124563795968

 

[juanrga]

Spoiler

..just thinking its been about 10 years to get from 4.0 Overclocked to 5.0 overclocked.

That has a name: frequency wall

Not quite. It's a self-imposed wall. You can get higher speeds if you're willing to consume more power to achieve it.

I can totally see IBM going for 6Ghz if they so want to if the silicon allows within a reasonable power envelope.

Wrong. The frequency wall is a well-know limit of the laws of physics. Wait sitting for those 6GHz IBM chips. LOL

This is self fulfilling prophecy for everything known, and unknown is physics! As with anything, it's only a matter of time before what was once thought impossible is made reality.

The laws of physics don't stop working when someone desires to ignore them.

Au contraire ! Actually I believe they can.. An here's two examples:

Quantum Entanglement springs to mind. I couldn't find this one it goes something like this:
If you split a photon of light with a prism an shoot one half of the photon through a shudder.. .When you observe the experiment what you think should happen does an one shadow is cast but if you ignore and record it they both cast the same shadow.

Also Here's another experiment I did find that proves you can alter reality or physics just by observing it.. The double slit experiment.
To quote Morgan Freeman in the video "This suggests that we can change the way reality behaves just by looking at it"
https://www.youtube.com/watch?v=ayvbKafw2g0

It appears that what we think has an effect on our reality.. That's "spooky" Interaction for sure.
So that's why something is always in the last place ya look... It ran out of probable places to hide in. Ah that makes perfect sense now !
Stay positive people lol

Quantum entanglement and double slit pattern is a consequence of laws of physics known for 100 years. Using those laws we can do a computer simulation of the double slit phenomena. Those are the trajectories of the quantum particles through the slits

 

[juanrga]

This guy does some great work creating these models.

Incorrect data, as shown in the Intel thread.

 

[juanrga]

Adding 50% moar cores and ~1GHz would increase the limit to around 350W.

This is what took you 5 minutes and is the basis for your argument for the slide being fake... .. . . . Care to explain? AM4 has a 140W+ TDP and the 1800X is a 8 core with base clock 3.6GHz and TDP of 95W.

That is only a part of my argument. The other part is where I mention the physical impossibility of 12LP reducing this wattage by a factor of ~3x to fit within the AM4 socket spec limits.

Per spec. the "AMD socket AM4 reference limit" is 128W.

140W is the rating for some coolers.

The '95W' of the 1800X is a marketing value that doesn't correspond to the real TDP. The real TDP is higher as confirmed by reviews.

TDP != Power draw

This has been discussed multiple times.

Does the 7900X have a TDP of 315W? No?

Case closed.

TDP = Power draw, by the first law of thermo.

The i9-7900X has a TDP of 140W. The R7-1800X has a TDP of 128W. The TR-1950X has a TDP of 180W. This is all well-known now.

Here's a great article from our very own Paul Alcorn very informative.

The Numbers Behind AMD's 2017 Ryzen Insurgency:
http://www.tomshardware.co.uk/amd-ryzen-intel-desktop-pc-market-share,news-57577.html

So he confirms what I said about Zen only providing 2% marketshare to AMD. 😀

 

[-Fran-]

TDP = Power draw, by the first law of thermo.

The i9-7900X has a TDP of 140W. The R7-1800X has a TDP of 128W. The TR-1950X has a TDP of 180W. This is all well-known now.

Spoiler

Sigh... The first law would only apply here if the system was closed... The measurements you take for CPUs is hardly what you could classify as a "closed" system where you have no loss of energy.

Here's a great article from our very own Paul Alcorn very informative.

The Numbers Behind AMD's 2017 Ryzen Insurgency:
http://www.tomshardware.co.uk/amd-ryzen-intel-desktop-pc-market-share,news-57577.html

So he confirms what I said about Zen only providing 2% marketshare to AMD. 😀

No, he doesn't.