Zen 3 is rumored to be flaunting monumental IPC gains in early testing

kamikazi

Gawd
Joined
Jan 19, 2006
Messages
659
If true, this has to be game over for Intel, especially on the server side. I know that Intel has the ecosystem and established IT infrastructure in place, but those kids of gains are too much to ignore.
 

Ready4Dis

2[H]4U
Joined
Nov 4, 2015
Messages
2,366
Probably not game over, but the bleeding will continue. Intel is supposedly getting some pretty decent IPC gains in their new generation.... who knows if/when it'll be available however and if they will ever make a desktop part :). I'm curious if their IPC gain is going to finally close the last gap they have, gaming. If their IPC gain closes this gap to Intel, there will be 0 reason to ever go with Intel, as opposed to right now where theirs a very small use case, but still hard to advise anyone to go with Intel in todays market.
 

kamikazi

Gawd
Joined
Jan 19, 2006
Messages
659
Probably not game over, but the bleeding will continue. Intel is supposedly getting some pretty decent IPC gains in their new generation.... who knows if/when it'll be available however and if they will ever make a desktop part :). I'm curious if their IPC gain is going to finally close the last gap they have, gaming. If their IPC gain closes this gap to Intel, there will be 0 reason to ever go with Intel, as opposed to right now where theirs a very small use case, but still hard to advise anyone to go with Intel in todays market.

You're right, I should have said game over for the next couple of years. Intel is a sleeping giant who for sure has been awakened at this point. No more screwing around with 10nm and milking the market with miniscule IPC upgrades. I don't see them striking back a vengeance for a couple of years, though.
 
  • Like
Reactions: kac77
like this

Revenant_Knight

Limp Gawd
Joined
Nov 18, 2011
Messages
431
Probably not game over, but the bleeding will continue. Intel is supposedly getting some pretty decent IPC gains in their new generation.... who knows if/when it'll be available however and if they will ever make a desktop part :). I'm curious if their IPC gain is going to finally close the last gap they have, gaming. If their IPC gain closes this gap to Intel, there will be 0 reason to ever go with Intel, as opposed to right now where theirs a very small use case, but still hard to advise anyone to go with Intel in todays market.


But what if someone’s favorite color is blue?! ;)
 

tangoseal

[H]F Junkie
Joined
Dec 18, 2010
Messages
8,942
If true, this has to be game over for Intel, especially on the server side. I know that Intel has the ecosystem and established IT infrastructure in place, but those kids of gains are too much to ignore.

Are you kidding me?

Intel can just drop 30 billion on new research and SMOKE amd.

upload_2019-12-30_1-0-40.png


upload_2019-12-30_1-0-51.png
 

Rizen

[H]F Junkie
Joined
Jul 16, 2000
Messages
9,330
Are you kidding me?

Intel can just drop 30 billion on new research and SMOKE amd.

View attachment 211805

View attachment 211806
Spending money on research doesn't automatically translate into a great product, though. You can invest a lot of money and still make a mistake (example: Intel 10nm).

A ton of Intel's R&D is already allocated to their fab technology, which AMD doesn't have to deal with. Obviously this has pros and cons but Intel's expenditures on R&D aren't solely on new architecture.

AMD is clearly the underdog, but the financials on their own don't tell the whole story.
 

tangoseal

[H]F Junkie
Joined
Dec 18, 2010
Messages
8,942
Spending money on research doesn't automatically translate into a great product, though. You can invest a lot of money and still make a mistake (example: Intel 10nm).

A ton of Intel's R&D is already allocated to their fab technology, which AMD doesn't have to deal with. Obviously this has pros and cons but Intel's expenditures on R&D aren't solely on new architecture.

AMD is clearly the underdog, but the financials on their own don't tell the whole story.

This is very true. All I was pointing out was the massive market cap Intel owns compared to AMD. However, AMD might catch up with Intel's market valuation soon enough if they keep this level of engineering new stuff up!
 

Gideon

2[H]4U
Joined
Apr 13, 2006
Messages
2,737
This is very true. All I was pointing out was the massive market cap Intel owns compared to AMD. However, AMD might catch up with Intel's market valuation soon enough if they keep this level of engineering new stuff up!

General Electric used to be worth a lot as well in stock valuation and now they are just trying to survive these days without going bankrupt. Market valuation has very little to do with how much free cash they have on hand to do such a investment and two research never guarantees any success. Often as a company becomes bigger they try to find more ways to diversify their income and make more, however that strains the company from their original focus and forces more of the R&D budget to go elsewhere.
 

toast0

Gawd
Joined
Jan 26, 2010
Messages
1,019
You're right, I should have said game over for the next couple of years. Intel is a sleeping giant who for sure has been awakened at this point. No more screwing around with 10nm and milking the market with miniscule IPC upgrades. I don't see them striking back a vengeance for a couple of years, though.

So... Jim Keller (re)joined AMD in 2012, and Zen 2 smoked Intel in 2019. He joined Intel in 2018, so I expect results by 2025 ;) ... Maybe sooner, since Intel of 2018 has a lot more resources than AMD of 2012. Intel's fab trouble has really set it back, because of the long design pipelines.
 

Mega6

2[H]4U
Joined
Aug 13, 2017
Messages
2,760
posted this in the Intel section, seems relevant here.

AMD balance sheet is actually superior to Intel.

Intel
total cash: 12.02B
total debt: 29.48B

AMD
total cash 1.21B
total debt 1.11B

And yes, intel is behind, and falling further away rapidly. the 5GHz / 14nm wall is hitting them hard. Revenue is falling, they are propping up the share price with buybacks in the billions instead of using it for R&D. Intel spent spent $4.5 billion on share repurchases during the third quarter versus $3.2 billion on R&D. Do you know that the CEO of intel is a Finance guy? Not an Engineer like the past. This shows you Intel's true direction.
 

somebrains

[H]ard|Gawd
Joined
Nov 10, 2013
Messages
1,360
General Electric used to be worth a lot as well in stock valuation and now they are just trying to survive these days without going bankrupt. Market valuation has very little to do with how much free cash they have on hand to do such a investment and two research never guarantees any success. Often as a company becomes bigger they try to find more ways to diversify their income and make more, however that strains the company from their original focus and forces more of the R&D budget to go elsewhere.

GE has been executing bizarre offerings with 70's era Big Blue ideology since the late 90's.

I know guys that worked on their public cloud offering. Oracle seems more plausible than the Garbo GE was wasting $ on.

GE sold their lightbulb concern.

That would be like Intel selling off their networking division.
 

thesmokingman

Supreme [H]ardness
Joined
Nov 22, 2008
Messages
6,229
posted this in the Intel section, seems relevant here.

AMD balance sheet is actually superior to Intel.

Intel
total cash: 12.02B
total debt: 29.48B

AMD
total cash 1.21B
total debt 1.11B

And yes, intel is behind, and falling further away rapidly. the 5GHz / 14nm wall is hitting them hard. Revenue is falling, they are propping up the share price with buybacks in the billions instead of using it for R&D. Intel spent spent $4.5 billion on share repurchases during the third quarter versus $3.2 billion on R&D. Do you know that the CEO of intel is a Finance guy? Not an Engineer like the past. This shows you Intel's true direction.

WOW, their debt is getting out of hand.
 
  • Like
Reactions: N4CR
like this

Dan_D

Extremely [H]
Joined
Feb 9, 2002
Messages
57,033
You're right, I should have said game over for the next couple of years. Intel is a sleeping giant who for sure has been awakened at this point. No more screwing around with 10nm and milking the market with miniscule IPC upgrades. I don't see them striking back a vengeance for a couple of years, though.

I don't think we'll see anything that can challenge AMD the way it used to, much less beat them for 12-18 months and that's probably being generous.
 

kamikazi

Gawd
Joined
Jan 19, 2006
Messages
659
I don't think we'll see anything that can challenge AMD the way it used to, much less beat them for 12-18 months and that's probably being generous.

That means a lot coming from you. Glad I bought some AMD stock a few years ago. Just wish I bought more.
 

FlawleZ

[H]ard|Gawd
Joined
Oct 20, 2010
Messages
1,105
I don't think we'll see anything that can challenge AMD the way it used to, much less beat them for 12-18 months and that's probably being generous.
Maybe. But worse things have been said in even darker times when the shoe was on the other foot.
 

Rvenger

2[H]4U
Joined
Sep 12, 2012
Messages
2,131
Are you kidding me?

Intel can just drop 30 billion on new research and SMOKE amd.

View attachment 211805

View attachment 211806


Money doesn't buy Magic. You have to play your cards right. AMD did this time. Intel fucked up. Intel has a lot more portfolio of products and their R&D does not revolve around everything CPUs.

30 billion can buy a shit ton of engineers, but it only takes a few really good ones to make the magic. Money doesn't guarantee a better product, good brains and time does.
 

IdiotInCharge

NVIDIA SHILL
Joined
Jun 13, 2003
Messages
14,712
which AMD doesn't have to deal with.

AMD has a bigger long-term question: will TSMC continue to be able to deliver?

What's the plan for when they inevitably do not?

What's the plan when TSMC decides that someone else is more deserving of their fab capacity?

This is why Intel stayed in the fab business; it's biting them in the ass today, but it has delivered long-term and will likely continue to deliver long-term.

At the very least, TSMCs current process family has been fairly successful, and thankfully while Intel is experiencing their rare stumble, AMD has something actually worth producing!
 

bigdogchris

Fully [H]
Joined
Feb 19, 2008
Messages
18,206
If true, this has to be game over for Intel, especially on the server side. I know that Intel has the ecosystem and established IT infrastructure in place, but those kids of gains are too much to ignore.
I'm absolutely looking at EPYC during the next refresh cycle.
 

kac77

2[H]4U
Joined
Dec 13, 2008
Messages
2,583
AMD has a bigger long-term question: will TSMC continue to be able to deliver?

What's the plan for when they inevitably do not?
Move to another fab.

What's the plan when TSMC decides that someone else is more deserving of their fab capacity?
Move to another fab.

This is why Intel stayed in the fab business; it's biting them in the ass today, but it has delivered long-term and will likely continue to deliver long-term.

It hasn't delivered long term. Everything fab-wise is off. They are capacity strained when with AMD coming along they should have some breathing room. But they don't. It really looks like they offered to fab for others without making sure they have enough for their own personal designs.

Now is there a possibility in the far future for the capacity problems to pay off? Sure. But they haven't proved it yet. It looks like most of the engineers were dedicated to opening up more fabs for third party designs then their own.
 

IdiotInCharge

NVIDIA SHILL
Joined
Jun 13, 2003
Messages
14,712
Move to another fab.

Which means redesigning ICs. Yes, it's the right answer, but it's not an easy thing to do. Only Samsung appears to be in the running for advanced large scale production outside of TSMC and Intel, and we don't really know how much Samsung is really in the running.

That's why the question will hang over AMDs head. Maybe there'll be someone to produce their designs, and maybe they'll have to improvise like Intel is with 14nm. History being a guide points more to the latter than the former, and that's without really considering just how much harder future die shrinks will be.

It hasn't delivered long term.

99% of the x86 market isn't 'delivering'?

They are capacity strained

Sure; they planned to have already been on 10nm for years now, and to already have 7nm shipping. Instead of moving forward and upgrading fabs, they've had to roll back some to 14nm and backport their newer designs to 14nm as well. That really throws a wrench in capacity!

when with AMD coming along they should have some breathing room.

No one knew if AMD would ever deliver again -- and no one knew if TSMC or anyone else would crack < 10nm either. Both happening is a great stroke of luck for AMD, and that happening while Intel stumbled at 10nm? Crazy.

It really looks like they offered to fab for others without making sure they have enough for their own personal designs.

Now is there a possibility in the far future for the capacity problems to pay off? Sure. But they haven't proved it yet. It looks like most of the engineers were dedicated to opening up more fabs for third party designs then their own.

I don't really get the 'third-party' angle. Yeah, they flirted with it, back when they were lying to themselves about how their own fab progression was going, but do you really expect them to prioritize third-party production over producing their products?

If they had their 10nm fabs running at full tilt right now, that they'd be producing something other than as many Intel CPUs as they possibly could?
 

UnknownSouljer

Supreme [H]ardness
Joined
Sep 24, 2001
Messages
6,740
Interview with CTO Mark Papermaster is up: https://www.anandtech.com/show/1526...-mark-papermaster-theres-more-room-at-the-top

I would say that for the most part it's a fluff piece. I can tell that basically all the responses are canned and no detail is really laid out nor does the interviewer really press for more info on any of those subjects. But there are a few notable quotes in there. Such as AMD's goal to beat the "industry standard" of 7% gains per year. And of course the immediate expectation of IPC gains for Zen 3 being much higher (or similar to gains made from Zen 1 to 2). And TSMC capacity is also briefly talked(glossed) over. There are a few other tid bits. The whole article can really be skimmed in a few minutes.
 

tangoseal

[H]F Junkie
Joined
Dec 18, 2010
Messages
8,942
Im in the pipe for a new shiny and smellin' like new chip, 4700x when they come out. I'll give it to my 6 year old to play his games on with me over the lan.
 

Gideon

2[H]4U
Joined
Apr 13, 2006
Messages
2,737
AMD has a bigger long-term question: will TSMC continue to be able to deliver?

What's the plan for when they inevitably do not?

What's the plan when TSMC decides that someone else is more deserving of their fab capacity?

This is why Intel stayed in the fab business; it's biting them in the ass today, but it has delivered long-term and will likely continue to deliver long-term.

At the very least, TSMCs current process family has been fairly successful, and thankfully while Intel is experiencing their rare stumble, AMD has something actually worth producing!

It always amazes me when you comment that AMD has tied their hopes to TSMC and how that is a looming question of a possible issue, then in another thread you have no Issue with Nvidia doing the same thing and has for years.
 

IdiotInCharge

NVIDIA SHILL
Joined
Jun 13, 2003
Messages
14,712
It always amazes me when you comment that AMD has tied their hopes to TSMC and how that is a looming question of a possible issue, then in another thread you have no Issue with Nvidia doing the same thing and has for years.

Nice to get personal!

And you're going to have to work pretty hard to even begin to show that I don't have a problem with Nvidia tying their hopes to TSMC. It's precisely because Nvidia has been held back by TSMC before that AMD using TSMC to compete with Intel is a reason for worry.
 

Gideon

2[H]4U
Joined
Apr 13, 2006
Messages
2,737
Here's the thing: TSMCs 7nm process works. Nvidia has been the leading GPU company basically since AMD bought ATi and spent years just trying to produce a GPU with a full featureset.

Nvidia has consistently moved to new nodes without drama, unlike AMD.

The only reason to introduce doubt is personal.

You only ever seem to doubt AMD, just like in this quote, you were even good with the idea Nvidia might use Samsung as well until Nvidia denied that rumor. It's a observation, it's only a attack in your mind.
 

tangoseal

[H]F Junkie
Joined
Dec 18, 2010
Messages
8,942
must be why 10nm is so successful.
They probably spent 500 million on that. If they dropped 10 billion into R&D 10nm would be just fine.

Intel will NOT allow AMD to lead for long. They are chipzilla forms reason. They have bank and investors galore.
 

Master_shake_

[H]F Junkie
Joined
Apr 9, 2012
Messages
13,770
They probably spent 500 million on that. If they dropped 10 billion into R&D 10nm would be just fine.

Intel will NOT allow AMD to lead for long. They are chipzilla forms reason. They have bank and investors galore.

their lack of products going forward kinda rule them out of a lead wouldn't you say?

well i guess they could release the same thing over and over agai.....oh wait.
 

Rizen

[H]F Junkie
Joined
Jul 16, 2000
Messages
9,330
They probably spent 500 million on that. If they dropped 10 billion into R&D 10nm would be just fine.

Intel will NOT allow AMD to lead for long. They are chipzilla forms reason. They have bank and investors galore.
That's not really how it works at this level though. This is cutting edge engineering on the nano-scale, they are running up against the laws of physics. You can't just throw money at the problem to make it work. They were too ambitious with their density goals for 10nm and more money wouldn't have made the process work. It's why everyone else moved to EUV processes.
 

Rizen

[H]F Junkie
Joined
Jul 16, 2000
Messages
9,330
AMD has a bigger long-term question: will TSMC continue to be able to deliver?

What's the plan for when they inevitably do not?

What's the plan when TSMC decides that someone else is more deserving of their fab capacity?

This is why Intel stayed in the fab business; it's biting them in the ass today, but it has delivered long-term and will likely continue to deliver long-term.

At the very least, TSMCs current process family has been fairly successful, and thankfully while Intel is experiencing their rare stumble, AMD has something actually worth producing!
Yeah like I said there are pros and cons - but fabs are extremely expensive to build, operate, and innovate with, and (IMO) AMD is better as a focused company working on a narrow scope of products with the cash they have available. That's basically what Lisa Su has done, and it's worked very well turning the company around.
 

dgz

Supreme [H]ardness
Joined
Feb 15, 2010
Messages
5,838
That's not really how it works at this level though. This is cutting edge engineering on the nano-scale, they are running up against the laws of physics. You can't just throw money at the problem to make it work. They were too ambitious with their density goals for 10nm and more money wouldn't have made the process work. It's why everyone else moved to EUV processes.

Are you trying to tell me hiring 5 more women won't make a baby develop faster in the womb? Guess I'll need to be more creative if I expect to convince her we need them
 

Jandor

Gawd
Joined
Dec 30, 2018
Messages
530
Are you kidding me?

Intel can just drop 30 billion on new research and SMOKE amd.

View attachment 211805

View attachment 211806
Let's analyze this with a comparison.

Boeing : Market Cap 187B$.
SpaceX : Market Cap... Private. Funding 3 B$, but not for what I'll be explaining (but for some even more weird stuff).
Frankly speaking SpaceX is so little compared to Boeing that it's crappy to compare. It's just ridiculous. They have no history vs Boeing. You wouldn't bet a dime on it if you're financial and look at the financial records. If you were a financial guy you would just say that SpaceX is a hoax.
NASA launched a competition between the two for a crowded ship and both answered. Boeing is a half failure, not yet ready, not sure will ever be, even missed a launch, will cost double (90M $) per launch. SpaceX is ready is on third test to be validated, costs 55M$ per launch.
 

tangoseal

[H]F Junkie
Joined
Dec 18, 2010
Messages
8,942
Let's analyze this with a comparison.

Boeing : Market Cap 187B$.
SpaceX : Market Cap... Private. Funding 3 B$, but not for what I'll be explaining (but for some even more weird stuff).
Frankly speaking SpaceX is so little compared to Boeing that it's crappy to compare. It's just ridiculous. They have no history vs Boeing. You wouldn't bet a dime on it if you're financial and look at the financial records. If you were a financial guy you would just say that SpaceX is a hoax.
NASA launched a competition between the two for a crowded ship and both answered. Boeing is a half failure, not yet ready, not sure will ever be, even missed a launch, will cost double (90M $) per launch. SpaceX is ready is on third test to be validated, costs 55M$ per launch.

SpaceX doesnt sell weapons to every country on Earth like Boeing. But quite possibly they will in the future.
 
Top