Article "AMD Has Lost"

[DigDog]

somebody please help me link this because my browser isn't being helpful:

Why AMD’s best days are behind it

AMD had a shot to catch Intel and blew it.

www.marketwatch.com

Accoridng to Daniel Newman, principal analyst at Futurum Reseach, AMD "has lost".

I'll directy you straight away to the .. humorous video of Gamers Nexus who are extremely perplexed at the amount of lies and bad facts in said article, to the point where nobody with a bit of knowledge of the current market could make those mistakes. Video starts at 2:40

I really, REALLY hope this isn't a marketing tactic sponsored by Intel because that would .. be disgusting, but also suggest that things are much worse than expected, in Intel.

I hope this is satisfactory.

Iron Woode
Super Moderator

 

[CHADBOGA]

What I said is there is no more AMD bias on the forum today than there was Intel bias a few years ago. We always had fans from both camps, but the tech status quo dictates who gets to be more vocal. The forum tends to resonate with arguments for the brand with better products. Look at the GPU forum: would you say there is more AMD bias there than Nvidia? At some point it all got so heated up they had to split the GPU section in half.

What's the common factor in the CPU and GPU forum wars?

 

[bigboxes]

You are absolutely spot on about how ridiculous the people who should have known better were, for suggesting that competition leads to less effective outcomes because of some utopian reasoning that would never apply, but you are mistaken to suggest there is no AMD bias in this forum.

Obviously there aren't many cases where I feel it currently makes sense to buy an Intel CPU(except for mobile), but that doesn't mean a good deal of the AMD advocates, aren't biased towards AMD.

AMD is finally getting their turn. I'm still running Intel and I think this is good development. This is great for competition and great for getting Intel off its duff. Intel owners have been dogging AMD owners for years on this forum. You know you have been one of those posters for Intel. So, now AMD has the halo chip and Intel is way too expensive. I think it's great!

 

[CHADBOGA]

You know you have been one of those posters for Intel.

Nonsense.

I have only ever tried to tackle the insane and have frequently stated that I am incapable of punishing myself by buying lesser value processors for preposterous "moral" reasons.

Anyway, Intel is off my radar for now as a CPU choice and I would be surprised if they come back before they get to 7nm and even then, possibly not.

 

[coercitiv]

What's the common factor in the CPU and GPU forum wars?

Human ego.

Anyway, let's agree to disagree on the AMD bias intensity in the CPU forums. Time will tell better than anyone of us can.

 

[bigboxes]

Nonsense.

I have only ever tried to tackle the insane and have frequently stated that I am incapable of punishing myself by buying lesser value processors for preposterous "moral" reasons.

Anyway, Intel is off my radar for now as a CPU choice and I would be surprised if they come back before they get to 7nm and even then, possibly not.

Nonsense. You HAVE been one of those posters (can't use the other name) and you belittled anyone that chose AMD. So much so that I starting glossing over your posts because you were always like this. You didn't want to discuss anything. Just shout them off the board for choosing AMD. You were like that nVidia guy Rollo. I'm not saying that you were a paid shill, but it sure did seem like it.

 

[Kocicak]

Quoting from the article:

AMD failed to capitalize on a perfect storm of opportunity when its chip-making rival was wounded.

I am not sure about the past tense here, Intel still IS seriously wounded and incapacitated with no hope for quick recovery. They are forced to deliver products with break neck frequency and bizarre power draw, in order to stay remotely competitive. And for some competing products they offer no alternative at all.

Other than that, I noticed too, that AMD was not very aggresive with advertising and pushing their products. I believe that the main reason for this is, that they would not have been able to deliver more than what they did because of the "tight supply" (quoted from AMD) of TSMC production.

I believe the "storm" from AMD comes this year with much improved supply from TSMC.

 

[DisEnchantment]

Let's be honest here, Michael Larabel has shown several times that there's no perceivable performance loss when you're using the mitigations=off parameter and if you only run your own code, or the code you totally trust, then there's no reason not to use it.

I have to agree with you that the companies which provide shared virtual environments (Amazon AWS, Linode, etc.) have been hit hard by these mitigations because they cannot disable them.

I have all Intel at work.
I have a thin Client for presentations and to run other corporate SW using Windows and a Workstation Laptop which runs Linux.
On my own WS I can change anything. I recompile the kernel with all the things I need.
I agree I can take out most of the performance loss from the mitigations.

But I dont compile stuff, run the CI, run simulations, deploy and run our SW tests etc on my WS it will take me several hours to compile and several days to run our tests.
We connect to our VMs deployed on Intel based clusters to do all that stuffs.
There is a big hit here.
On top of that, we are using the services of a major cloud provider from around Seattle area for the backend services and another Massachusets based Company provides us the CDN.
Our workflow is like
Develop code locally on network shares -> Build on VMs hosted on our build/test clusters --> run tests/etc --> push changes to Cloud --> Cloud Infra spawns VM to run build and CI in docker containers --> Deploy to Worker services for Backend and to API gateways --> Deploy to CDN --> Push to IOT --> End SW runs on IOT and connect to our cloud API frontend and CDNs.
Our IOT is not exactly an IOT but a complete Vehicle. We have components running on the backend and also running on edge devices.

Many steps of our workflow is impacted and it really annoying especially when you have to worry about pushing bug fixes to the edge devices and any additional bottleneck (caused by loss of performace or otherwise) in the workflow on top of what we already have delays deployment of changes.

 

[ksec]

I speak as someone who wants AMD to win and loathe Intel, but Let's be fair,

That analysts wasn't exactly wrong. There are many issues at play here, and he is not the only one within the industry who were disappointed with AMD. ( Even though he may be biased )

AMD currently trade at 2019 P/E of 180, and Intel trade at 2019 P/E of 14.2

Just let that sink in for a min.

Even if you accounted for the 200M they paid in debt that could have counted as profits, the P/E is still ~120. Which means even if AMD manage to grow 50% every year for ALL segment in the next 4 years ( 5x ) , they will still be at a P/E of 24.

Now given the sort of dismal market share they have right now in Laptop and Server market. Growing 5x isn't exactly impossible. The problem is this completely ignores the GPU, Embedded and Semi-Custom, do you see those line getting 5x the size few years down the line?

So AMD needs to grow 5x+ ( 6-7x ) of its CPU revenue in the next 4 - 5 years to even make its current price feasible. Which isn't impossible considering that means AMD would needs around 35% Server Market shares.

So what is the problem?

1. AMD themselves only project revenue of 25-30% growth for FY2020.
2. Even if this is a conservative projection, it is the growth happening during Intel's perfect storm.
3. You are basically betting Intel dont have much answer to AMD's offering in the next 4- 5 years timeline.

While I still dont think Intel has pick up their slack yet ( We will know for sure when they ship Tiger Lake and IceLake Server ), you have to give credit to Intel for their defence, from Marketing, Sales to Software / Open Source Contribution. And while people are blaming TSMC for tight supply, when it is now very clear it was AMD's conservative estimate that failed them. Semi-Conductor Fabs ( TSMC ) doesn't work on Just in Time production, everything is literally a quarter or two quarter in advance. And a failure to capture more market shares when Intel is running at 100% capacity is also the failure ( comparatively speaking ) of AMD's sales, marketing, and forecasting.

Of course, the no one gets fired for buying Intel is still a very strong perception, and most consumer stills prefer Intel.

The only thing that could turn the whole thing around, ( in my opinion ) is if AMD managed to win Apple over for its Mac. No one gets fired for using something Apple is using.

So while I do think AMD still has a bright future, it is a little hard to be overly optimistic given the current valuation.

Edit: I would also like to remind people to be careful with those so called market shares figures and statistic. Example even if Newegg report 100% of their CPU sold are AMD, it would still represent less than 1% of the total PC market shipment. ~75% of all PC shipment are Notebooks, and majority of the 25% Desktop are Business uses. Which sort of give you an idea why winning much of the enthusiast market hasn't resulted in significant uptake in overall market shares.

 

[Atari2600]

That guy really puts the anal in analyst all right.

The clown thinks a new CEO can snap their fingers and have process & architecture appear as if by magic after a couple of months.

It'll take Intel several years to turn around design approach from a 2018 datum.... and that only really can occur when process is sorted and when talent is not constrained by (mis)management. There also is no Yonah waiting in the wings for Intel to roll into desktop/server.

The fundamentals of AMD in 2020 compared to AMD in 2005 is that now AMD is riding the wave of the process giant (TSMC & all its mobile clients), rather than being alone vs. chipzilla.

Of course, anal-yst doesn't understand that, so insists that the earth is flat.

Intel will probably make a come back, but it is not this side of 2023.


AMD need a few things to build penetration significantly:
- An APU (ironic) that uses their cutting edge architectures; you'd have thought the MCM approach would lend itself to this, but obviously not so far.
- Time. Time for continued execution to demonstrate to business that EPYC is a platform worth investing in.
- Software. The Intel compilers and support stack are miles better than AMD. Hopefully continued revenue growth will allow more resource to be devoted to this.

 

[amrnuke]

I speak as someone who wants AMD to win and loathe Intel, but Let's be fair,

That analysts wasn't exactly wrong. There are many issues at play here, and he is not the only one within the industry who were disappointed with AMD. ( Even though he may be biased )

AMD currently trade at 2019 P/E of 180, and Intel trade at 2019 P/E of 14.2

Just let that sink in for a min.

Even if you accounted for the 200M they paid in debt that could have counted as profits, the P/E is still ~120. Which means even if AMD manage to grow 50% every year for ALL segment in the next 4 years ( 5x ) , they will still be at a P/E of 24.

Now given the sort of dismal market share they have right now in Laptop and Server market. Growing 5x isn't exactly impossible. The problem is this completely ignores the GPU, Embedded and Semi-Custom, do you see those line getting 5x the size few years down the line?

So AMD needs to grow 5x+ ( 6-7x ) of its CPU revenue in the next 4 - 5 years to even make its current price feasible. Which isn't impossible considering that means AMD would needs around 35% Server Market shares.

So what is the problem?

1. AMD themselves only project revenue of 25-30% growth for FY2020.
2. Even if this is a conservative projection, it is the growth happening during Intel's perfect storm.
3. You are basically betting Intel dont have much answer to AMD's offering in the next 4- 5 years timeline.

While I still dont think Intel has pick up their slack yet ( We will know for sure when they ship Tiger Lake and IceLake Server ), you have to give credit to Intel for their defence, from Marketing, Sales to Software / Open Source Contribution. And while people are blaming TSMC for tight supply, when it is now very clear it was AMD's conservative estimate that failed them. Semi-Conductor Fabs ( TSMC ) doesn't work on Just in Time production, everything is literally a quarter or two quarter in advance. And a failure to capture more market shares when Intel is running at 100% capacity is also the failure ( comparatively speaking ) of AMD's sales, marketing, and forecasting.

Of course, the no one gets fired for buying Intel is still a very strong perception, and most consumer stills prefer Intel.

The only thing that could turn the whole thing around, ( in my opinion ) is if AMD managed to win Apple over for its Mac. No one gets fired for using something Apple is using.

So while I do think AMD still has a bright future, it is a little hard to be overly optimistic given the current valuation.

Edit: I would also like to remind people to be careful with those so called market shares figures and statistic. Example even if Newegg report 100% of their CPU sold are AMD, it would still represent less than 1% of the total PC market shipment. ~75% of all PC shipment are Notebooks, and majority of the 25% Desktop are Business uses. Which sort of give you an idea why winning much of the enthusiast market hasn't resulted in significant uptake in overall market shares.

There's nothing majorly wrong with AMD's P/E in my opinion. Of course it's high - there is a lot in the future that investors are excited about. (Also, they paid over $500m in debt - "we reduced gross debt by $524 million". That put them over $1 billion in debt paydown for the year.) Ryzen 4000 mobile, Zen3 and the next wave of server and mainstream chips. Continued work on RDNA that has brought them near equivalency with Nvidia with respect to efficiency and perf/W. The reality is that with AMD paying down debt, solid cash flow, increase in R&D/marketing expenditure, and 14+% growth in CPU/GPU market along with a huge pipeline of anticipated advances to continue to boost earnings, things look great. This is an agile, less burdened company.

For Intel, the P/E reflects that they are a behemoth struggling to navigate a new landscape, with little in the pipeline that is exciting. Their P/E ratio is 2/3 that of the market as a whole, telling you just how disenchanted investors are with their projected growth and future earnings.

Intel has a larger market share. When you look at narrow markets like this, with CPU duopoly, Intel has 85% of the CPU market and AMD 15%. The growth of the market in general will continue. Investors are simply putting almost all of their bets on AMD to continue to take market share and generate earnings. There is nothing wrong with that. I'd be more worried if Intel's P/E were 40 and AMD's were 180. But with Intel's being below market, the investors are expecting shrinkage, with the majority of Intel's value lying in dividends rather than earnings growth. AMD is far different.

 

[rbk123]

If that's true, he may be violating some SEC regulations? Maybe? It would certainly look bad if he came out swinging as a financial analyst on a site like marketwatch trying to manipulate people into defending his positions.

Happens all the time in the Financial Services sector. Impossible to prosecute.

ksec hit the nail on the head - their P/E ratio is stratospheric so it's now a speculators stock and not fundamentals based. It can't continue at that level indefinitely, but that doesn't mean it won't stay that way for a good bit of time. That certainly doesn't mean AMD has "lost", though. Big companies don't change vendors easily - that is the hardest nut for AMD to crack.

 

[rbk123]

Nonsense. You HAVE been one of those posters (can't use the other name) and you belittled anyone that chose AMD.

Oh yeah. Chad was right up there with The Elf and Zucker2k in "Intel is the greatest and can do no wrong" posts. The kool-aid does seem to have worn off for him of late, though; but the AMD versions of The Elf and co are equally painful. So give Chad credit for trying to be a better person, and help him get through his recovery.

 

[CHADBOGA]

Oh yeah. Chad was right up there with The Elf and Zucker2k in "Intel is the greatest and can do no wrong" posts. The kool-aid does seem to have worn off for him of late, though; but the AMD versions of The Elf and co are equally painful. So give Chad credit for trying to be a better person, and help him get through his recovery.

You won't be able to find posts that back up your incorrect assertions.

I just tried to save people from Bulldozer.

 

[Ottonomous]

The real question is if anyone has stock or is financially compensated to some measure by the three companies.

 

[DigDog]

i just want to chime in that, i find disturbing when people discuss servers and games in the same breath.

If you are a gamer, you will be - and should be - unconcerned with Spectre / Meltdown exploits beause you are not likely to ever be attacked. IIRC these exploits cause a portion of the memory to be dumped out, which isn't exactly something a consumer-targeting hacker has much interest in. They want your CC details and Amazon login so keyloggers, fake game .exe, social exploits and ringing you up to offer a free virus sweep should offer far more returns.
of course a server that stores credit card info for a bank, that's another matterm but that is also a platform which doesnt concern itself with how well you can play Minecraft on it.

second; Intel architecture isn't bad because it suffers from exploits. AMD's architecture isn't great because it doesnt. AMD's architecture could RIGHT NOW be suffering from ludicrous exploits that neither AMD nor the hacker knows exists yet. The same reasoning applies to Intel, nobody knew there was an exploit until the exploit was found. It's a matter of discovery, not of invention, and there is no divining if your archi design is or isnt exploit-free.

 

[amrnuke]

The same reasoning applies to Intel, nobody knew there was an exploit until the exploit was found. It's a matter of discovery, not of invention, and there is no divining if your archi design is or isnt exploit-free.

That's just inherently misleading. Intel knew about the problems with TLB for over a decade before the exploit was formally released. They continued to bake the same exploits into their chips. They have also actively tried to withhold release of exploit reports with other vulnerabilities.

Intel almost certainly knew about their TLB issues and the Meltdown-like exploits in 1995. In 2012 Apple uncovered the issues, and had to bake in a kernel fix to secure the systems because Intel still hadn't fixed things. Same with Linux in 2014 with KASLR. (Then, once made public, Intel told the Chinese government about the issues before telling the US government. I'm not here to play politics though.) In any case, this is the kind of blind eye Intel has turned toward known architectural issues with respect to security. To say that nobody knew there was an exploit until it was found is horribly misleading. Intel knew about the exploits and continued to make new architectures with the same issues, for years, while trying to suppress release of exploit reports.

 

[DigDog]

That's just inherently misleading. Intel knew about the problems with TLB for over a decade before the exploit was formally released. They continued to bake the same exploits into their chips. They have also actively tried to withhold release of exploit reports with other vulnerabilities.

Intel almost certainly knew about their TLB issues and the Meltdown-like exploits in 1995. In 2012 Apple uncovered the issues, and had to bake in a kernel fix to secure the systems because Intel still hadn't fixed things. Same with Linux in 2014 with KASLR. (Then, once made public, Intel told the Chinese government about the issues before telling the US government. I'm not here to play politics though.) In any case, this is the kind of blind eye Intel has turned toward known architectural issues with respect to security. To say that nobody knew there was an exploit until it was found is horribly misleading. Intel knew about the exploits and continued to make new architectures with the same issues, for years, while trying to suppress release of exploit reports.

that .. .is very different than what i read last year, when Spectre was all over the press.

 

[fkoehler]

I just wanted to say kudo's for this.

I well remember 4-5 years ago when Intel was simply owning AMD, and AMD was in the midst of Bulldozer. There were a number of high profile posters who couldn't help but pi** on AMD and crow about how Intel would never lose, AMD was done, etc.
I'm sure most of us remember that, and some here were doing exactly that.

Funny, now their tunes have changed, and many of the worst are gone, or just here under a different handle.

I'm preferential to AMD, unless there is a really good reason to look at another.
I also was following the market as a non-pro, and in the past year and a half made enough on trades to buy a nice house outside of the East/West coast metro areas.
If I had simply held it, I'd have been halfway to millionaire status.

However, I think holding AMD long-term is simply to risky.
Right now at 48-52/share, you need a lot of capital to make any money, and AMD's P/E is simply outrageous.
Secondly, everything AMD has built recently has a single point of failure.
Any real earthquake in SK has the potential to totally screw TSMC, and many other of its clients.
I'm sure they've got lots of earthquake proofing, etc, however they don't have geographical redundancy like Intel. Toshiba just lost a chunk of their quarterly production due to a 1 min electrical outage.

OK, so they're playing with Samsung also, which is located where?

I'm very happy to see AMD kicking Goliath square in the teeth. They deserve it, along with nVidia.
Honestly, anyone feeling butthurt about Intel getting some karma probably needs a hug or something.

2-3 years from now, Intel may be back on top, and every discussion regarding AMD will have the same sort of replies.

Its called growing a pair or thicker skin.

Who knows, might be discussing Apple as the new Intel/AMD in the future.

BTW, is butthurt also on the list of disallowed words? Where is that list anyways?

Keep in mind this is an open forum. You might be talking to a 70 year old engineer with a background in IT, or a 17 year old enthusiastic student. Some will be fans, some will use hyperbole, some will confuse the silicon tech field with celebrity sport matches.

Keep calm and read the tech threads you're interested in. We don't own any of these brands our allegiance.

 

[maddie]

That's just inherently misleading. Intel knew about the problems with TLB for over a decade before the exploit was formally released. They continued to bake the same exploits into their chips. They have also actively tried to withhold release of exploit reports with other vulnerabilities.

Intel almost certainly knew about their TLB issues and the Meltdown-like exploits in 1995. In 2012 Apple uncovered the issues, and had to bake in a kernel fix to secure the systems because Intel still hadn't fixed things. Same with Linux in 2014 with KASLR. (Then, once made public, Intel told the Chinese government about the issues before telling the US government. I'm not here to play politics though.) In any case, this is the kind of blind eye Intel has turned toward known architectural issues with respect to security. To say that nobody knew there was an exploit until it was found is horribly misleading. Intel knew about the exploits and continued to make new architectures with the same issues, for years, while trying to suppress release of exploit reports.

To some, what seems like an exploit might just be a feature. What you wrote here should make readers challenge their assumptions. I for one, hold the view that Intel is not that negligent and so it must be intentional. Do we really think this only occurs in spy fiction?

Remarkably, we have this from yesterday:

https://www.washingtonpost.com/graphics/2020/world/national-security/cia-crypto-encryption-machines-espionage/

‘The intelligence coup of the century’

For more than half a century, governments all over the world trusted a single company to keep the communications of their spies, soldiers and diplomats secret.

The company, Crypto AG, got its first break with a contract to build code-making machines for U.S. troops during World War II. Flush with cash, it became a dominant maker of encryption devices for decades, navigating waves of technology from mechanical gears to electronic circuits and, finally, silicon chips and software.

The Swiss firm made millions of dollars selling equipment to more than 120 countries well into the 21st century. Its clients included Iran, military juntas in Latin America, nuclear rivals India and Pakistan, and even the Vatican.

But what none of its customers ever knew was that Crypto AG was secretly owned by the CIA in a highly classified partnership with West German intelligence. These spy agencies rigged the company’s devices so they could easily break the codes that countries used to send encrypted messages.

 

[amrnuke]

that .. .is very different than what i read last year, when Spectre was all over the press.

Spectre is a different beast, but spec-ex cache timing attacks on Intel x86 CPUs were demonstrated in 2005, and the uarch continued, from 2010 through to 2018, to include kernel memory exploits as well as no hardware hardening to protect against exploits. AMD did have potential exploit with Spectre but this was able to be rectified with a software patch, while Intel required software and firmware/uop patching.

 

[zinfamous]

Most financial analysts in the tech industry honestly don't understand squat about the industry they claim to be experts in. So you get a lot of either skepticism for a non-blue blood player, or you get an over eagerness to jump on hot up and comers with big promises and little technical foundation to support those promises.

lol--there was an "article" the other day, on this clearly fake rumor that "Apple might be choosing AMD over Intel for their CPUs!" where they mentioned: "Apple might be using AMD processors in their Apple iPhone, which currently uses Intel processors."

I mean...that's all you need to read to know that the majority of these people are just flat out lying to meet their click quotas.

 

[DigDog]

in reply to @fkoehler
there is difference in someone saying "Intel will never lose because they are amazebawls" and saying "i predict that the larger, more experienced company with a deathgrip on the market will be the winner"

the origin of the "fanboy" word is when someone is .. literally, a fan. Someone who loves a team JUST BECAUSE, and they will support their idol without any special reason aside from "i like this thing".

being a fan of AMD when their chips were destroying Intel on IPC AND price isnt fanboyism. Being a fan of AMD during bulldozer instead is fanboyism.


Also, as a very broad observation, when you're the underdog you will reach for tactics which actually work, like: listening to your customers, delivering targeted products, and undercutting the market leader. When instead you are in the lead, you'll use dominance tactics, such as other-brand tie-in, lying in the press, exclusives, promotions, you basiaclly burn a portion of your earnings to maintain the leadership.

I have said this before, and i am confident it is accurate, Intel's biggest problem is that they have been too successful for too long - it's an old, big big company, with many directors, assistant directors, secretaries, department heads, regulations, protocols, and i want to bet, internal legacy systems; their developers dont rub shoulders, dont eat ramen in the same canteen, they are not hungry for success, but have too many sub-contractors whose job rotates around Intel maintaining what they have rather than risking a new direction.










im gonna guess the decision to release the 9900 went like this:

"Boss, AMD have got this new chip, it's got more cores, uses less power, has this massive boost, but also stays cool under normal loads, AND costs less than ours; what should we do?"
Boss: what options do we have?
Engineer: well, we should obviously re-design our architecture. And re-negotiate our contracts wth our technology partners. And increase the R&D budget. and.
Boss: no.
Engineer: what ?
Boss: no. take what we have, and push it to .. 5Ghz. Overclock it.
Engineer: but, the efficiency..
Boss: And make it eight cores.
Engineer: .. but, that would take the power envelope to ..
Boss: and .. NEW BOX. Get marketing on it!




right now AMD "make" chips, while Intel "sells" chips. It's obvious that a forum of technology enthusiasts will prefer AMD over Intel. Intel obviously make the decisions which make more sense to them as a market leader; you wanna sell a "gaming" chip, dont call it the DF8445 NE, or the PUEA442, call it the TX99 Terminator.
Are they selling a pre-overclocked chip that requires realistically a $100 AIO to run as described? yes they do. Are people buying it? You bet.

it's unrealistic to expect Intel to be on the bleeding edge of technology because they do not have incentives to do it. The opposite is true for AMD.

However, it worries me, when Intel - or anyone else - decides to use unfair market tactics, because that impacts me directly. I am totally ok with a company saying (or not saying, but me being aware of) "consumers are stupid, so this is the product that is most profitable for me to produce", but if they then attack the company that targets products at me, then i am directly impacted by Intel's bloatedness.

 

[potato masher]

Security mitigations that make the users on older slower hardware want to $$$ upgrade.. that motive there tho.

 

[jpiniero]

Sony Is Struggling With PlayStation 5 Price Due to Costly Parts

Scarce components have pushed the manufacturing costs for Sony Corp.’s next PlayStation to around $450 per unit, forcing a difficult price-setting decision in its battle with Microsoft Corp., according to people with knowledge of the matter.

www.bloomberg.com

Sort of tangently related, Sony is having problems deciding on the PS5's price. Obviously this is bad news for AMD if they end up selling the PS5 at $499.

 

[BigDaveX]

Sony Is Struggling With PlayStation 5 Price Due to Costly Parts

Scarce components have pushed the manufacturing costs for Sony Corp.’s next PlayStation to around $450 per unit, forcing a difficult price-setting decision in its battle with Microsoft Corp., according to people with knowledge of the matter.

www.bloomberg.com

Sort of tangently related, Sony is having problems deciding on the PS5's price. Obviously this is bad news for AMD if they end up selling the PS5 at $499.

Good thing AMD is providing the SoCs for all incarnations of the PS4, PS5, Xbox One, and Xbox Series X, then - even if the newer consoles have a slower start, their predecessors will have one last hurrah this holiday season, before the market transitions next year when the prices on the newer consoles will hopefully drop.

Well, that is unless everyone just starts buying the Switch instead - that would be bad news for AMD!