Intel CEO Doesn't See Arm-based Chips as Competition in the PC Sector

erek

erek

[H]F Junkie
Joined
Dec 19, 2005
Messages
11,046
Seems like a mistake in leadership, that Apple M chip was pretty impressive, right?

“Additionally, the CEO noted: "When thinking about other alternative architectures like Arm, we also say, wow, what a great opportunity for our foundry business." If the adoption of Arm-based CPUs for Windows PCs becomes more present, Intel plans to compete with its next-generation x86 offerings like Meteor Lake, Arrow Lake, Lunar Lake, and even Panther Lake in the future. As stated, the CEO expects the competition to manufacture its chips at Intel's foundries so that Intel can provide a platform for these companies to serve the PC ecosystem.”

https://www.techpowerup.com/315228/...m-based-chips-as-competition-in-the-pc-sector
 
Gelsinger is delusional. He's also gone on record saying AMD is in the rear-view mirror after Alder Lake launched.

This time is approaching faster than he thinks, the Snapdragon X Elite is looking pretty impressive and should see some decent adoption. As is the Apple M chip you mentioned erek . Don't forget Microsoft is now teaching developers how to code for Arm as x86 end of life approaches.

Edit: Tested: Qualcomm’s Snapdragon X Elite CPU looks like a legit Intel rival. This is in the mobile/laptop sector, but don't kid yourself it's going to trickle into desktop/server segments too. If Qualcomm is achieving performance like this at 80W, imagine how it could perform tuned for higher power draw and more robust cooling. Both Intel and AMD do indeed have to worry about this.
 
Last edited:
Gideon said:
No one will worry about it until a desktop chip is launched and sells well, server will likely become a custom thing if ARM is used. The death of X86 has been foretold far too many times.
Click to expand...
And what of the unforetold death of RISC-V huh? 🤔 🧐

Wonder if Intel has a post-x86 plan or roadmap?

Beyond the previous failed architectures of ia64 and the one other I can’t remember

Or is he truly embracing x86_64 til the very end?
 
If there's any company that wouldn't be able to acknowledge and process a trend in technology, it's Intel. Granted, they have too much invested in x86 at this point to want a challenge to that form of computing, but simply dismissing that one can come would be an error. I don't know if x86 will for sure be going away any time soon or in the future, but there's no question that ARM is becoming increasingly more relevant, and more and more companies are going in that direction. Intel has grown so accustomed to resting on their laurels while the rest of the world moves on from them that I wouldn't be surprised if that's the ultimate result here, as well.
 
Dopamin3 said:
Edit: Tested: Qualcomm’s Snapdragon X Elite CPU looks like a legit Intel rival. This is in the mobile/laptop sector, but don't kid yourself it's going to trickle into desktop/server segments too. If Qualcomm is achieving performance like this at 80W, imagine how it could perform tuned for higher power draw and more robust cooling. Both Intel and AMD do indeed have to worry about this.
Click to expand...

A lot of those benchmarks have "Qualcomm estimate" beside the CPU model lol. I'm kind of curious just how accurate of an estimate that is.

Either way I'll hold off on an opinion on ARM-based architectures until I see some benchmarks of actual real world use cases that I care about. All I remember about ARM is that I think it had a somewhat compressed instruction set, bit wise, or something, which made it ideal for the mobile space. However it's sort of taken a life of its own lately. I think AMD is already developing some ARM based chips anyway. Intel... Intel is either burying its head in the sand, or just treading water while trying to get the ship stable.

But I'm just a consumer. If Qualcomm shows that the next Crysis-like titles just play better on its architecture, and it somehow has backwards compatibility back to the old Thief games on Windows, then fine I'll just buy it for my next build. But there's gonna be a long time between "well we showed this works on some mobile stuff, which is where we specialize anyway" and "we made this 300 watt behemoth that just shreds AMD and Intel and is fully backwards compatible, and we have contracts lined up with motherboard manufacturers so you can build your own PC with it." Although knowing Qualcomm what I'm worried about is that the world of PC building is going to move towards integrated motherboard/CPU/RAM combos where everything is soldered on. The GPU slot will still be separate because they can't compete with Nvidia or AMD in that space (yet).
 
Dopamin3 said:
This time is approaching faster than he thinks, the Snapdragon X Elite is looking pretty impressive and should see some decent adoption. As is the Apple M chip you mentioned erek .
Click to expand...
But is it better than AMD or Intel? As it stands right now, AMD and Intel easily beat the M2 Max in performance and AMD nearly matches it in power efficiency. Intel will soon release their Meteor Lake chips, which are meant to be power efficient. This brings up the question as to why bother dealing with ARM at all? So you can either hunt for software that's compatible or rewrite every line of code so it's natively ARM? Sorry but Intel's right. If you buy an ARM based laptop, you are punishing yourself.
Don't forget Microsoft is now teaching developers how to code for Arm as x86 end of life approaches.
Click to expand...
That's to Microsoft's benefit, not towards yours. Microsoft has been trying to use ARM to lock users to their store since Windows RT.
Edit: Tested: Qualcomm’s Snapdragon X Elite CPU looks like a legit Intel rival. This is in the mobile/laptop sector, but don't kid yourself it's going to trickle into desktop/server segments too. If Qualcomm is achieving performance like this at 80W, imagine how it could perform tuned for higher power draw and more robust cooling. Both Intel and AMD do indeed have to worry about this.
Click to expand...
They tested against one Intel CPU using nothing but synthetic benchmarks. Apparently the only applications ARM can run are synthetic benchmarks. Also, the GPU performance of the Snapdragon X Elite was admitted to be poor.
erek said:

That's what happens when you completely misread the market​


https://www.theregister.com/2023/10/30/arm_intel_comment/
Click to expand...
Someone misread the market, but I doubt it was Intel. Intel's Meteor Lake is entirely meant to make it seem like Apple misread the market. If Meteor Lake can be as power efficient or better than Apple's Silicon, then there's no point sticking with ARM. I've said it before and I'll say it again, Apple going ARM was a mistake.
Screenshot-2023-10-10-at-9.05.04-AM.png
 
Gideon said:
No one will worry about it until a desktop chip is launched and sells well, server will likely become a custom thing if ARM is used. The death of X86 has been foretold far too many times.
Click to expand...
I agree that the death of x86 has been foretold too many times, and that x86 has plenty of life left in it.

That said I also believe it is a mistake by Gelsinger to not consider various ARM designs a real competitor.

These two things are not mutually exclusive.

While I tend to believe x86 will be around for quite some time, it also seems pretty evident that we are entering a period of time when the market will have a mix of different instruction sets competing with each other. Primarily it will be x86 and ARM, but there may be others as well over time.

See, the truth is that outside of software/binary compatibility, the instruction set you use doesn't really matter all that much. Every instruction set in use these days breaks down instructions to RISC-like micro-ops. This is true whether you are using x86 or ARM or just about anything else.

The popular belief is that x86 is outdated, and it is inherently less efficient than ARM, but that just isn't true. You can achieve similar perf/watt figures regardless of the instruction set. The instruction set really isn't THAT important in the grand scheme of things. The real magic happens on the back end where individual CPU architecture processes the micro-ops. The worst penalty you are going to get from x86 is probably slight, and it comes from the need of a bigger decode stage, which instruction sets that are natively closer to RISC don't need as much of. (Though, I think they still need them, I'm reaching the limits of my laymans understanding here)

While we all know and love x86 for its raw power over the years, x86 can compete (and has competed) with ARM in the low power perf/watt space as well. I had an Asus Zenfone 2 with Intel's Moorefield Atom Z3580 SoC in it back in 2016, and it was a great phone. Performance wise - on average - it wasn't that far behind the contemporaneous LG G4 with its Qualcomm MSM8992 Snapdragon 808 Hexa-core (4x1.4 GHz Cortex-A53 & 2x1.8 GHz Cortex-A57) meeting it in some benchmarks and falling behind it in a few others. In some areas the Intel smartphone did exceptionally well. In Browser performance it blew everyone else out of the water. In phone gaming performance it landed well above the iPhone 6, somewhere between the LG G4 and the Samsung Galaxy S6.

Battery life was also on par with the LG G4 and the Galaxy S6 as well, both of which were pretty similar.

But keep in mind, this was Intel's first foray into phone class SoC's, whereas Qualcomm had many generations of iterative refinement behind theirs. Also, they were running in an adapted Android ecosystem mostly optimized around ARM, which put them at a disadvantage.

Had they kept it up, I'm sure they could be competing on a similar footing with modern ARM chips in the phone/tablet space.

I'm not an Intel fanboy or anyhting, but all I am trying to express is that Instruction set really isnt terribly important to either raw performance, power use, or low power performance. Most of that is up to the underlying architecture. You can make similarly performing chips with either x86 or ARM, and probably once it is ready, RISCV as well. And this applies to high end desktop applications, server applications, and mobile applications.

The biggest problem with x86 is the licensing. Arm might be slightly better there, since Arm Holdings will license the instruction set (and even their own architectures) to pretty much anyone who pays them, but long term, who knows if they will still be doing that. As the Nvidia acquisition attempt showed us, it is a real risk (no pun intended)

A license free instruction set that maintains binary compatibility, while anyone can develop their own proprietary micro-op architecture underneath it would be the most desirable outcome, but this is one the Nvidia's, Apples, Microsofts, and other device makers of the world hate. They want to lock in users, and force them to not have choice, and that is a real problem.

Cross-compatibility needs to be universal and based on license-free industry standards, or this little hobby of ours is a s good as dead.
 
Last edited:
Zarathustra[H] said:
I agree that the death of x86 has been foretold too many times, and that x86 has plenty of life left in it.

That said I also believe it is a mistake by Gelsinger to not consider various ARM designs a real competitor.

These two things are not mutually exclusive.
Click to expand...

I won't get into all of the things, but this is a post of yours that I for the most part completely agree with. Death of x86 definitely isn't happening. But ARM at this point can't be ignored either.
It's happening at scale in Amazon data centers (as one example) and there are ARM chips for consumers that are doing very well. Not to mention nVidia + ARM with their Grace chip.

Intel, if they choose to stay x86, will still be alive for a minimum of 10 years. Maybe even 20+ as legacy stuff will still be useful for a long long time. However the ARM pie is likely to grow at the cost of the x86 pie.
 
It is not because they say it to project confidence to the investors that they really think it.

They already lost a nice client (Apple PCs) to arm, if a company would be aware of the competition it would be them
DukenukemX said:
. Intel's Meteor Lake is entirely meant to make it seem like Apple misread the market.
Click to expand...
If meteor lake would be in response to Apple arm decision, that wouild mean

1) Intel read ARM as a valid threat
2) Read that the market loved ARM solution
3) Responded to that reading of the market
 
In all fairness to Pat here, he’s not wrong.
You see that for the most part ARM and x86 trade blows for single and multi core performance and a lot of ARM’s advantage goes away there when using non LPDDR modules that match the x86 variants, so memory speed mismatches there play a big part. Upcoming changes to memory formats will level that out.

Where ARM SOC’s are trouncing is when dealing with anything GPU accelerated or when running tasks that can take advantage of the fixed function AI/ML accelerators on the ARM chips but that primarily a software hurdle not a hardware one.

Intel and AMD need to step up their iGPU game and it shows.

ARM gets a platform advantage because of its SOC nature, AMD or Intel solutions running in similar configs could meet or beat the ARM solutions but those solutions would be everything we buy PC platforms to avoid so I don’t welcome it. PC needs some new socket and memory designs to get prove speeds and latency while still giving us upgrade flexibility.
 
Lakados said:
In all fairness to Pat here, he’s not wrong.
You see that for the most part ARM and x86 trade blows for single and multi core performance and a lot of ARM’s advantage goes away there when using non LPDDR modules that match the x86 variants, so memory speed mismatches there play a big part. Upcoming changes to memory formats will level that out.

Where ARM SOC’s are trouncing is when dealing with anything GPU accelerated or when running tasks that can take advantage of the fixed function AI/ML accelerators on the ARM chips but that primarily a software hurdle not a hardware one.

Intel and AMD need to step up their iGPU game and it shows.

ARM gets a platform advantage because of its SOC nature, AMD or Intel solutions running in similar configs could meet or beat the ARM solutions but those solutions would be everything we buy PC platforms to avoid so I don’t welcome it. PC needs some new socket and memory designs to get prove speeds and latency while still giving us upgrade flexibility.
Click to expand...
Yes.

The other place ARM can compete in that x86 cannot is simply in the size of the silicon necessary to do what it does. Not because ARM is magically more efficient, but because ARM easily and neatly can kill the 40 years of x86 baggage. Apple has even gone far enough to remove 32-bit from their ARM processors.
This nets not only less expensive equivalents, but also more efficient in terms of power usage, and therefore easier to cool.

If x86 could be as ruthless (kill off all the legacy stuff even and up-to 32-bit) then it would be very similar to ARM. The problem is, the politics won't let that happen, and part of the reason why people are using x86 in the first place (maybe even a big part) is in fact all of the legacy code that big business wants to support. Despite the fact that it could just as easily be emulated on more modern hardware. "Emulation" isn't enough, they want native.
 
Lakados said:
ARM gets a platform advantage because of its SOC nature, AMD or Intel solutions running in similar configs could meet or beat the ARM solutions but those solutions would be everything we buy PC platforms to avoid so I don’t welcome it. PC needs some new socket and memory designs to get prove speeds and latency while still giving us upgrade flexibility.
Click to expand...

Aaaand this is why I don't want Qualcomm at all in the PC space lol as I said before.

But AMD and Intel probably are sleeping on their progress a bit too much regardless. Mobile gaming has taken off. As much as I hate the title, this might be driven by Gentrash Shitpact mostly, but the demand for that sort of GPU performance in your hands has really made Qualcomm get their shit together in the IGPU game...
 
StoleMyOwnCar said:
Aaaand this is why I don't want Qualcomm at all in the PC space lol as I said before.

But AMD and Intel probably are sleeping on their progress a bit too much regardless. Mobile gaming has taken off. As much as I hate the title, this might be driven by Gentrash Shitpact mostly, but the demand for that sort of GPU performance in your hands has really made Qualcomm get their shit together in the IGPU game...
Click to expand...
Lots of PC tablets that are already non upgradable, Chromebooks, Surface, lots of ultra portable flip-book blah blah blah, if Qualcomm can offer a superior package in that space then power to them.

If they can smack around an N4500 platform or Omnissiah willing the 1260P then Intel is not going to be happy loosing that volume segment.

AMD is the Ferrari of the CPU space, small batch high performance.

Intel is our Ford, nothing too fancy but it’s easy to mass produce and it gets the job done.

Intel makes its money with quantity and overall platform quality, so loosing big contracts for small things has big ripples for them.
 
Tangential to this is the Intel Foundry Services, Intel does a lot of low margin bulk sales, if Intel can get foundry contracts with similar margins for other companies products then their overall income doesn’t change. Make $3 from the sale of a Celeron N4500 ok, make 2.99 from building Qualcomm their chip but not have to deal with any of the packaging, product placement, support, etc, then technically they come out ahead on that deal.
 
Lakados said:
Tangential to this is the Intel Foundry Services, Intel does a lot of low margin bulk sales, if Intel can get foundry contracts with similar margins for other companies products then their overall income doesn’t change. Make $3 from the sale of a Celeron N4500 ok, make 2.99 from building Qualcomm their chip but not have to deal with any of the packaging, product placement, support, etc, then technically they come out ahead on that deal.
Click to expand...

Yep. Now that they have fixed their fabs, they can definitely make some money in this space. There is still a shortage of current gen fab capacity. Can't let TSMC and Samsung have all the fun.

I just wonder if they are too proud for that, even if it makes financial sense.
 
StoleMyOwnCar said:
what I'm worried about is that the world of PC building is going to move towards integrated motherboard/CPU/RAM combos where everything is soldered on.
Click to expand...
Kind of true in the laptop world. It wouldn't surprise me if the Dells and Lenovos take that approach for their pre-built desktops. Servers? I dunno.

For lots of other one-time add-ins, there are now soldered on chips on motherboards. BT, Ethernet, WiFi, sound, USB. But for self-builders, I doubt that would work for CPU and RAM. Too many options for CPU (speeds), and RAM (speeds and size). Of course, ask me again in 20 years. Maybe by then no one will be building their own.

StoleMyOwnCar said:
The GPU slot will still be separate because they can't compete with Nvidia or AMD in that space (yet).
Click to expand...
 
For me, an ARM PC would be of no interest. I'm not a laptop guy so I'm not power sensitive. Since retiring, my biggest concern is backwards compatibility (games, games and more games!). The ARM PC would need to run all my old games and applications at basically the same speed of my AMD PC.
 
erek said:
Seems like a mistake in leadership, that Apple M chip was pretty impressive, right?

“Additionally, the CEO noted: "When thinking about other alternative architectures like Arm, we also say, wow, what a great opportunity for our foundry business." If the adoption of Arm-based CPUs for Windows PCs becomes more present, Intel plans to compete with its next-generation x86 offerings like Meteor Lake, Arrow Lake, Lunar Lake, and even Panther Lake in the future. As stated, the CEO expects the competition to manufacture its chips at Intel's foundries so that Intel can provide a platform for these companies to serve the PC ecosystem.”

https://www.techpowerup.com/315228/...m-based-chips-as-competition-in-the-pc-sector
Click to expand...
According to that asshat AMD is in the rear view mirror…
 
UnknownSouljer said:
Intel, if they choose to stay x86, will still be alive for a minimum of 10 years. Maybe even 20+ as legacy stuff will still be useful for a long long time. However the ARM pie is likely to grow at the cost of the x86 pie.
Click to expand...
The thing about ARM is that it displaced MIPS because again those are cheap power efficient designs. PowerPC was also ahead of x86, until it wasn't. The only reason ARM would ever take the place of x86 is because the difference of cost would have to be huge. Last I checked, an Apple Macbook isn't exactly cheap to buy. You still have Apple M1/M2 owners with problems running their favorite software on their hardware because there just isn't a lot of interest in porting to ARM. This is the same problem with Windows on ARM, which is why Microsoft is trying to encourage developers to make ports for ARM.

LukeTbk said:
It is not because they say it to project confidence to the investors that they really think it.

They already lost a nice client (Apple PCs) to arm, if a company would be aware of the competition it would be them

If meteor lake would be in response to Apple arm decision, that wouild mean

1) Intel read ARM as a valid threat
2) Read that the market loved ARM solution
3) Responded to that reading of the market
Click to expand...
Intel today is not the same Intel from 2019. Hopefully not the same Intel from 2019. Intel did no innovation for years, mainly because AMD their competitor did not have the resources to make a decent CPU until Ryzen. Even after Ryzen was created, it took a while before AMD was able to create mobile chips based on Ryzen. Intel obviously didn't like losing Apple as a customer, so their response has been to mimic them. Not because they fear losing market share to ARM or Apple, but because they want Apple back.

It's more like this. Lets be honest here, in that Apple's market share didn't increase going ARM but has decreased. It's such a problem that Apple has for the first time gotten involved in pushing developers to port games to MacOS. This is probably due to feedback Apple has gotten from costumer complaints.
  1. Intel sees they lost Apple as a customer and are angry.
  2. Sees that Apple customers have problems losing x86 compatibility.
  3. Puts tons of money into ARC graphics.
  4. Will soon release Meteor Lake that's built specifically to discredit Apple's M-Series chips.

Red Falcon said:
Intel better get its act together, things are quickly changing with ARM, and x86-64 more than has its work cut out for it.
Click to expand...
Nothing good has changed for ARM. Remember that ARM is still trying to fight bankruptcy by increasing their license fees to everyone but Apple and Samsung. ARM is where it is because of Apple and Android. You aren't seeing an ARM revolution anywhere else, beyond these type of devices. It's not like Microsoft hasn't tried to bring ARM to their Surface devices, and failed. Apple is certainly pushing for ARM, but no amount of Apple zealots are enough to prevent the recent market share decline. As a reminder, Apple lost 40% earlier this year. This quarter, Apple's Mac sales dropped 23%. HP who sold twice as many computers, actually saw at least 6% growth. I'm not seeing this ARM revolution like these Apple zealots are seeing.
AMD is the only corporation pushing the x86-64 ISA forward, they just aren't big enough to take it over the market entirely.
Click to expand...
Intel's biggest recent innovation was AVX-512. I was really excited for this, until Intel started to remove this instruction set in newer CPU's. Then AMD brought it back with their Ryzen 7000 series. I'm still willing to bet that Apple's M3 will not only be ARMv9, but will support SVE2.
 
DukenukemX said:
ARM is where it is because of Apple and Android. You aren't seeing an ARM revolution anywhere else, beyond these type of devices.
Click to expand...
Apple is the megacorp that drove Apple in the mainstream market, which in turn fast-tracked the development cycle and adoption rate beyond simply mobile devices and onto desktop.
However, ARM in the server space is a completely different beast, and the megacorp that fast-tracked this was Intel itself simply by hubris alone.

DukenukemX said:
Intel's biggest recent innovation was AVX-512. I was really excited for this, until Intel started to remove this instruction set in newer CPU's.
Click to expand...
Agreed, and once again Intel shoots itself in its proverbial foot.
As I have stated before, this isn't a design or engineering issue, it is purely an upper-management issue.
 
philb2 said:
Kind of true in the laptop world. It wouldn't surprise me if the Dells and Lenovos take that approach for their pre-built desktops. Servers? I dunno
Click to expand...
Upcoming AMD and Intel server offerings have HBM3 on the package and it can function entirely within that if desired, no need to expand out into system ram from there so it’s happening.

I expect socketed or slotted consumer SOC packages in the not to distant future.
 
Red Falcon said:
As I have stated before, this isn't a design or engineering issue, it is purely an upper-management issue.
Click to expand...
Considering how much of their problem is product not on time, node reduction lost to TSMC, I am not so sure.
 
LukeTbk said:
Considering how much of their problem is product not on time, node reduction lost to TSMC, I am not so sure.
Click to expand...
Upper-management can always cause issues at the mid to low tiers as well.


I would timestamp the video, but the whole video is relevant as to Intel's current and near-future competition.


View: https://youtu.be/dA0xT5Bf_cc
 
Last edited:
Red Falcon said:
Agreed, and once again Intel shoots itself in its proverbial foot.
As I have stated before, this isn't a design or engineering issue, it is purely an upper-management issue.
Click to expand...
That's Intel for the past decade, a giant upper management problem. Them losing Apple was probably the best jump start they needed.

LukeTbk said:
M released in 2020

Apple Q3 2019 share of the market was significantly lower than now no ?
https://photos5.appleinsider.com/gallery/33147-57803-pc-pr-table-2019-q3-xl.jpg

7.6%, your table show a bit over 10% now, their place specially relative to Lenovo-HP-Dell-Acer seem much higher now
Click to expand...
Everyone loves to mention the past 3 years, but also forgets that the market went nuts for the past 3 years. Yes Apple got more sales, but not more relative to other manufacturers. This year especially, Apple is losing the most sales. Apple's ARM based laptops had their honeymoon period, and now it's over. The reality is setting in that Apple M-Series chips are less compatible with Windows x86 applications and are much slower. There are now AMD and Intel x86 based laptops that are just as power efficient and faster without any sacrifice with Windows x86 compatibility. Why would Intel be worried about ARM? Especially since they're probably a few months away from releasing Meteor Lake? No doubt that Apple's M3 will be a much bigger improvement over the M2's, but if Apple's venture into 3nm has shown us with the A17 Pro, it'll probably run a lot hotter and consume more power. I would expect active cooling being required with Apple's new M3's.

https://www.statista.com/statistics/263393/global-pc-shipments-since-1st-quarter-2009-by-vendor/
 
When you own one thing, why would you ever support the "other" thing?
Biz is biz. Don't get the "silly"!
 
"Neither RedBox nor Netflix are even on the radar screen in terms of competition," Blockbuster CEO Jim Keyes 2008

"Five hundred dollars? Fully subsidized? With a plan? I said that is the most expensive phone in the world.... And it doesn't appeal to business customers because it doesn't have a keyboard. Which makes it not a very good email machine."
“There’s no chance that the iPhone is going to get any significant market share.” Steve Ballmer 2007

"Screw the Nano. What the hell does the Nano do? Who listens to 1,000 songs?" Ed Zander Motorola 2006

Will be able to add Patrick P. Gelsinger to this list sooner than I expected.
 
ChadD said:
"Neither RedBox nor Netflix are even on the radar screen in terms of competition," Blockbuster CEO Jim Keyes 2008

"Five hundred dollars? Fully subsidized? With a plan? I said that is the most expensive phone in the world.... And it doesn't appeal to business customers because it doesn't have a keyboard. Which makes it not a very good email machine."
“There’s no chance that the iPhone is going to get any significant market share.” Steve Ballmer 2007

"Screw the Nano. What the hell does the Nano do? Who listens to 1,000 songs?" Ed Zander Motorola 2006

Will be able to add Patrick P. Gelsinger to this list sooner than I expected.
Click to expand...

Brings back some memories. Only 15-17 years ago now, but still memories. I had one of the old iPod Nanos. Not enough capacity, but was a pretty good music player. Also one of my first smartphones as a Windows one iirc. It sucked... barely operable lol. Didn't have a keyboard either way, not that I know who can actually use one of those tiny keyboards on a Blackberry.
 
  • Like
Reactions: ChadD
like this
StoleMyOwnCar said:
Brings back some memories. Only 15-17 years ago now, but still memories. I had one of the old iPod Nanos. Not enough capacity, but was a pretty good music player. Also one of my first smartphones as a Windows one iirc. It sucked... barely operable lol. Didn't have a keyboard either way, not that I know who can actually use one of those tiny keyboards on a Blackberry.
Click to expand...
I had a nano I loved as well. It had just enough space, small fast nice usable screen and had great audio quality. It was the crappy MP3 player perfected.
I also had a stupid windows phone. Keyboard that popped out and ya it was terrible... and creaky.
I actually loved my blackberries. Before Wifi everywhere and actually decent cell data they were just so ahead of every other option. My blackberries got email instantly everywhere... my windows phone would push me 2 hour old emails. MS couldn't even get the most basic thing Balmer was saying was all that mattered right. Makes the iphone is a sub par email machine quote even funnier.
 
sadsteve said:
For me, an ARM PC would be of no interest. I'm not a laptop guy so I'm not power sensitive. Since retiring, my biggest concern is backwards compatibility (games, games and more games!). The ARM PC would need to run all my old games and applications at basically the same speed of my AMD PC.
Click to expand...

I think it is mistake to associate ARM with "low power / mobile". That is certainly where it got its start, but there is nothing about the instruction set that makes it less suited to performance applications or more suited to mobile applications. ARM is just another instruction set. It is the underlying architecture that makes the big difference. Provided with the right underlying architecture, x86 can pretty much to everything ARM does (we saw this with Intel's Moorefield SoC's), and ARM can pretty much do everything x86 can do (as Apple has demonstrated with their M series chips)

Any migration to a new instruction set will of course have its challenges, but the truth is that modern CPU's have provided to be surprisingly receptive to translation layers. I'd imagine that IF the PC market shifts to ARM (I'm not yet convinced it will, but it might) then popular more recent titles will see ports, and older titles will likely run adequately using some sort of x86 to ARM wrapper.

It would likely not run as fast as native code, but since it would mostly apply to older titles, you probably wouldn't need it to run as fast either. Its sort of like running old DOS titles in DOSBox. Sure it loses some efficiency not running natively, but even a CPU in the late 90's was fast enough that this didn't matter when running 80's and early 90's DOS games.

Heck, it would even make sense that at least initially ARM CPU's intended for PC use include some sort of hardware x86 decode stage allowing them to semi-natively handle that process when called upon to run x86 code.

It doesn't seem like this would be too far fetched because this is essentially what x86 chips already do, decoding x86 instructions to RISC-like micro-ops.

In fact, I vaguely remember this being discussed WAY back when ARM architectures were first hitting the market. There was talk of ARM CPU's serving as the back end, having any number of existing instruction sets decoded to run on them.

In the end, if a new instruction set comes, I just don't see it being a game changer either way. It will be more of an evolutionary transition that won't really stop you from doing anything you are currently doing.

And it could be a good thing. ARM could open up competition to more players resulting in more more options at higher performance. Or it could be a mishmash of incompatible boards and CPU's making things even more of a mess.

Something truly open, like RISC-V would be even better on the competition side, though some more controls on feature levels would be necessary to ensure binary compatibility.

The important part in this transition is to not allow vendors to go down the road of proprietizing everything resulting in lock-ins and lock-outs. It is incumbent upon us, the users to push for cross compatibility of things like RAM standards/ form factors, PCIe, USB, drive form factors, no locked boot loaders, etc. such that the vendors can't pull their bullshit.

Because they will with 100% certainty try. The PC market thrives because of cross-compatibility, but every single vendor out there, Intel, Nvidia, Microsoft, Ape and even AMD benefits from market manipulations that hurt the customer and force them into making purchasing choices based on unnecessary compatibility lock-ins (like how if you bought a G-Sync screen, you now pretty much have to buy a Nvidia GPU, etc.)

We have to fight the industry to prevent the PC market from sliding into a proprietized dystopia. Based on history, I don't have too much confidence consumers will actually succeed...
 
Last edited:
UnknownSouljer said:
Yes.

The other place ARM can compete in that x86 cannot is simply in the size of the silicon necessary to do what it does. Not because ARM is magically more efficient, but because ARM easily and neatly can kill the 40 years of x86 baggage. Apple has even gone far enough to remove 32-bit from their ARM processors.
This nets not only less expensive equivalents, but also more efficient in terms of power usage, and therefore easier to cool.

If x86 could be as ruthless (kill off all the legacy stuff even and up-to 32-bit) then it would be very similar to ARM. The problem is, the politics won't let that happen, and part of the reason why people are using x86 in the first place (maybe even a big part) is in fact all of the legacy code that big business wants to support. Despite the fact that it could just as easily be emulated on more modern hardware. "Emulation" isn't enough, they want native.
Click to expand...

That's the biggest factor at play going forward here, IMO. x86 is still supporting instruction sets from way back in the day. Ultimately the chips will get larger in a world where everything else is getting smaller. ARM is a remedy to that situation.

I'm not a computer engineer, but it's something I had been thinking about when comparing the technologies.
 
Lakados said:
Lots of PC tablets that are already non upgradable, Chromebooks, Surface, lots of ultra portable flip-book blah blah blah, if Qualcomm can offer a superior package in that space then power to them.

If they can smack around an N4500 platform or Omnissiah willing the 1260P then Intel is not going to be happy loosing that volume segment.

AMD is the Ferrari of the CPU space, small batch high performance.

Intel is our Ford, nothing too fancy but it’s easy to mass produce and it gets the job done.

Intel makes its money with quantity and overall platform quality, so loosing big contracts for small things has big ripples for them.
Click to expand...

Intel's biggest advantage remains legacy contracts with large PC OEMs like Dell and a perception among less technically inclined PC users who think of Intel being the standard and AMD being the equivalent of a "cheap IBM clone" if you will back in the day. That can only take you so far though. People are increasingly discovering Ryzen while Intel rested on that and now Intel is flat-footed and trying to catch up. They tried to accelerate that by hiring Jim Keller, but he left fairly quickly. Why? Not sure, but not a good sign.
 
ChadD said:
"Neither RedBox nor Netflix are even on the radar screen in terms of competition," Blockbuster CEO Jim Keyes 2008

"Five hundred dollars? Fully subsidized? With a plan? I said that is the most expensive phone in the world.... And it doesn't appeal to business customers because it doesn't have a keyboard. Which makes it not a very good email machine."
“There’s no chance that the iPhone is going to get any significant market share.” Steve Ballmer 2007

"Screw the Nano. What the hell does the Nano do? Who listens to 1,000 songs?" Ed Zander Motorola 2006

Will be able to add Patrick P. Gelsinger to this list sooner than I expected.
Click to expand...
The difference here is that x86's death has been predicted many times before. Even Intel tried to replace x86 with Titanium, only to find that AMD created 64-bit instructions for x86 and the performance difference was huge.

“When we look at the future roadmaps projected out to mid-2006 and beyond,” Jobs said, “what we see is the PowerPC gives us sort of 15 units of performance per Watt, but the Intel roadmap in the future gives us 70. So this tells us what we have to do.” -- Steve Jobs

"I'm going to destroy Android, because it's a stolen product. I'm willing to go thermonuclear war on this," --Steve Jobs
 
You must log in or register to reply here.
Back
Top